6 SQL Statements

This chapter provides information about the SQL statements available in TimesTen.

SQL statements are generally considered to be either data manipulation language (DML) statements or data definition language (DDL) statements.

DML statements modify database objects. INSERT, UPDATE and DELETE are examples of DML statements. The SELECT statement retrieves data from one or more tables or views.

DDL statements modify the database schema. CREATE TABLE and DROP TABLE are examples of DDL statements.

In addition to an alphabetical listing of all statements, this chapter also contains:

Summary of SQL statements supported in TimesTen

Table 6-1 shows a summary of the SQL statements in TimesTen. The second column indicates if the statement is supported in TimesTen Scaleout. Every statement except ALTER SEQUENCE is supported in TimesTen Classic.

Table 6-1 SQL statements supported in TimesTen

SQL statement Supported in TimesTen Scaleout?

ALTER ACTIVE STANDBY PAIR

N

ALTER CACHE GROUP

Y

ALTER FUNCTION

N

ALTER PACKAGE

N

ALTER PROFILE

Y

ALTER PROCEDURE

N

ALTER REPLICATION

N

ALTER SEQUENCE

Y

Not supported in TimesTen Classic.

ALTER SESSION

Y

ALTER TABLE

Y

Unsupported clauses:

Aging and column-based compression

Unsupported data types:

LOB columns are not supported in tables. LOB variables are supported in PL/SQL programs.

ALTER USER

Y

CALL

Y

COMMIT

Y

CREATE ACTIVE STANDBY PAIR

N

CREATE CACHE GROUP

Y static read-only with incremental autorefresh

CREATE FUNCTION

Y

CREATE INDEX

Y

CREATE MATERIALIZED VIEW

Y with restrictions

CREATE PACKAGE

Y

CREATE PACKAGE BODY

Y

CREATE PROCEDURE

Y

CREATE PROFILE

Y

CREATE REPLICATION

N

CREATE SEQUENCE

Y with TimesTen Scaleout specific BATCH clause.

CREATE SYNONYM

Y

CREATE TABLE

Y including CREATE TABLE... AS SELECT

Unsupported clauses:

Aging and column-based compression

Unsupported data types:

LOBs and ROWID

Distribution clause is not supported for global temporary tables.

CREATE USER

Y

CREATE VIEW

Y

DELETE

Y

DROP ACTIVE STANDBY PAIR

N

DROP CACHE GROUP

Y static read-only cache groups with incremental autorefresh

DROP FUNCTION

Y

DROP INDEX

Y

DROP MATERIALIZED VIEW

Y

DROP PACKAGE [BODY]

Y

DROP PROCEDURE

Y

DROP PROFILE

Y

DROP REPLICATION

N

DROP SEQUENCE

Y

DROP SYNONYM

Y

DROP TABLE

Y

DROP USER

Y

DROP VIEW

Y

FLUSH CACHE GROUP

N

GRANT

Y

INSERT

Y

INSERT...SELECT

Y

LOAD CACHE GROUP

Y static read-only cache groups with incremental autorefresh

MERGE

N

REFRESH CACHE GROUP

Y static read-only cache groups with incremental autorefresh

REVOKE

Y

ROLLBACK

Y

SELECT

Y

TRUNCATE TABLE

Y, but TRUNCATE TABLE is similar to a DDL statement that invalidates all commands that depend on the table being truncated. It is preferable to use the DELETE statement rather than the TRUNCATE statement to delete all rows in a table.

UNLOAD CACHE GROUP

Y static read-only cache groups with incremental autorefresh

UPDATE

Y

Comments within SQL statements

A comment can appear between keywords, parameters, or punctuation marks in a statement. You can include a comment in a statement in two ways:

  • Begin the comment with a slash and an asterisk (/*). Proceed with the text of the comment. The text can span multiple lines. End the comment with an asterisk and a slash. (*/). You do not need to separate the opening and terminating characters from the text by a space or line break.

  • Begin the comment with two hyphens (--). Proceed with the text of the comment. The text cannot extend to a new line. End the comment with a line break.

Optimizer hints

Optimizer hints are instructions that are passed to the TimesTen query optimizer. The optimizer considers these hints when choosing the best execution plan for your query. Most of the hints are supported both in TimesTen Scaleout and in TimesTen Classic. There are also hints that are supported only in TimesTen Scaleout. See "Optimizer hints supported in TimesTen Scaleout only" for information.

TimesTen supports three levels of optimizer hints:

  • Statement level optimizer hints: When specified, the optimizer considers the hint for the particular statement. See "Statement level optimizer hints" for details.

  • Transaction level optimizer hints: When specified (by calling the appropriate built-in procedure), the optimizer considers the hint for the entire transaction. See "Use optimizer hints to modify the execution plan" in the Oracle TimesTen In-Memory Database Operations Guide.

  • Connection level optimizer hints: When specified, the optimizer considers the hint for the entire connection. See "Use optimizer hints to modify the execution plan" in the Oracle TimesTen In-Memory Database Operations Guide and "OptimizerHint" in the Oracle TimesTen In-Memory Database Reference for details.

The order of precedence for optimizer hints is statement level hints, transaction level hints and then connection level hints. Table 6-2 provides a summary of the statement, transaction, and connection level optimizer hints.

Table 6-2 Summary of statement, transaction, and connection level optimizer hints

Statement level optimizer hint Transaction level optimizer hint Connection level optimizer hint

You specify the hint within the comment syntax and after a SQL VERB in a SQL statement.

You specify the hint by calling the ttOptSetFlag, or the ttOptSetOrder, or the ttOptUseIndex built-in procedure.

You specify the hint in the OptimizerHint general connection attribute.

The hint is scoped to the SQL statement.

The hint is scoped to the transaction.

The hint is scoped to the connection.

The autocommit setting has no effect. After the statement containing the hint is executed, the hint has no effect on future statements or queries.

The autocommit setting has an effect. You must set autocommit to off. Doing so ensures the hint is in effect for the duration of your transaction (until you issue a commit or rollback). If you do not set autocommit to off, the statement is executed in its own transaction and the hint only has an effect on the statement.

The autocommit setting has no effect. The hint is in effect for the duration of the connection.

The optimizer considers the hint for the statement only.

The optimizer considers the hint for all statements in the transaction.

The optimizer considers the hint for all statements in the connection.

The hint is supported in the CREATE TABLE...AS SELECT statement.

The hint is not supported in the CREATE TABLE...AS SELECT statement. This is a DDL statement and TimesTen implicitly commits a DDL statement.

The hint is not supported in the CREATE TABLE...AS SELECT statement. This is a DDL statement and TimesTen implicitly commits a DDL statement.

If you specify the hint in a transaction in which transaction level optimizer hints or connection level optimizer hints are specified, the statement level optimizer hint overrides the transaction level hint or the connection level hint for the SQL statement. After TimesTen executes the SQL statement:

  • The original transaction level optimizer hint remains in effect for the duration of the transaction or

  • The original connection level optimizer hint remains in effect for the duration of the connection.

The hint is in effect for the duration of the transaction. If you specify a statement level optimizer hint in a SQL statement, the statement level optimizer hint is in effect for the statement and the optimizer does not use the transaction level hint for the statement. After TimesTen executes the statement, the original transaction level optimizer hint remains in effect for the duration of the transaction.

A hint specified at this level overrides the same hint specified at the connection level.

The hints are in effect for the duration of the connection. The order of precedence is statement level, transaction level, and then connection level.

You use the statement level optimizer hints if you want to influence the optimizer for a specific statement. You must specify the hint for each statement in which you want to influence the optimizer. This could result in multiple alterations to your statements.

You use the transaction level optimizer hints to influence the optimizer for all statements in a transaction. You do not have to specify a hint for each statement. The hint applies to all statements in the transaction. The hint can be overridden by specifying the hint at the statement level.

You use the connection level optimizer hint to influence the optimizer for all statements in the connection. The hint can be overridden by specifying the hint at the transaction or at the statement level.

Statement level optimizer hints

Statement level optimizer hints are comments in a SQL statement that pass instructions to the TimesTen query optimizer. The optimizer considers these hints when choosing the best execution plan for your query. It analyzes the SQL statements and generates a query plan which is then used by the SQL execution engine to execute the query and return the data.

See "Use optimizer hints to modify the execution plan" in Oracle TimesTen In-Memory Database Operations Guide for information about statement level optimizer hints.

SQL Syntax

A SQL statement can have one comment that includes one or more statement level optimizer hints.
These hints are only supported in TimesTen Cache:
  • TT_DynamicLoadMultiplePKs
  • TT_DynamicLoadRootTbl
  • TT_DynamicPassThrough

Some hints are not supported in certain SQL statements:

  • TT_CommitDMLOnSuccess is supported in the DELETE, INSERT, and UPDATE statements. It is also valid in the INSERT...SELECT statement and must follow the SELECT keyword. This hint is supported in TimesTen Scaleout only.

  • The TT_GridQueryExec and TT_PartialResult hints are supported in the SELECT, INSERT...SELECT, and CREATE TABLE... AS SELECT SQL statements only and these hints must follow the SELECT keyword. These hints are supported in TimesTen Scaleout only.

  • The remaining hints are supported in the DELETE, INSERT, MERGE, SELECT, UPDATE, INSERT...SELECT, and CREATE TABLE...AS SELECT SQL statements and these hints must follow the DELETE, INSERT, MERGE, SELECT, or UPDATE keyword.

Table 6-3 shows the proper placement of hints in a SQL statement.

You embed statement level optimizer hints in comment syntax. TimesTen supports hints in comments that span one line and in comments that span more than one line. If your comment that contains the hint spans one or more lines, use the comment syntax, /*+...*/. If your comment that contains the hint spans one line, use the comment syntax, --+.

Syntax:

SQL VERB {/*+ [CommentText] hint [{hint|CommentText} [...]] */ |
          --+ [CommentText] hint [{hint|CommentText} [...]] }
hint::= ScaleoutHint | CacheHint | JoinOrderHint | IndexHint| FlagHint
ScaleoutHint::= TT_CommitDMLOnSuccess({0|1})|TT_GridQueryExec({LOCAL|GLOBAL})|
                TT_PartialResult(0|1)
CacheHint::= TT_DynamicLoadMultiplePKs ({0|1})|TT_DynamicLoadRootTbl ({0|1})|
             TT_DynamicPassthrough(N) 
JoinOrderHint::= TT_JoinOrder (CorrelationName CorrelationName [...])
IndexHint::= TT_Index (CorrelationName,IndexName,{0|1} [;...])
FlagHint::= FlagName (0|1)
FlagName::= TT_BranchAndBound|TT_CountAsInt|TT_DynamicLoadEnable|
            TT_DynamicLoadErrorMode| TT_FirstRow|TT_ForceCompile|
            TT_GenPlan|TT_HashGb|TT_HashScan|TT_IndexedOr|TT_MergeJoin|
            TT_NestedLoop|TT_NoRemRowIdOpt|TT_Range|TT_Rowid|TT_RowLock|
            TT_ShowJoinOrder|TT_TblLock|TT_TblScan|TT_TmpHash|TT_TmpRange|
            TT_TmpTable|TT_UseBoyerMooreStringSearch|

Parameters

Parameter Description

SQL VERB

SQL VERB refers to one of the keywords: DELETE, INSERT, MERGE, SELECT, or UPDATE. You embed a statement level optimizer hint in comment syntax and if the comment syntax contains a statement level optimizer hint, the comment syntax must follow the SQL VERB.

The TT_GridQueryExec and TT_PartialResult hints are valid for the SELECT keyword only.

/*+ hint */

One or more hints that are embedded in comment syntax. The comment syntax can span one or more lines. The plus sign (+) denotes the start of a statement level optimizer hint.

Make sure there is no space between the star (*) and the plus sign (+).

--+ hint

One or more hints that are embedded in comment syntax. The comment syntax can only span one line. The plus sign (+) denotes the start of a statement level optimizer hint.

Make sure there is no space between the dash (-) and the plus sign (+).

hint

A statement level optimizer hint. A SQL statement supports one or more statement level optimizer hints as one comment string. For one SQL statement, you can specify one comment that contains one or more hints and that comment must follow a DELETE, INSERT, MERGE, SELECT, or UPDATE keyword (or for TT_GridQueryExec and TT_PartialResult, the SELECT keyword). TT_CommitDMLOnSuccess must follow a DELETE, INSERT, or UPDATE keyword and in the INSERT...SELECT statement, it must follow the SELECT keyword.

If you specify more than one hint within the comment, make sure there is a space between the hints.

Statement level optimizer hints are scoped to a SQL statement and have per query semantics.

For hints other than TT_GridQueryExec, TT_PartialResult, or TT_CommitDMLOnSuccess:

  • The name and type of statement level optimizer hints map to the transaction level optimizer hints. Transaction level optimizer hints are invoked by calling the built-in procedures ttOptSetFlag, ttOptSetOrder, or ttOptUseIndex.

  • Transaction level hints are scoped to the transaction and have transaction semantics. You must set autocommit to 0 and the transaction level optimizer hint is in effect for the duration of your transaction.

  • For more information, see "ttOptSetFlag," "ttOptSetOrder" or "ttOptUseIndex" in the Oracle TimesTen In-Memory Database Reference.

CommentText

Text within a comment string. You can use both statement level optimizer hints and commenting text within one comment. Make sure to include a space between the hint and the commenting text.

ScaleoutHint

ScaleoutHint refers to the TT_CommitDMLOnSuccess statement level hint as well as the TT_GridQueryExec and the TT_PartialResult statement level optimizer hints. These hints are supported in TimesTen Scaleout only.

  • See "TT_CommitDMLOnSuccess optimizer hint" for detailed information.

  • TT_GridQueryExec (LOCAL|GLOBAL) returns query results for the local element (if LOCAL) or for all elements (if GLOBAL). If K-safety is set to 2 and GLOBAL is specified, the results include the data in all elements in a replica set.

    LOCAL: Queries are executed in the local element only. If the local element does not have a full copy of the data, TimesTen Scaleout returns partial results.

    GLOBAL: TimesTen Scaleout retrieves data from all elements, including copies of the rows from all tables from all replica sets to generate the results. This results in duplicate data returned if K-safety is set to 2 or if tables have a duplicate distribution scheme.

    The default is neither local nor global. If you do not specify this hint, the query is executed in one logical data space. Exactly one full copy of the data is used to compute the query.

  • TT_PartialResult (0|1) returns an error (if 0) or partial results (if 1) when data is not available.

    0: Returns error if the required data is not available in the case where all elements in a replica set are not available. If at least one element from each replica set is available or the data required by the query is available, the optimizer returns the query result correctly without error. This is the default.

    1: Returns partial results if all elements in a replica set are not available.

SELECT /*+TT_GridQueryExec(LOCAL)*/   
        COUNT(*), elementId# FROM t
        GROUP BY elementId#;
SELECT /*+TT_GridQueryExec(GLOBAL)*/   
        COUNT(*), elementId# FROM t 
        GROUP BY elementId#;
SELECT /*+TT_PartialResult(0)*/ COUNT  
       (*), elementId# FROM t GROUP BY
       elementId#;
SELECT /*+TT_PartialResult(1)*/ COUNT  
      (*), elementId# FROM t GROUP BY
      elementId#;
CacheHint CacheHint refers to the supported optimizer hints for TimesTen Cache. These hints are TT_DynamicLoadMultiplePKs, TT_DynamicLoadRootTbl, and TT_DynamicPassthrough. These hints are described later in this table (in alphabetical order).

JoinOrderHint::= TT_JoinOrder ( CorrelationName CorrelationName [...] )

JoinOrderHint refers to the syntax for the TT_JoinOrder statement level optimizer hint. The TT_JoinOrder hint instructs the optimizer to join your tables in a specified order. The join order is in effect for the statement only.

Specify TT_JoinOrder and two or more correlation names. Make sure to enclose the correlation names in parentheses.

CorrelationName refers to the alias of your table specified in the query or subquery of your SQL statement. Make sure to use the correlation name rather than the actual table name when using this hint.

For example, if you are joining the employees and departments tables and you specify the emps correlation name for the employees table and the depts correlation name for the departments table, then to specify the TT_JoinOrder hint in ttIsql:

Command> SELECT /*+ TT_JoinOrder
           (EMPS DEPTS)*/...

If your TT_JoinOrder hint contains syntax errors, TimesTen ignores the hint. If your TT_JoinOrder hint does not contain a closing parenthesis, then the remainder of the comment string is ignored. So if you specify additional statement level optimizer hints after the TT_JoinOrder hint, and the TT_JoinOrder hint is missing the closing parenthesis, these additional statement level optimizer hints are ignored.

You can execute the built-in procedure, ttOptSetOrder, to specify a join order for the duration of your transaction. The TT_JoinOrder statement level optimizer hint is in effect for the statement only.

For more information on ttOptSetOrder, see "ttOptSetOrder" in the Oracle TimesTen In-Memory Database Reference.

IndexHint::= TT_INDEX (CorrelationName IndexName {0|1} [;...] )

IndexHint refers to the syntax for the TT_INDEX statement level optimizer hint. Use the TT_INDEX hint to direct the optimizer to use or not use an index for your table. The index hint is in effect for the statement only.

CorrelationName refers to the alias of your table specified in the query or subquery of your SQL statement. Make sure to use the correlation name rather than the actual table name. If you omit the correlation name, the setting affects all tables with the specified index name.

IndexName refers to the index you want the optimizer to consider or not consider for the table If you omit the index name, the setting applies to all indexes of the specified correlation.

Specify a value of 0 to ask the optimizer not to consider the index. Specify a value of 1 to ask the optimizer to consider the index.

For example, To direct the optimizer to use the index emp_name_ix for a query on the employees table and you specify the correlation name, e, for the employees table, then to specify the TT_INDEX hint in ttIsql:

Command> SELECT /*+ TT_INDEX
          (E,EMP_NAME_IX,1) */ ...

Use a semicolon (;) to include more than one CorrelationName, IndexName, {0|1} for the TT_INDEX hint. You must specify each of the parameters: the CorrelationName, the IndexName, and either 0 or 1.

If your TT_Index hint contains syntax errors, TimesTen ignores the hint. If your TT_Index hint does not contain a closing parenthesis, then the remainder of the comment string is ignored. So if you specify additional statement level optimizer hints after the TT_Index hint and the TT_Index hint is missing the closing parenthesis, these additional statement level optimizer hints are ignored.

You can execute the built-in procedure, ttOptUseIndex, to specify an index hint for the duration of your transaction. The TT_INDEX statement level optimizer hint is in effect for the statement only.

For more information on ttOptUseIndex, see "ttOptUseIndex" in the Oracle TimesTen In-Memory Database Reference.

FlagHint

FlagHint refers to statement level optimizer flags that you enable or disable to influence the execution plan of the TimesTen query optimizer. These flags map to the flags used in the ttOptSetFlag built-in procedure.

Statement level optimizer hint flags are in effect for the statement only whereas transaction level optimizer hint flags are in effect for the duration of your transaction.

TT_BranchAndBound

Flag that maps to the flag BranchAndBound in the ttOptSetFlag built-in procedure.

TT_CountAsInt

This hint controls the return data type for the COUNT function (used in a query). Specify 1 to have the return data type be TT_INTEGER. Specify 0 to have the return data type be TT_BIGINT. If you do not specify this hint, the default return data type is TT_BIGINT. This hint is supported at the statement and at the connection levels. See "COUNT" for information on the COUNT function.

This hint is provided for backward compatibility. If you specify the hint with a value of 1, it may result in an unexpected integer overflow. New applications should not specify this hint. This ensures TimesTen uses the default return data type of TT_BIGINT for the COUNT function.

This example specifies a value of 1 for the hint. The return data type is TT_INTEGER:

Command> describe SELECT /*+TT_CountAsInt(1)*/ COUNT (*) FROM dual;

Prepared Statement:
  Columns:
    EXP                             TT_INTEGER NOT NULL

This example specifies a value of 0 for the hint. The return data type is TT_BIGINT.

Command> describe SELECT /*+TT_CountAsInt(0)*/ COUNT (*) FROM dual;

Prepared Statement:
  Columns:
    EXP                             TT_BIGINT NOT NULL

This example does not set the optimizer hint. The default return data type is TT_BIGINT.

describe SELECT COUNT (*) FROM dual;

Prepared Statement:
  Columns:
    EXP                             TT_BIGINT NOT NULL

TT_DynamicLoadEnable

Flag that maps to the flag DynamicLoadEnable in the ttOptSetFlag built-in procedure.

TT_DynamicLoadErrorMode

Flag that maps to the flag DynamicLoadErrorMode in the ttOptSetFlag built-in procedure.

TT_DynamicLoadMultiplePKs{ 0|1} TimesTen Cache optimizer hint, supported in TimesTen Classic. This hint enables (if set to 1) or disables (if set to 0), the ability to dynamically load multiple cache instances on a single table cache group. The dynamic load operation must be triggered by a qualified SELECT statement that contains a WHERE clause, in which the WHERE clause references multiple primary key values of the root table of the cache group. The default is 1. When both the TT_DynamicLoadMultiplePKs and the TT_DynamicLoadRootTbl hints are specified, the TT_DynamicLoadMultiplePKs take precedence. See "Dynamically loading multiple cache instances with multiple primary keys" in the Oracle TimesTen In-Memory Database Cache Guide for details.
TT_DynamicLoadRootTbl TimesTen Cache optimizer hint, supported in TimesTen Classic. This hint enables (if set to 1) or disables (if set to 0), the ability to dynamically load multiple cache instances on a single table cache group. The dynamic load operation must be triggered by a qualified SELECT statement that contains a WHERE clause, in which the WHERE clause does not reference multiple primary key values of the root table of the cache group. The default is 0. When both the TT_DynamicLoadMultiplePKs and the TT_DynamicLoadRootTbl hints are specified, the TT_DynamicLoadMultiplePKs take precedence. See "Dynamically loading multiple cache instances without multiple primary keys" in the Oracle TimesTen In-Memory Database Cache Guide for details.
TT_DynamicPassThrough(N) TimesTen Cache optimizer hint, supported in TimesTen Classic. If specified, this hint limits the number of rows that can be dynamically loaded into a TimesTen cache instance. Specifically, if a dynamic load operation triggered by a qualified SELECT statement results in a number of rows that is greater than the specified N row limit, the cache instance is not loaded and instead the query is passed to the Oracle database. The dynamic load must be triggered by a qualified SELECT statement and the cache group must not have a WHERE clause. The hint is ignored for non-SELECT statements. Set this hint to the maximum number of rows you want dynamically loaded. If you set the hint to a value less than or equal to 0 or if you do not specify the hint, the dynamic load has no row limit. In this case, there is not a limit in the number of rows can be loaded into the cache instance. See "Automatic passthrough of dynamic load to the Oracle database" in the Oracle TimesTen In-Memory Database Cache Guide for details.

TT_FirstRow

Flag that maps to the flag FirstRow in the ttOptSetFlag built-in procedure.

TT_ForceCompile

Flag that maps to the flag ForceCompile in the ttOptSetFlag built-in procedure.

TT_GenPlan

Flag that maps to the flag GenPlan in the ttOptSetFlag built-in procedure.

TT_HashGb

Flag that maps to the flag HashGb in the ttOptSetFlag built-in procedure.

TT_HashScan

Flag that maps to the flag Hash in the ttOptSetFlag built-in procedure.

TT_IndexedOr

Flag that maps to the flag IndexedOr in the ttOptSetFlag built-in procedure.

TT_MergeJoin

Flag that maps to the flag MergeJoin in the ttOptSetFlag built-in procedure.

TT_NestedLoop

Flag that maps to the flag NestedLoop in the ttOptSetFlag built-in procedure.

TT_NoRemRowIdOpt

Flag that maps to the flag NoRemRowIdOpt in the ttOptSetFlag built-in procedure.

TT_Range

Flag that maps to the flag Range in the ttOptSetFlag built-in procedure.

TT_Rowid

Flag that maps to the flag Rowid in the ttOptSetFlag built-in procedure.

TT_RowLock

Flag that maps to the flag Rowlock in the ttOptSetFlag built-in procedure.

TT_ShowJoinOrder

Flag that maps to the flag ShowJoinOrder in the ttOptSetFlag built-in procedure.

TT_TblLock

Flag that maps to the flag TblLock in the ttOptSetFlag built-in procedure.

TT_TblScan

Flag that maps to the flag Scan in the ttOptSetFlag built-in procedure.

TT_TmpHash

Flag that maps to the flag TmpHash in the ttOptSetFlag built-in procedure.

TT_TmpRange

Flag that maps to the flag TmpRange in the ttOptSetFlag built-in procedure.

TT_TmpTable

Flag that maps to the flag TmpTable in the ttOptSetFlag built-in procedure.

TT_UseBoyerMooreStringSearch

Flag that maps to the flag UseBoyerMooreStringSearch in the ttOptSetFlag built-in procedure.

Note:

For descriptions of flags discussed in the preceding table, see "ttOptSetFlag" in the Oracle TimesTen In-Memory Database Reference

Description

  • Embed statement level optimizer hints in comment syntax. Begin the comment with either /* or --. Follow the beginning comment syntax with a plus sign (+). The plus sign (+) signals TimesTen to interpret the comment as a list of hints. The plus sign (+) must follow immediately after the comment delimiter. (For example, after /* or after --). No space is permitted between the comment delimiter and the plus sign (+).

    In the following example, there is a space between the star (*) and the plus sign (+), so the hint is ignored:

    Command> SELECT /* + TT_TblScan (1) This hint is ignored because there is a
              space between the star (*) and the plus (+) sign. */ ...
    
  • A hint is one of the statement level optimizer hints supported by TimesTen. There can be a space between the plus sign (+) and the hint. If the comment contains multiple hints, separate the hints by at least one space. For example, to specify two hints on one line:

    Command> SELECT --+ TT_MergeJoin (0) TT_NestedLoop (1) 
              ...
    
  • You can intersperse commenting text with hints in a comment. For example,

    Command> SELECT /*+ TT_HashScan (1) This demonstrates a hint followed by a
               comment string. */ ...
    
  • TimesTen ignores hints and does not return an error if:

    • Your hint does not follow the DELETE, INSERT, MERGE, SELECT or UPDATE keyword (or for TT_GridQueryExec or TT_PartialResult, the SELECT keyword). TT_CommitDMLOnSuccess must follow the DELETE, INSERT, UPDATE keyword and for INSERT...SELECT, it must follow the SELECT keyword.

    • Your hint contains misspellings or syntax errors. If you have hints that are within the same comment and some hints are correct syntactically and some hints are incorrect syntactically, TimesTen ignores the incorrect hints and accepts the correct hints.

    • You use either the TT_JoinOrder or TT_Index hint and you do not supply a closing parenthesis, the remainder of the hint string is ignored.

  • For hints that conflict with each other, TimesTen uses the rightmost hint in the comment. For example, if the comment string is /*+TT_TblScan (0)...TT_TblScan (1) */, the rightmost hint, TT_TblScan(1), is used.

  • Statement level optimizer hints override conflicting transaction level optimizer hints. If you specify a transaction level optimizer hint that conflicts with a statement level optimizer hint, the statement level optimizer hint overrides the conflicting transaction level optimizer hint. For example, if you call ttOptSetFlag, and enable the Range flag and then you issue a SQL query and disable the statement level optimizer flag, TT_Range, TimesTen disables the range flag for the query. After the query is executed, the original range flag setting that was in place in the transaction before the query was executed remains in effect for the duration of the transaction. For more information, see "Using statement level optimizer hints for a SELECT query". The TT_GridQueryExec, TT_PartialResult, TT_CommitDMLOnSuccess, and TT_CountAsInt hints are not supported at the transaction level.

  • Do not use statement level optimizer hints in a subquery.

  • The TimesTen query optimizer does not recognize statement level optimizer hints for passthrough statements. TimesTen passes the SQL text for passthrough statements to the Oracle database and the SQL text is processed according to the SQL rules of the Oracle database. Passthrough statements are not supported in TimesTen Scaleout.

SQL statements that support statement level optimizer hints

You can specify statement level optimizer hints in SQL statements. Not all hints are supported in all statements. You must specify the hint within comment syntax and the comment syntax must immediately follow the SQL VERB. (For example, SELECT /*+ hint */...) Table 6-3 shows the correct placement of the statement level hint. It also indicates if a hint is not supported in the statement.

Table 6-3 Placement of statement level hints in SQL statements

SQL statement Placement of hint

CREATE TABLE... AS SELECT

CREATE TABLE...AS SELECT [hint]...

Do not use transaction level hints with the CREATE TABLE...AS SELECT statement.

TT_CommitDMLOnSuccess is not supported in this statement.

DELETE

DELETE [hint]...

The TT_GridQueryExec and TT_PartialResult hints are not supported in this statement.

INSERT

INSERT [hint]...

The TT_GridQueryExec and TT_PartialResult hints are not supported in this statement.

INSERT...SELECT

INSERT...SELECT [hint]...

MERGE

MERGE [hint]...

The TT_GridQueryExec and TT_PartialResult hints are not supported.

SELECT

SELECT [hint]...

Do not specify a hint in a subquery.

The TT_CommitDMLOnSuccess hint is not supported in this statement.

SELECT...{UNION|MINUS| INTERSECT} SELECT...

SELECT [hint] {UNION|MINUS|INTERSECT} SELECT...

The TT_CommitDMLOnSuccess hint is not supported in this statement.

UPDATE

UPDATE [hint]...

The TT_GridQueryExec and TT_PartialResult hints are not supported in this statement.

Understanding hints

Use optimizer hints to influence the TimesTen query optimizer in determining the choice of the execution plan for your query.

TT_GridQueryExec, TT_PartialResult and TT_CommitDMLOnSuccess are supported at the connection and statement levels only. This section is not valid for these hints.

To view transaction level optimizer hints, execute the built-in procedure, ttOptSetFlag. For more information on the built-in procedure, ttOptGetFlag, see "ttOptGetFlag" in Oracle TimesTen In-Memory Database Reference.

Examples

For TT_CommitDMLOnSuccess examples, see "TT_CommitDMLOnSuccess optimizer hint" for information.

For TT_GridQueryExec and TT_PartialResult examples:

The following examples illustrate usages of statement level and transaction level optimizer hints. The TimesTen optimizer is a cost based query optimizer and generates what it thinks is the most optimal execution plan for your statement. This plan differs from release to release. The plan is based on the indexes that exist on the referenced tables as well as the column and table statistics that are available. When you recompute statistics or change indexes, the TimesTen optimizer may change the execution plan based on the recomputed statistics and index changes. Because the execution plan may vary, these examples are included for demonstration purposes only. Examples include:

Using statement level optimizer hints for a SELECT query

View the execution plan for a query. Then use statement level optimizer hints to influence the optimizer to choose a different execution plan. Consider the query:

Command> SELECT r.region_name, c.country_name
         FROM regions r, countries c
         WHERE r.region_id = c.region_id
         ORDER BY c.region_id;

Use the ttIsql EXPLAIN command to view the plan generated by the optimizer. Note:

  • The optimizer performs two range scans using table level locking for both scans.

  • The optimizer uses the MergeJoin operation to join the two tables.

Command> EXPLAIN SELECT r.region_name, c.country_name
         FROM regions r, countries c
         WHERE r.region_id = c.region_id
         ORDER BY c.region_id;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               2
  OPERATION:           TblLkRangeScan
  TBLNAME:             COUNTRIES
  IXNAME:              COUNTR_REG_FK
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>
 
 
  STEP:                2
  LEVEL:               2
  OPERATION:           TblLkRangeScan
  TBLNAME:             REGIONS
  IXNAME:              REGIONS
  INDEXED CONDITION:   R.REGION_ID >= C.REGION_ID
  NOT INDEXED:         <NULL>
 
  STEP:                3
  LEVEL:               1
  OPERATION:           MergeJoin
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   C.REGION_ID = R.REGION_ID
  NOT INDEXED:         <NULL>

Now use statement level optimizer hints to direct the optimizer to perform the scans using row level locking and to use a NestedLoop operation to join the tables. Set autocommit to on to illustrate that the autocommit setting has no effect because statement level optimizer hints are scoped to the SQL statement.

Command> autocommit on;
Command> EXPLAIN SELECT /*+ TT_RowLock (1), TT_TblLock (0), TT_MergeJoin (0),
         TT_NestedLoop (1) */
         r.region_name, c.country_name
         FROM regions r, countries c
         WHERE r.region_id = c.region_id
         ORDER BY c.region_id;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               3
  OPERATION:           RowLkRangeScan
  TBLNAME:             REGIONS
  IXNAME:              REGIONS
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>
 
  STEP:                2
  LEVEL:               3
  OPERATION:           RowLkRangeScan
  TBLNAME:             COUNTRIES
  IXNAME:              COUNTR_REG_FK
  INDEXED CONDITION:   C.REGION_ID = R.REGION_ID
  NOT INDEXED:         <NULL>
 
  STEP:                3
  LEVEL:               2
  OPERATION:           NestedLoop
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>
 
  STEP:                4
  LEVEL:               1
  OPERATION:           OrderBy
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>

Prepare the query again without statement level optimizer hints. The optimizer reverts back to the original execution plan because statement level optimizer hints are scoped to the SQL statement.

Command> EXPLAIN SELECT r.region_name, c.country_name
         FROM regions r, countries c
         WHERE r.region_id = c.region_id
         ORDER BY c.region_id;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               2
  OPERATION:           TblLkRangeScan
  TBLNAME:             COUNTRIES
  IXNAME:              COUNTR_REG_FK
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>
 
  STEP:                2
  LEVEL:               2
  OPERATION:           TblLkRangeScan
  TBLNAME:             REGIONS
  IXNAME:              REGIONS
  INDEXED CONDITION:   R.REGION_ID >= C.REGION_ID
  NOT INDEXED:         <NULL>
 
  STEP:                3
  LEVEL:               1
  OPERATION:           MergeJoin
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   C.REGION_ID = R.REGION_ID
  NOT INDEXED:         <NULL>

Using on and off hinting

This example illustrates the importance of directing the optimizer to specifically enable or disable hints that perform a similar function. For example, the hash and range hints direct the optimizer to use either a hash or range access path for the table. In order to ensure the optimizer chooses the specific access path, enable one hint and disable all other related hints.

Create a table and create a hash index on the first column of the table and a range index on the second column.

Command> CREATE TABLE test (col1 NUMBER, col2 NUMBER);
Command> CREATE HASH INDEX h_index ON test (col1);
Command> CREATE INDEX hr_index ON test (col2);

Set autocommit to off and execute the built-in procedure, ttOptGetFlag, to review the current transaction level optimizer hint settings for the transaction. A setting of 1 means the flag is enabled.

Command> autocommit off;
Command> CALL ttOptGetFlag ('Hash');
< Hash, 1 >
1 row found.
Command> CALL ttOptGetFlag ('Scan');
< Scan, 1 >
1 row found.

Use the ttIsql EXPLAIN command to review the plan for a SELECT query using a WHERE clause and dynamic parameters. The optimizer uses a hash scan.

Command> EXPLAIN SELECT * FROM test WHERE col1 = ? and col2 = ?;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               1
  OPERATION:           RowLkHashScan
  TBLNAME:             TEST
  IXNAME:              H_INDEX
  INDEXED CONDITION:   TEST.COL1 = _QMARK_1
  NOT INDEXED:         TEST.COL2 = _QMARK_2

Use the statement level optimizer hint TT_Range to direct the optimizer to use a range scan. Note that the optimizer ignores the TT_Range hint and uses a hash scan because you did not direct the optimizer to disable the hash scan. Alter the statement and direct the optimizer to use a range scan and not use a hash scan. To accomplish this, enable the statement level optimizer hint TT_Range and disable the statement level optimizer hint TT_HashScan. The optimizer no longer ignores the TT_Range hint.

Command> EXPLAIN SELECT --+ TT_Range (1) Single line comment to set TT_Range
         * FROM TEST WHERE col1 = ? and col2 = ?;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               1
  OPERATION:           RowLkHashScan
  TBLNAME:             TEST
  IXNAME:              H_INDEX
  INDEXED CONDITION:   TEST.COL1 = _QMARK_1
  NOT INDEXED:         TEST.COL2 = _QMARK_2

Command> EXPLAIN SELECT /*+ TT_Range (1) TT_HashScan (0)
          Multiple line comment to enable TT_Range and disable TT_HashScan */
          * FROM TEST WHERE col1 = ? and col2 = ?;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               1
  OPERATION:           RowLkRangeScan
  TBLNAME:             TEST
  IXNAME:              HR_INDEX
  INDEXED CONDITION:   TEST.COL2 = _QMARK_2
  NOT INDEXED:         TEST.COL1 = _QMARK_1

Prepare the query again without using statement level optimizer hints and without issuing a commit or rollback. The optimizer uses the transaction level optimizer hints settings that were in effect before executing the query. The optimizer uses transaction level optimizer hints because statement level optimizer hints are scoped to the SQL statement.

Command> EXPLAIN SELECT * FROM TEST WHERE col1 = ? and col2 = ?;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               1
  OPERATION:           RowLkHashScan
  TBLNAME:             TEST
  IXNAME:              H_INDEX
  INDEXED CONDITION:   TEST.COL1 = _QMARK_1
  NOT INDEXED:         TEST.COL2 = _QMARK_2

Using TT_JoinOrder to specify a join order

Use the statement level optimizer hint TT_JoinOrder to direct the optimizer to use a specific join order. First use a transaction level optimizer hint to direct the optimizer to use a specific join order for the transaction. Then use a statement level optimizer hint to direct the optimizer to change the join order for the statement only.

Command> CALL ttOptSetOrder ('e d j');
Command> EXPLAIN SELECT *
         FROM employees e, departments d, job_history j
         WHERE e.department_id = d.department_id
         AND e.hire_date = j.start_date;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               3
  OPERATION:           TblLkRangeScan
  TBLNAME:             EMPLOYEES
  IXNAME:              EMP_DEPT_FK
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>
 
  STEP:                2
  LEVEL:               3
  OPERATION:           TblLkRangeScan
  TBLNAME:             DEPARTMENTS
  IXNAME:              DEPARTMENTS
  INDEXED CONDITION:   D.DEPARTMENT_ID >= E.DEPARTMENT_ID
  NOT INDEXED:         <NULL>
 
  STEP:                3
  LEVEL:               2
  OPERATION:           MergeJoin
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   E.DEPARTMENT_ID = D.DEPARTMENT_ID
  NOT INDEXED:         <NULL>
 
  STEP:                4
  LEVEL:               2
  OPERATION:           TblLkRangeScan
  TBLNAME:             JOB_HISTORY
  IXNAME:              JOB_HISTORY
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         E.HIRE_DATE = J.START_DATE
 
  STEP:                5
  LEVEL:               1
  OPERATION:           NestedLoop
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>

Use the statement level optimizer hint, TT_JoinOrder, to direct the optimizer to override the transaction level join order optimizer hint for the SQL statement only.

Command> EXPLAIN SELECT --+ TT_JoinOrder (e j d)
          *
          FROM employees e, departments d, job_history j
          WHERE e.department_id = d.department_id
          AND e.hire_date = j.start_date;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               3
  OPERATION:           TblLkRangeScan
  TBLNAME:             EMPLOYEES
  IXNAME:              EMP_DEPT_FK
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>
 
  STEP:                2
  LEVEL:               3
  OPERATION:           TblLkRangeScan
  TBLNAME:             JOB_HISTORY
  IXNAME:              JOB_HISTORY
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         E.HIRE_DATE = J.START_DATE
 
  STEP:                3
  LEVEL:               2
  OPERATION:           NestedLoop
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>
 
  STEP:                4
  LEVEL:               2
  OPERATION:           TblLkRangeScan
  TBLNAME:             DEPARTMENTS
  IXNAME:              DEPARTMENTS
  INDEXED CONDITION:   D.DEPARTMENT_ID >= E.DEPARTMENT_ID
  NOT INDEXED:         <NULL>
 
  STEP:                5
  LEVEL:               1
  OPERATION:           MergeJoin
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   E.DEPARTMENT_ID = D.DEPARTMENT_ID
  NOT INDEXED:         <NULL>

Prepare the query again to verify that the join order that was in effect for the transaction remains in effect.

Command> EXPLAIN SELECT *
          FROM employees e, departments d, job_history j
          WHERE e.department_id = d.department_id
          AND e.hire_date = j.start_date;
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               3
  OPERATION:           TblLkRangeScan
  TBLNAME:             EMPLOYEES
  IXNAME:              EMP_DEPT_FK
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>
 
  STEP:                2
  LEVEL:               3
  OPERATION:           TblLkRangeScan
  TBLNAME:             DEPARTMENTS
  IXNAME:              DEPARTMENTS
  INDEXED CONDITION:   D.DEPARTMENT_ID >= E.DEPARTMENT_ID
  NOT INDEXED:         <NULL>
 
  STEP:                3
  LEVEL:               2
  OPERATION:           MergeJoin
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   E.DEPARTMENT_ID = D.DEPARTMENT_ID
  NOT INDEXED:         <NULL>
 
  STEP:                4
  LEVEL:               2
  OPERATION:           TblLkRangeScan
  TBLNAME:             JOB_HISTORY
  IXNAME:              JOB_HISTORY
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         E.HIRE_DATE = J.START_DATE
 
  STEP:                5
  LEVEL:               1
  OPERATION:           NestedLoop
  TBLNAME:             <NULL>
  IXNAME:              <NULL>
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         <NULL>

Using the statement level optimizer hint TT_INDEX

Perform a query on the employees table that uses the index, emp_name_ix. Then use the statement level optimizer hint TT_INDEX to direct the optimizer not to use this index. First run the ttIsql command, indexes, to view the indexes for the employees table.

Command> indexes employees;
 
Indexes on table TESTUSER.EMPLOYEES:
  EMPLOYEES: unique range index on columns:
    EMPLOYEE_ID
    (referenced by foreign key index JHIST_EMP_FK on table TESTUSER.JOB_HISTORY)
  TTUNIQUE_0: unique range index on columns:
    EMAIL
  EMP_DEPT_FK: non-unique range index on columns:
    DEPARTMENT_ID
    (foreign key index references table TESTUSER.DEPARTMENTS(DEPARTMENT_ID))
  EMP_JOB_FK: non-unique range index on columns:
    JOB_ID
    (foreign key index references table TESTUSER.JOBS(JOB_ID))
  EMP_NAME_IX: non-unique range index on columns:
    LAST_NAME
    FIRST_NAME
  5 indexes found.
 
5 indexes found on 1 table.

Use the ttIsql command, EXPLAIN, to view the execution plan for a SELECT query on the employees table that uses a WHERE clause on the last_name column.

Command> EXPLAIN SELECT e.first_name
           FROM employees e
           WHERE e.last_name BETWEEN 'A' AND 'B';
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               1
  OPERATION:           RowLkRangeScan
  TBLNAME:             EMPLOYEES
  IXNAME:              EMP_NAME_IX
  INDEXED CONDITION:   E.LAST_NAME >= 'A' AND E.LAST_NAME <= 'B'
  NOT INDEXED:         <NULL>

Use the statement level optimizer hint, TT_INDEX, to direct the optimizer not to use the index, emp_name_ix.

Command> EXPLAIN SELECT --+ TT_INDEX (E,EMP_NAME_IX,0)
           e.first_name
           FROM employees e
           WHERE e.last_name BETWEEN 'A' AND 'B';
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               1
  OPERATION:           TblLkRangeScan
  TBLNAME:             EMPLOYEES
  IXNAME:              EMPLOYEES
  INDEXED CONDITION:   <NULL>
  NOT INDEXED:         E.LAST_NAME <= 'B' AND E.LAST_NAME >= 'A'

Optimizer hints supported in TimesTen Scaleout only

These optimizer hints are only supported in TimesTen Scaleout. They are valid at the statement and at the connection levels.

See "OptimizerHint" in the Oracle TimesTen In-Memory Database Reference for information on hints at the connection level and "Statement level optimizer hints" in this book for information on statement level optimizer hints.

TT_GridQueryExec optimizer hint

The TT_GridQueryExec optimizer hint enables you to specify whether the query should return data from the local element or from all elements, including the elements in a replica set when K-safety is set to 2.

If you do not specify this hint, the query is executed in one logical data space. It is neither local nor global. This means that exactly one full copy of the data is used to compute the query. Use this hint in cases where obtaining some result is more important than obtaining the correct result (for example, where one or more replica sets are unavailable). Valid options for this hint are LOCAL and GLOBAL.

For more information, see:

  • "TT_GridQueryExec" in the Oracle TimesTen In-Memory Database Scaleout User's Guide for information on using this hint.

  • "OptimizerHint" in the Oracle TimesTen In-Memory Database Reference for information on using this hint at the connection level.

  • "Statement level optimizer hints" for information on using this hint at the statement level.

This example illustrates how to use the TT_GridQueryExec(GLOBAL) hint on the dual table to determine the ids of all elements, replica sets, and dataspaces.

Command> SELECT /*+TT_GridQueryExec(GLOBAL)*/ elementId#, replicasetId#,
           dataspaceId# FROM dual ORDER BY elementId#,replicasetId#,dataspaceId#;
 
ELEMENTID#, REPLICASETID#, DATASPACEID#
< 1, 1, 1 >
< 2, 1, 2 >
< 3, 2, 1 >
< 4, 2, 2 >
< 5, 3, 1 >
< 6, 3, 2 >
6 rows found.

See "TT_GridQueryExec" in the Oracle TimesTen In-Memory Database Scaleout User's Guide for more examples.

TT_PartialResult optimizer hint

The TT_PartialResult optimizer hint enables you to specify whether the query should return partial results if some data is not available.

Use TT_PartialResult(1) to direct the query to return partial results if all elements in a replica set are not available.

Use TT_PartialResult(0) to direct the query to return an error if the required data is not available in the case where all elements in a replica set are not available. If at least one element from each replica set is available or the data required by the query is available, the optimizer returns the query result correctly without error.

The default is TT_PartialResult(0).

For more information, see:

  • "TT_PartialResult" in the Oracle TimesTen In-Memory Database Scaleout User's Guide for information on using this hint and for examples.

  • "OptimizerHint" in the Oracle TimesTen In-Memory Database Reference for information on using this hint at the connection level.

  • "Statement level optimizer hints" for information on using this hint at the statement level.

TT_CommitDMLOnSuccess optimizer hint

Use the TT_CommitDMLOnSuccess hit to enable or disable a commit operation as part of DML execution.

  • At the statement level, TT_CommitDMLOnSuccess is used in a DML statement (DELETE, INSERT, INSERT... SELECT, and UPDATE) to enable or disable the commit behavior of the transaction when the DML operation is executed. For the INSERT...SELECT statement, specify TT_CommitDMLOnSuccess after the SELECT keyword.

    TT_CommitDMLOnSuccess is valid in DML operations only. It is not valid for queries or DDL operations and, if specified in a non-DML statement, is ignored and no error is returned. See "Statement level optimizer hints" for information on the syntax and semantics.

  • At the connection level, TT_CommitDMLOnSuccess is also used to enable or disable the commit behavior of the transaction when a DML operation is executed. However, you specify TT_CommitDMLOnSuccess as a parameter to the OptimizerHint connection attribute. See "OptimizerHint" in the Oracle TimesTen In-Memory Database Reference for information on using TT_CommitDMLOnSuccess at the connection level.

At both levels, valid options are 0 and 1. If you do not specify TT_CommitDMLOnSuccess, there are no changes to the normal commit behavior. The order of precedence is statement level followed by connection level.

The TT_CommitDMLOnSuccess commit behavior at the statement level is:

  • TT_CommitDMLOnSuccess(1) commits the current transaction if the DML statement in which the hint is specified is executed successfully. If there are open cursors at commit time, all cursors are closed and the transaction is committed. If the statement with this hint fails, the transaction is not committed.

  • TT_CommitDMLOnSuccess(0) disables the commit of the current transaction if the DML statement in which the hint is specified is executed successfully.

Table 6-4 shows the commit behavior when not setting TT_CommitDMLOnSuccess as well as setting TT_CommitDMLOnSuccess to 0 and 1 at the statement and connection levels. The table shows the commit behavior when autocommit is set to 0.

Table 6-5 shows the commit behavior when not setting TT_CommitDMLOnSuccess as well as setting TT_CommitDMLOnSuccess to 0 and 1 at the statement and connection levels. The table shows the commit behavior when autocommit is set to 1.

Table 6-4 TT_CommitDMLOnSuccess commit behavior: Autocommit 0

Blank Not set at connection level Set to 0 at connection level Set to 1 at connection level

Not set at statement level

  • If there are no open cursors, and DML execution is successful, the transaction is not committed.

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is not committed.

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed due to the connection level setting of 1.

  • If there are open cursors and DML execution is successful, the cursors are closed and the transaction is committed due to the connection level setting of 1.

  • If DML execution is not successful, the transaction is not committed.

Set to 0 at statement level

  • If there are no open cursors, and DML execution is successful, the transaction is not committed up until the DML statement that includes the hint due to the autocommit 0 setting. This DML statement is not committed due to disabling the commit behavior at the statement level. (This is the same behavior as when TT_CommitDMLOnSuccess is not set at the statement level.)

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed. (This is the same behavior as when TT_CommitDMLOnSuccess is not set at the statement level.)

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is not committed.

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed up until the DML statement that includes the hint. This DML statement is not committed due to disabling the commit behavior at the statement level.

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed.

  • If DML execution is not successful, the transaction is not committed.

Set to 1 at statement level

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, cursors are closed and the transaction is committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, cursors are closed and the transaction is committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, cursors are closed and the transaction is committed.

  • If DML execution is not successful, the transaction is not committed.

Table 6-5 TT_CommitDMLOnSuccess commit behavior: Autocommit 1

Blank Not set at connection level Set to 0 at connection level Set to 1 at connection level

Not set at statement level

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, the cursors are closed and the transaction is committed.

  • If DML execution is not successful, the transaction is not committed.

Set to 0 at statement level

  • If there are no open cursors, and DML execution is successful, the transaction is committed. (This is the same behavior as when TT_CommitDMLOnSuccess is not set at the statement level.)

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed. (This is the same behavior as when TT_CommitDMLOnSuccess is not set at the statement level.)

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed.

  • If DML execution is not successful, the transaction is not committed

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, the cursors are not closed and the transaction is not committed.

  • If DML execution is not successful, the transaction is not committed.

Set to 1 at statement level

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, cursors are closed and the transaction is committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, cursors are closed and the transaction is committed.

  • If DML execution is not successful, the transaction is not committed.

  • If there are no open cursors, and DML execution is successful, the transaction is committed.

  • If there are open cursors and DML execution is successful, cursors are closed and the transaction is committed.

  • If DML execution is not successful, the transaction is not committed.

For more information, see:

These examples illustrate the use of the TT_CommitDMLOnSuccess optimizer hint:

Setting TT_CommitDMLOnSuccess to 1

This example first creates the mytable table. It then sets autocommit to 0 and inserts a row into the mytable table. A second connection (conn2) connects to the database and issues a SELECT query against the mytable table. The query returns 0 rows. The ttIsql use command returns the application to the first connection (database1) and issues a second INSERT operation, setting TT_CommitDMLOnSuccess to 1 at the statement level. A second ttIsql use command returns the application to the conn2 connection. A SELECT query shows two rows have been inserted into the mytable table. This example illustrates that issuing TT_CommitDMLOnSuccess(1) commits the transaction after the successful execution of the second INSERT operation (which set the hint).

Command> CREATE TABLE mytable (col1 TT_INTEGER, col2 VARCHAR2(4000));
Command> autocommit 0;
Command> INSERT INTO mytable VALUES (10, 'ABC');
1 row inserted.

Establish a second connection (conn2)

Command> connect as conn2;
Using the connection string of connection database1 to connect...
...
(Default setting AutoCommit=1)

Issue a SELECT query and expect 0 rows due to autocommit set to 0.

conn2: Command> SELECT * FROM mytable;
0 rows found.

Return to the first connection (database1) and issue an INSERT operation with TT_CommitDMLOnSuccess set to 1.

conn2: Command> use database1;
database1: Command> INSERT /*+TT_CommitDMLOnSuccess(1)*/ 
                       INTO mytable VALUES (10, 'ABC');
1 row inserted.

Return to the second connection (conn2) and issue a SELECT query. Expect 2 rows (due to the two INSERT statements. (The transaction is committed due to the TT_CommitDMLOnSuccess statement level hint set to 1 and the successful execution of the two INSERT operations.)

database1: Command> use conn2
conn2: Command> SELECT * FROM mytable;
< 10, ABC >
< 10, ABC >
2 rows found.

Using TT_CommitDMLOnSuccess at connection level

This example first creates the mytable table. It then uses PL/SQL to insert 1000 rows into the table. There is a second connection to the database (conn2) and this connection connects with TT_CommitDMLOnSuccess set to 1 at the connection level. Various operations are performed to illustrate the behavior of TT_CommitDMLOnSuccess at both the statement and connection levels.

Command>  CREATE TABLE mytable (col1 TT_INTEGER NOT NULL PRIMARY KEY, 
            col2 VARCHAR2 (4000));
Command> BEGIN
           FOR i in 1..1000
           LOOP
            INSERT INTO mytable VALUES (i,i);
          END LOOP;
        END;
        /
PL/SQL procedure successfully completed.

Establish a second connection (conn2) and connect setting TT_CommitDMLOnSuccess at the connection level to 1.

Command> CONNECT adding "OptimizerHint=TT_CommitDMLOnSuccess(1)" as conn2;
Connection successful: 
...

Set autocommit to 0 and issue a DELETE operation.

conn2: Command> autocommit 0;
conn2: Command> DELETE FROM mytable WHERE col1=1000;
1 row deleted.

Return to the original connection (database1) and issue a SELECT query to see if the DELETE operation was committed. The operation was committed due to the TT_CommitDMLOnSuccess setting of 1 at the connection level.

conn2: Command> use database1;
database1: Command> SELECT * FROM mytable WHERE col1=1000;
0 rows found.

Return to the second connection (conn2) and issue an INSERT operation. Then return to the original connection (database1). The transaction containing the INSERT operation was committed.

database1: Command> use conn2;
conn2: Command> INSERT INTO mytable VALUES (1000,1000);
1 row inserted.
conn2: Command> use database1
database1: Command> SELECT * FROM mytable WHERE col1=1000;
< 1000, 1000 >
1 row found.

Return to the second connection (conn2) and issue a DELETE operation, followed by an INSERT operation, and then a second INSERT operation where TT_CommitDMLOnSuccess is set to 0 at the statement level (the second INSERT).

database1: Command> use conn2;
conn2: Command> DELETE FROM mytable WHERE col1=1000;
1 row deleted.
conn2: Command> INSERT INTO mytable VALUES (1001,1001);
1 row inserted.
conn2: Command> INSERT /*+TT_CommitDMLOnSuccess(0)*/ INTO mytable 
                  VALUES (1002,1002);
1 row inserted.

Issue a SELECT query and notice the results of the query. The one DELETE operation and the two INSERT operations were successful.

conn2: Command> SELECT * FROM mytable where col1 >= 1000;
< 1001, 1001 >
< 1002, 1002 >
2 rows found.

Return to the original connection (database1) and issue the same SELECT query. Observe that the one DELETE statement and the first INSERT operation were committed. This is due to the TT_CommitDMLOnSuccess setting of 1 at the connection level. The second INSERT statement was not committed due to the TT_CommitDMLOnSuccess setting of 0 for this second INSERT statement.

conn2: Command> use database1;
database1: Command> SELECT * FROM mytable where col1 >= 1000;
< 1001, 1001 >
1 row found.

Return to the second connection (conn2) and issue a third INSERT operation. Then issue a SELECT query and observe the results.

database1: Command> use conn2;
conn2: Command> INSERT INTO mytable VALUES (1003,1003);
1 row inserted.
conn2: Command> SELECT * FROM mytable where col1 >= 1000 ORDER BY col1;
< 1001, 1001 >
< 1002, 1002 >
< 1003, 1003 >
3 rows found.

Return to the original connection (database1) and issue the same SELECT query. Note the results are the same as in the conn2 connection. The transaction is committed due to the TT_CommitDMLOnSuccess setting of 1 at the connection level and the successful execution of the second and third INSERT operations.

conn2: Command> use database1
database1: Command> SELECT * FROM mytable where col1 >= 1000 ORDER BY col1;
< 1001, 1001 >
< 1002, 1002 >
< 1003, 1003 >
3 rows found.

ALTER ACTIVE STANDBY PAIR

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

You can change an active standby pair by:

  • Adding or dropping a subscriber database

  • Altering store attributes

    Only the PORT and TIMEOUT attributes can be set for subscribers.

  • Including tables, sequences or cache groups in the replication scheme

  • Excluding tables, sequences or cache groups from the replication scheme

See "Making other changes to an active standby pair" in Oracle TimesTen In-Memory Database Replication Guide.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

ALTER ACTIVE STANDBY PAIR { 
      SubscriberOperation | 
      StoreOperation | InclusionOperation |
      NetworkOperation } [...]

Syntax for SubscriberOperation:

{ADD | DROP } SUBSCRIBER FullStoreName

Syntax for StoreOperation:

ALTER STORE FullStoreName SET StoreAttribute

Syntax for InclusionOperation:

    [{ INCLUDE | EXCLUDE }{TABLE [[Owner.]TableName [,...]]|
         CACHE GROUP [[Owner.]CacheGroupName [,...]]|
         SEQUENCE [[Owner.]SequenceName [,...]]} [,...]]

Syntax for NetworkOperation:

ADD ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName
      { { MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost }
          PRIORITY Priority } [...] 
DROP ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName
      { MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost } [...] 

Parameters

Parameter Description

ADD SUBSCRIBER FullStoreName

Indicates a subscriber database. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

DROP SUBSCRIBER FullStoreName

Indicates that updates should no longer be sent to the specified subscriber database. This operation fails if the replication scheme has only one subscriber. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

ALTER STORE FullStoreName SET StoreAttribute

Indicates changes to the attributes of a database. Only the PORT and TIMEOUT attributes can be set for subscribers. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

See "CREATE ACTIVE STANDBY PAIR" For information on StoreAttribute clauses.

FullStoreName

The database, specified as one of the following:

  • SELF

  • The prefix of the database file name

For example, if the database path is directory/subdirectory/data.ds0, then data is the database name that should be used.

This is the database file name specified in the DataStore attribute of the DSN description with optional host ID in the form:

DataStoreName [ON Host]

Host can be either an IP address or a literal host name assigned to one or more IP addresses, as described in "Configuring the network" in Oracle TimesTen In-Memory Database Replication Guide. Host names containing special characters must be surrounded by double quotes. For example: "MyHost-500".

{INCLUDE|EXCLUDE}

{[TABLE [Owner.]TableName[,...]|

CACHE GROUP

[[Owner.]CacheGroupName]|[,...]

SEQUENCE [[Owner.]SequenceName [,...]}

[,...]

Includes in or excludes from replication the tables, sequences or cache groups listed.

INCLUDE adds the tables, sequences or cache groups to the replication scheme. Use one INCLUDE clause for each object type (table, sequence or cache group).

EXCLUDE removes the tables, sequences or cache groups from replication. Use one EXCLUDE clause for each object type (table, sequence or cache group).

You cannot use the EXCLUDE clause for AWT cache groups.

ADD ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName

Adds NetworkOperation to replication scheme. Enables you to control the network interface that a master store uses for every outbound connection to each of its subscriber stores. In the context of the ADD ROUTE clause, each master database is a subscriber of the other master database and each read-only subscriber is a subscriber of both master databases.

Can be specified more than once.

For FullStoreName, "ON host" must be specified.

DROP ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName

Drops NetworkOperation from replication scheme.

Can be specified more than once.

For FullStoreName, "ON host" must be specified.

MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost

MasterHost and SubscriberHost are the IP addresses for the network interface on the master and subscriber stores. Specify in dot notation or canonical format or in colon notation for IPV6.

Clause can be specified more than once. Valid for both ADD and DROP ROUTE MASTER.

PRIORITY Priority

Variable expressed as an integer from 1 to 99. Denotes the priority of the IP address. Lower integral values have higher priority. An error is returned if multiple addresses with the same priority are specified. Controls the order in which multiple IP addresses are used to establish peer connections.

Required syntax of NetworkOperation clause. Follows MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost clause.

Description

  • You must stop the replication agent before altering an active standby pair. The exceptions are for those objects and statements that are automatically replicated and included based on the values of the DDL_REPLICATION_LEVEL and DDL_REPLICATION_ACTION attributes. See "ALTER SESSION" for more information.

  • You may only alter the active standby pair replication scheme on the active database. See "Making other changes to an active standby pair" in Oracle TimesTen In-Memory Database Replication Guide for more information.

  • You may not use ALTER ACTIVE STANDBY PAIR when using Oracle Clusterware with TimesTen. See "Restricted commands and SQL statements" in Oracle TimesTen In-Memory Database Replication Guide for more information.

    Instead, perform the tasks described in "Changing the schema" section of the Oracle TimesTen In-Memory Database Replication Guide.

  • Use ADD SUBSCRIBER FullStoreName to add a subscriber to the replication scheme.

  • Use DROP SUBSCRIBER FullStoreName to drop a subscriber from the replication scheme.

  • Use the INCLUDE or EXCLUDE clause to include the listed tables, sequences or cache groups in the replication scheme or to exclude them from the replication scheme. Use one INCLUDE or EXCLUDE clause for each object type (table, sequence or cache group). The ALTER ACTIVE STANDBY statement is not necessary for those objects and statements that are automatically replicated and included based on the values of the DDL_REPLICATION_LEVEL and DDL_REPLICATION_ACTION attributes. See "ALTER SESSION" for more information. However, if DDL_REPLICATION_LEVEL is 2 or greater and DDL_REPLICATION_ACTION="EXCLUDE", use the INCLUDE clause to include replicated objects into the replication scheme.

  • Do not use the EXCLUDE clause for AWT cache groups.

  • When DDL_REPLICATION_LEVEL is 2 or greater, the INCLUDE clause can only be used with empty tables on the active database. The contents of the corresponding tables on the standby and any subscribers will be truncated before the table is added to the replication scheme.

Examples

Add a subscriber to the replication scheme.

ALTER ACTIVE STANDBY PAIR
    ADD SUBSCRIBER rep4;

Drop two subscribers from the replication scheme.

ALTER ACTIVE STANDBY PAIR
    DROP SUBCRIBER rep3
    DROP SUBSCRIBER rep4;

Alter the store attributes of the rep3 and rep4 databases.

ALTER ACTIVE STANDBY PAIR
    ALTER STORE rep3 SET PORT 23000 TIMEOUT 180
    ALTER STORE rep4 SET PORT 23500 TIMEOUT 180;

Add a table, a sequence and two cache groups to the replication scheme.

ALTER ACTIVE STANDBY PAIR
    INCLUDE TABLE my.newtab
    INCLUDE SEQUENCE my.newseq
    INCLUDE CACHE GROUP my.newcg1, my.newcg2;

Add NetworkOperation clause to active standby pair:

ALTER ACTIVE STANDBY PAIR
ADD ROUTE MASTER rep1 ON "machine1" SUBSCRIBER rep2 ON "machine2"
MASTERIP "1.1.1.1" PRIORITY 1 SUBSCRIBERIP "2.2.2.2" PRIORITY 1;

ALTER CACHE GROUP

The ALTER CACHE GROUP statement modifies the state, interval and mode of AUTOREFRESH for a cache group.

Updates on the Oracle Database tables can be propagated back to the TimesTen cache group with the use of AUTOREFRESH. AUTOREFRESH can be enabled when the cache group is a user managed cache group or is defined as READONLY with an AUTOREFRESH clause.

Any values or states set by ALTER CACHE GROUP are persistent. They are stored in the database and survive daemon and cache agent restarts.

Required privilege

No privilege is required for the cache group owner.

ALTER ANY CACHE GROUP for another user's cache group.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

This statement changes the AUTOREFRESH mode of the cache group, which determines which rows are updated during an autorefresh operation. You cannot use the ALTER CACHE GROUP...SET AUTOREFRESH MODE clause in TimesTen Scaleout.

ALTER CACHE GROUP [Owner.]GroupName
        SET AUTOREFRESH MODE
        {INCREMENTAL | FULL}

This statement changes the AUTOREFRESH interval on the cache group. You cannot use the ALTER CACHE GROUP...SET AUTOREFRESH INTERVAL clause in TimesTen Scaleout.

ALTER CACHE GROUP [Owner.]GroupName 
        SET AUTOREFRESH INTERVAL IntervalValue 
        {MINUTE[S] | SECOND[S] | MILLISECOND[S]}

This statement alters the AUTOREFRESH state:

ALTER CACHE GROUP [Owner.]GroupName 
        SET AUTOREFRESH STATE
        {ON | OFF | PAUSED}

Parameters

Parameter Description

[Owner.]GroupName

Name assigned to the new cache group.

AUTOREFRESH

Indicates that changes to the Oracle Database tables should be automatically propagated to TimesTen.

MODE

Determines which rows in the cache are updated during an autorefresh. If the INCREMENTAL clause is specified, TimesTen refreshes only rows that have been changed on the Oracle Database since the last propagation. If the FULL clause is specified or if there is neither FULL nor INCREMENTAL clause specified, TimesTen updates all rows in the cache with each autorefresh. The default mode is INCREMENTAL.

INTERVAL

IntervalValue

An integer value that specifies how often AUTOREFRESH should be scheduled, in minutes, seconds or milliseconds. The default value is five minutes. An autorefresh interval set to 0 milliseconds enables continuous autorefresh, where the next autorefresh cycle is scheduled immediately after the last autorefresh cycle has ended. See "Automatically refreshing a cache group" in the Oracle TimesTen In-Memory Database Cache Guide for more information.

If the specified interval is not long enough for an AUTOREFRESH to complete, a runtime warning is generated and the next AUTOREFRESH waits until the current one finishes. An informational message is generated in the support log if the wait queue reaches 10.

STATE

Specifies whether AUTOREFRESH should be changed to on, off or paused. By default, the AUTOREFRESH STATE is ON.

ON

AUTOREFRESH is scheduled to occur at the specified interval.

OFF

A scheduled AUTOREFRESH is canceled, and TimesTen does not try to maintain the information necessary for an INCREMENTAL refresh. Therefore if AUTOREFRESH is turned on again at a later time, the first refresh is FULL.

PAUSED

A scheduled AUTOREFRESH is canceled, but TimesTen tries to maintain the information necessary for an INCREMENTAL refresh. Therefore if AUTOREFRESH is turned on again at a later time, a full refresh may not be necessary.

Description

  • A refresh does not occur immediately after issuing ALTER CACHE GROUP...SET AUTOREFRESH STATE. This statement only changes the state of AUTOREFRESH. When the transaction that contains the ALTER CACHE GROUP statement is committed, the cache agent is notified to schedule an AUTOREFRESH immediately, but the commit goes through without waiting for the completion of the refresh. The scheduling of the autorefresh operation is part of the transaction, but the refresh itself is not.

  • If you issue an ALTER CACHE GROUP... SET AUTOREFRESH STATE OFF statement and there is an autorefresh operation currently running, then:

    • If LockWait interval is 0, the ALTER statement fails with a lock timeout error.

    • If LockWait interval is nonzero, then the current autorefresh transaction is rolled back, and the ALTER statement continues. This affects all cache groups with the same autorefresh interval.

  • Replication cannot occur between cache groups with AUTOREFRESH and cache groups without AUTOREFRESH.

  • If the ALTER CACHE GROUP statement is part of a transaction that is being replicated, and if the replication scheme has the RETURN TWOSAFE attribute, the transaction may fail.

  • You cannot execute the ALTER CACHE GROUP statement when performed under the serializable isolation level. An error message is returned when attempted.

ALTER FUNCTION

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

The ALTER FUNCTION statement recompiles a standalone stored function. Explicit recompilation eliminates the need for implicit runtime recompilation and prevents associated runtime compilation errors and performance overhead.

To recompile a function that is part of a package, recompile the package using the ALTER PACKAGE statement.

Required privilege

No privilege is required for the PL/SQL function owner.

ALTER ANY PROCEDURE for another user's function.

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

ALTER FUNCTION [Owner.]FunctionName COMPILE
      [CompilerParametersClause [...]] 
      [REUSE SETTINGS]

Parameters

Parameter Description

[Owner.]FunctionName

Name of the function to be recompiled.

COMPILE

Required keyword that causes recompilation of the function. If the function does not compile successfully, use the ttIsql command SHOW ERRORS to display the compiler error messages.

CompilerParametersClause

Use this optional clause to specify a value for one of the PL/SQL persistent compiler parameters. The PL/SQL persistent compiler parameters are PLSQL_OPTIMIZE_LEVEL and NLS_LENGTH_SEMANTICS.

You can specify each parameter once in the statement.

If you omit a parameter from this clause and you specify REUSE SETTINGS, then if a value was specified for the parameter in an earlier compilation, TimesTen uses that earlier value. If you omit a parameter and either you do not specify REUSE SETTINGS or no value has been specified for the parameter in an earlier compilation, then TimesTen obtains the value for the parameter from the session environment.

REUSE SETTINGS

Use this optional clause to prevent TimesTen from dropping and reacquiring compiler switch settings. When you specify REUSE SETTINGS, TimesTen preserves the existing settings and uses them for the compilation of any parameters for which values are not specified.

Description

  • The ALTER FUNCTION statement does not change the declaration or definition of an existing function. To redeclare or redefine a function, use the CREATE FUNCTION statement.

  • TimesTen first recompiles objects upon which the function depends, if any of those objects are invalid.

  • TimesTen also invalidates any objects that depend on the function, such as functions that call the recompiled function or package bodies that define functions that call the recompiled function.

  • If TimesTen recompiles the function successfully, then the function becomes valid. If recompiling the function results in compilation errors, then TimesTen returns an error and the function remains invalid. Use the ttIsql command SHOW ERRORS to display compilation errors.

  • During recompilation, TimesTen drops all persistent compiler settings, retrieves them again from the session, and stores them at the end of compilation. To avoid this process, specify the REUSE SETTINGS clause.

See also

ALTER PACKAGE

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

The ALTER PACKAGE statement explicitly recompiles a package specification, package body, or both. Explicit recompilation eliminates the need for implicit runtime recompilation and prevents associated runtime compilation errors.

This statement recompiles all package objects together. You cannot use the ALTER PROCEDURE or ALTER FUNCTION statement to individually recompile a procedure or function that is part of a package.

Required privilege

No privilege is required for the package owner.

ALTER ANY PROCEDURE for another user's package.

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

ALTER PACKAGE [Owner.]PackageName COMPILE
      [PACKAGE|SPECIFICATION|BODY]
      [CompilerParametersClause [...]] 
      [REUSE SETTINGS]

Parameters

Parameter Description

[Owner.]PackageName

Name of the package to be recompiled.

COMPILE

Required clause used to force the recompilation of the package specification, package body, or both.

[PACKAGE|SPECIFICATION|BODY]

Specify PACKAGE to recompile both the package specification and the body. Specify SPECIFICATION to recompile the package specification. Specify BODY to recompile the package body.

PACKAGE is the default.

CompilerParametersClause

Use this optional clause to specify a value for one of the PL/SQL persistent compiler parameters. The PL/SQL persistent compiler parameters are PLSQL_OPTIMIZE_LEVEL and NLS_LENGTH_SEMANTICS.

You can specify each parameter once in the statement.

If you omit a parameter from this clause and you specify REUSE SETTINGS, then if a value was specified for the parameter in an earlier compilation, TimesTen uses that earlier value. If you omit a parameter and either you do not specify REUSE SETTINGS or no value has been specified for the parameter in an earlier compilation, then TimesTen obtains the value for the parameter from the session environment.

REUSE SETTINGS

Use this optional clause to prevent TimesTen from dropping and reacquiring compiler switch settings. When you specify REUSE SETTINGS, TimesTen preserves the existing settings and uses them for the compilation of any parameters for which values are not specified.

Description

  • When you recompile a package specification, TimesTen invalidates local objects that depend on the specification, such as procedures that call procedures or functions in the package. The body of the package also depends on the specification. If you subsequently reference one of these dependent objects without first explicitly recompiling it, then TimesTen recompiles it implicitly at runtime.

  • When you recompile a package body, TimesTen does not invalidate objects that depend on the package specification. TimesTen first recompiles objects upon which the body depends, if any of those objects are invalid. If TimesTen recompiles the body successfully, then the body become valid.

  • When you recompile a package, both the specification and the body are explicitly recompiled. If there are no compilation errors, then the specification and body become valid. If there are compilation errors, then TimesTen returns an error and the package remains invalid.

See also

ALTER PROCEDURE

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

The ALTER PROCEDURE statement recompiles a standalone stored procedure. Explicit recompilation eliminates the need for implicit runtime recompilation and prevents associated runtime compilation errors and performance overhead.

To recompile a procedure that is part of a package, recompile the package using the ALTER PACKAGE statement.

Required privilege

No privilege is required for the procedure owner.

ALTER ANY PROCEDURE for another user's procedure.

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

ALTER PROCEDURE [Owner.]ProcedureName COMPILE
      [CompilerParametersClause [...]] 
      [REUSE SETTINGS]

Parameters

Parameter Description

[Owner.]ProcedureName

Name of the procedure to be recompiled.

COMPILE

Required keyword that causes recompilation of the procedure. If the procedure does not compile successfully, use the ttIsql command SHOW ERRORS to display the compiler error messages.

CompilerParametersClause

Use this optional clause to specify a value for one of the PL/SQL persistent compiler parameters. The PL/SQL persistent compiler parameters are PLSQL_OPTIMIZE_LEVEL and NLS_LENGTH_SEMANTICS.

You can specify each parameter once in the statement.

If you omit a parameter from this clause and you specify REUSE SETTINGS, then if a value was specified for the parameter in an earlier compilation, TimesTen uses that earlier value. If you omit a parameter and either you do not specify REUSE SETTINGS or no value has been specified for the parameter in an earlier compilation, then TimesTen obtains the value for the parameter from the session environment.

REUSE SETTINGS

Use this optional clause to prevent TimesTen from dropping and reacquiring compiler switch settings. When you specify REUSE SETTINGS, TimesTen preserves the existing settings and uses them for the compilation of any parameters for which values are not specified.

Description

  • The ALTER PROCEDURE statement does not change the declaration or definition of an existing procedure. To redeclare or redefine a procedure, use the CREATE PROCEDURE statement.

  • TimesTen first recompiles objects upon which the procedure depends, if any of those objects are invalid.

  • TimesTen also invalidates any objects that depend on the procedure, such as procedures that call the recompiled procedure or package bodies that define procedures that call the recompiled procedure.

  • If TimesTen recompiles the procedure successfully, then the procedure becomes valid. If recompiling the procedure results in compilation errors, then TimesTen returns an error and the procedure remains invalid. Use the ttIsql command SHOW ERRORS to display compilation errors.

  • During recompilation, TimesTen drops all persistent compiler settings, retrieves them again from the session, and stores them at the end of compilation. To avoid this process, specify the REUSE SETTINGS clause.

Examples

Query the system view USER_PLSQL_OBJECT_SETTINGS to check PLSQL_OPTIMIZE_LEVEL for procedure query_emp. Alter query_emp by changing PLSQL_OPTIMIZE_LEVEL to 3. Verify results.

Command> SELECT PLSQL_OPTIMIZE_LEVEL FROM user_plsql_object_settings WHERE  name = 'QUERY_EMP';
< 2 >
1 row found.

Command> ALTER PROCEDURE query_emp COMPILE PLSQL_OPTIMIZE_LEVEL = 3;
 
Procedure altered.
 
Command> SELECT PLSQL_OPTIMIZE_LEVEL FROM user_plsql_object_settings WHERE  name = 'QUERY_EMP';
< 3 >
1 row found.

See also

ALTER PROFILE

The ALTER PROFILE statement adds, modifies, or removes one or more password parameters in a profile.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

ALTER PROFILE profile LIMIT password_parameters

password_parameters::=
[FAILED_LOGIN_ATTEMPTS password_parameter_options]
[PASSWORD_LIFE_TIME password_parameter_options]
[PASSWORD_REUSE_TIME password_parameter_options]
[PASSWORD_REUSE_MAX password_parameter_options]
[PASSWORD_LOCK_TIME password_parameter_options]
[PASSWORD_GRACE_TIME password_parameter_options]
[{PASSWORD_COMPLEXITY_CHECKER|PASSWORD_VERIFY_FUNCTION} password_checker_options]

password_parameter_options::=
UNLIMITED|DEFAULT|constant

password_checker_options::=
function|NULL|DEFAULT

function::=
TT_VERIFY_FUNCTION|TT_STRONG_VERIFY_FUNCTION|TT_STIG_VERIFY_FUNCTION

Parameters

Parameter Description

profile

Name of the profile.

LIMIT password_parameters

The LIMIT clause sets the limits for the password parameters. The LIMIT keyword is required.

The password parameters consist of the name of the password parameter and the value (or limit) for the password parameter. This includes the password complexity checker functions. All the parameters (with the exception of FAILED_LOGIN_ATTEMPTS and PASSWORD_REUSE_MAX) set lengths of time and are interpreted in number of days. You can use a decimal value (for example, you can use .0833 to denote approximately one hour). The minimum value is 1 second. The maximum value is 106,751,991 days. The constant value must be expressed in days. For example, to set a value of 5 minutes, specify the constant value of 0.0034722222222222 (5/1440 days). For FAILED_LOGIN_ATTEMPTS and PASSWORD_REUSE_MAX, you must specify an integer.

If you do not specify a password parameter after the LIMIT clause, the limit for that password parameter is based on the limit defined in the DEFAULT profile. In addition, if you only specify the LIMIT keyword with no additional parameters, the limits for the profile are based on the limits of the DEFAULT profile.

FAILED_LOGIN_ATTEMPTS

Specifies the number of consecutive failed attempts to connect to the database by a user before that user's account is locked.

PASSWORD_LIFE_TIME

Specifies the number of days that a user can use the same password for authentication. If you also set a value for PASSWORD_GRACE_TIME, then the password expires if it is not changed within the grace period. In such a situation, future connections to the database are rejected.

PASSWORD_REUSE_TIME and PASSWORD_REUSE_MAX

These two parameters must be used together.

  • PASSWORD_REUSE_TIME specifies the number of days that must pass before a user can reuse a password. For example, if you specify a value of 30, then after 30 days the user can reuse a previous password.

  • PASSWORD_REUSE_MAX specifies the number of password changes that are required before the current password can be reused.

You must specify a value for both parameters for them to have any effect. Specifically:

  • If you specify a value for both parameters: A user cannot reuse a password until the password has been changed the number of times specified for PASSWORD_REUSE_MAX during the number of days specified for PASSWORD_REUSE_TIME. For example, if you specify a value of 30 for PASSWORD_REUSE_TIME and a value of 10 for PASSWORD_REUSE_MAX, then the user can reuse the password after 30 days if the password has been changed 10 times.

  • If you specify a value for one parameter and specify a value of UNLIMITED for the second parameter, then the user can never reuse a password.

  • If you specify a value of UNLIMITED for both parameters, then TimesTen ignores both values, indicating that the password can be reused.

PASSWORD_LOCK_TIME

Specifies the number of days the user account is locked after the specified number of consecutive failed connection attempts.

PASSWORD_GRACE_TIME

Specifies the number of days after the grace period begins during which TimesTen issues a warning, but allows the connection to the database. If the password is not changed during the grace period, the password expires. This parameter is associated with the PASSWORD_LIFE_TIME parameter.

UNLIMITED

Indicates that there is no limit for the password parameter. If you specify UNLIMITED, it must follow the password parameter. For example, FAILED_LOGIN_ATTEMPTS UNLIMITED.

DEFAULT

Indicates that you want to omit a limit for the password parameter in this profile. A user that is assigned this profile is subject to the limit defined in the DEFAULT profile for this password parameter.

If you specify DEFAULT, it must follow the password parameter. For example, FAILED_LOGIN_ATTEMPTS DEFAULT.

constant

Indicates the value of the password parameter if you do not specify UNLIMITED or DEFAULT. If specified, it must follow the password parameter. For example, FAILED_LOGIN_ATTEMPTS 3.

{PASSWORD_COMPLEXITY_CHECKER| PASSWORD_VERIFY_FUNCTION} {function|NULL|DEFAULT}

Indicates if password verification is done on passwords and, if so, the function used for verification. You can specify either the PASSWORD_COMPLEXITY_CHECKER or the PASSWORD_VERIFY_FUNCTION password parameter. They are synonymous.

function refers to one of the three supported password complexity checker functions. Specify one of these functions to direct TimesTen to perform password verification. Valid values:
  • TT_VERIFY_FUNCTION
  • TT_STRONG_VERIFY_FUNCTION
  • TT_STIG_VERIFY_FUNCTION

NULL indicates that there is not a password verification function assigned for the profile.

DEFAULT indicates that the user is subject to the limits defined by the DEFAULT profile. The DEFAULT profile initially has a value of NULL.

If you do not specify the PASSWORD_COMPLEXITY_CHECKER password parameter, the value defaults to the limits defined for the DEFAULT profile.

Description

  • Use the ALTER PROFILE statement to modify a previously created profile. See "CREATE PROFILE" for information on creating a profile.

  • Changes made using the ALTER PROFILE statement takes effect the next time any affected user connected to the database. The exception is when you modify the PASSWORD_COMPLEXITY_CHECKER password parameter. Password verification is only done on newly created passwords (on the password provided in the IDENTIFIED BY clause of the CREATE USER or ALTER USER statement). Therefore, a user can connect to the database with an old password. See "ALTER the PASSWORD_COMPLEXITY_CHECKER password parameter" for an example.

  • You can alter the DEFAULT profile. However, you cannot drop the DEFAULT profile. See "Alter the DEFAULT profile" for an example of altering the DEFAULT profile.

  • You cannot alter the password parameters of the SYSTEM profile. This profile is assigned to system users, including the instance administrator.

  • You can alter the profile to change the password verification that is done on the passwords of users that are assigned the profile. See "About password complexity checker verification" for information on password verification and the password complexity checker verification functions.

Examples

ALTER the PASSWORD_COMPLEXITY_CHECKER password parameter

This example creates the myprofile_alterpw1 profile and specifies TT_VERIFY_FUNCTION for the PASSWORD_COMPLEXITY_CHECKER password parameter. The example then creates the sampleuser_alterpw1 user and assigns the myprofile_alterpw1 profile to the sampleuser_alterpw1 user. The example alters the profile, specifying TT_STIG_VERIFY_FUNCTION for the PASSWORD_COMPLEXITY_CHECKER password parameter. The sampleuser_alterpw1 attempts to connect to the database with the original password. The connection is successful. TimesTen does not perform password verification on old passwords. The example then uses the ALTER USER statement to change the sampleuser_alterpw1 user password to meet the requirements of the TT_STIG_VERIFY_FUNCTION. The ALTER USER statement succeeds and the user's password is changed.

Command> CREATE PROFILE myprofile_alterpw1 LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_VERIFY_FUNCTION;

Profile created.

Command> CREATE USER sampleuser_alterpw1 
           IDENTIFIED BY "%aabb2L90" PROFILE myprofile_alterpw1;

User created.

Alter the myprofile_alterpw1 profile, changing the value of PASSWORD_COMPLEXITY_CHECKER to TT_STIG_VERIFY_FUNCTION. Connect to the database as the sampleuser_alterpw1 user. The connection succeeds.

Command> ALTER PROFILE myprofile_alterpw1 LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_STIG_VERIFY_FUNCTION;

Profile altered.

Command> GRANT CONNECT TO sampleuser_alterpw1;
Command> connect adding "UID=sampleuser_alterpw1;PWD=%aabb2L90" as sampleuser;
Connection successful: DSN=access1;UID=sampleuser_alterpw1;
DataStore=/scratch/sampleuser/mydatabase1;DatabaseCharacterSet=AL32UTF8;
ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Alter the sampleuser_alterpw1 user specifying the same password. The ALTER USER statement fails. The newly created password does not meet the requirements of the TT_STIG_VERIFY_FUNCTION function. Alter the sampleuser_alterpw1 again, specifying a password that meets the requirements of the TT_STIG_VERIFY_FUNCTION function. The ALTER USER statement succeeds. See "TT_STIG_VERIFY_FUNCTION" for information on the TT_STIG_VERIFY_FUNCTION function.

Command> ALTER USER sampleuser_alterpw1 
           IDENTIFIED BY  "%aabb2L90";
15186: Password complexity check for the specified password failed
15188: TT-20001: Password length less than 15
The command failed.

Command> ALTER USER sampleuser_alterpw1 
           IDENTIFIED BY  "%aabb2L##mf5Fn!";

User altered.

Alter the DEFAULT profile

This example verifies the values of the password parameters in the DEFAULT profile. It then alters the profile with different values. Users that are assigned the DEFAULT profile will inherit the modified values at the user's next connection to the database.

Command> SELECT * FROM dba_profiles WHERE profile='DEFAULT' AND 
           resource_type='PASSWORD';
< DEFAULT, FAILED_LOGIN_ATTEMPTS, PASSWORD, 10 >
< DEFAULT, PASSWORD_LIFE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_MAX, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, NULL >
< DEFAULT, PASSWORD_LOCK_TIME, PASSWORD, .0034 >
< DEFAULT, PASSWORD_GRACE_TIME, PASSWORD, UNLIMITED >
7 rows found.

Create the user1 user and do not specify a profile. User1 is assigned the DEFAULT profile. Use the ALTER PROFILE statement to change the value of the FAILED_LOGIN_ATTEMPTS password parameter to 5 and the value of the PASSWORD_LOCK_TIME password parameter to 1 for the DEFAULT profile. Enclose DEFAULT in double quotation marks as DEFAULT is a reserved word. Connect to the database five times as user1 supplying an incorrect password each time. On the sixth attempt, the user1 account is locked.

Command> CREATE USER user1 IDENTIFIED BY user1;
 
User created.
Command> GRANT CONNECT TO user1;

Query the dba_users system view to verify that user1 is assigned the DEFAULT profile.

Command> SELECT profile FROM dba_users WHERE username='USER1';
< DEFAULT >
1 row found.

Use the ALTER PROFILE statement to modify the DEFAULT profile.

Command> ALTER PROFILE "DEFAULT" LIMIT 
           FAILED_LOGIN_ATTEMPTS 5 
           PASSWORD_LOCK_TIME 1;
 
Profile altered.
 

Query the dba_profiles system view to verify the values are changed (represented in bold).

Command> SELECT * FROM dba_profiles WHERE profile='DEFAULT' AND 
   resource_type='PASSWORD';
< DEFAULT, FAILED_LOGIN_ATTEMPTS, PASSWORD, 5 >
< DEFAULT, PASSWORD_LIFE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_MAX, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, NULL >
< DEFAULT, PASSWORD_LOCK_TIME, PASSWORD, 1 >
< DEFAULT, PASSWORD_GRACE_TIME, PASSWORD, UNLIMITED >
7 rows found.

Attempt to connect to the database as user1. Supply an incorrect password. On the sixth attempt, the user1 account is locked.

Command> connect adding "uid=user1;pwd=user1_test1" as user1;
 7001: User authentication failed
The command failed.
none: Command> connect adding "uid=user1;pwd=user1_test2" as user1;
 7001: User authentication failed
The command failed.
none: Command> connect adding "uid=user1;pwd=user1_test3" as user1;
 7001: User authentication failed
The command failed.
none: Command> connect adding "uid=user1;pwd=user1_test4" as user1;
 7001: User authentication failed
The command failed.
none: Command> connect adding "uid=user1;pwd=user1_test5" as user1;
 7001: User authentication failed
The command failed.
none: Command> connect adding "uid=user1;pwd=user1_test6" as user1;
15179: the account is locked
The command failed.

Create a profile then alter the profile

This example creates the profile1 profile and specifies values for the FAILED_LOGIN_ATTEMPTS, the PASSWORD_LIFE_TIME, the PASSWORD_LOCK_TIME, and the PASSWORD_GRACE_TIME password parameters. It then alters the profile1 profile to modify the PASSWORD_REUSE_TIME and the PASSWORD_REUSE_MAX password parameters.

Command> CREATE PROFILE profile1 LIMIT 
           FAILED_LOGIN_ATTEMPTS 3 
           PASSWORD_LIFE_TIME 90 
           PASSWORD_LOCK_TIME 30 
           PASSWORD_GRACE_TIME 10;
 
Profile created.

Query the dba_profiles system view to verify the values for the password parameters. Note that the PASSWORD_REUSE_TIME and the PASSWORD_REUSE_MAX password parameters each have a value of DEFAULT (represented in bold). These password parameters were not specified in the CREATE PROFILE definition, so TimesTen assigns a value of DEFAULT to each parameter. The values for these parameters are derived from the values in the DEFAULT profile.

Command> SELECT * FROM dba_profiles WHERE profile = 'PROFILE1' AND
           resource_type= 'PASSWORD';
< PROFILE1, FAILED_LOGIN_ATTEMPTS, PASSWORD, 3 >
< PROFILE1, PASSWORD_LIFE_TIME, PASSWORD, 90 >
< PROFILE1, PASSWORD_REUSE_TIME, PASSWORD, DEFAULT >
< PROFILE1, PASSWORD_REUSE_MAX, PASSWORD, DEFAULT >
< PROFILE1, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, DEFAULT >
< PROFILE1, PASSWORD_LOCK_TIME, PASSWORD, 30 >
< PROFILE1, PASSWORD_GRACE_TIME, PASSWORD, 10 >
7 rows found.

Alter the profile1 profile, specifying a value of 20 for the PASSWORD_REUSE_TIME password and a value of 15 for the PASSWORD_REUSE_MAX password parameter (represented in bold). A user assigned this profile can reuse the same password after 20 days if the password has been changed 15 times.

Command> ALTER PROFILE profile1 LIMIT 
           PASSWORD_REUSE_TIME 20 
           PASSWORD_REUSE_MAX 15;
 
Profile altered.

Query the dba_profiles system view to verify the values for the password parameters are changed (represented in bold).

Command> SELECT * FROM dba_profiles WHERE profile = 'PROFILE1' AND 
           resource_type= 'PASSWORD';
< PROFILE1, FAILED_LOGIN_ATTEMPTS, PASSWORD, 3 >
< PROFILE1, PASSWORD_LIFE_TIME, PASSWORD, 90 >
< PROFILE1, PASSWORD_REUSE_TIME, PASSWORD, 20 >
< PROFILE1, PASSWORD_REUSE_MAX, PASSWORD, 15 >
< PROFILE1, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, DEFAULT >
< PROFILE1, PASSWORD_LOCK_TIME, PASSWORD, 30 >
< PROFILE1, PASSWORD_GRACE_TIME, PASSWORD, 10 >
7 rows found.

ALTER REPLICATION

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

The ALTER REPLICATION statement adds, alters, or drops replication elements and changes the replication attributes of participating databases involved in a classic replication scheme.

Most ALTER REPLICATION operations are supported only when the replication agent is stopped (ttAdmin -repStop). However, it is possible to dynamically add a subscriber database to a replication scheme while the replication agent is running. See "Altering a Classic Replication Scheme" in Oracle TimesTen In-Memory Database Replication Guide for more information.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

The ALTER REPLICATION statement has the syntax:

ALTER REPLICATION [Owner.]ReplicationSchemeName
  ElementOperation [...] | StoreOperation |
  NetworkOperation [...]

Specify ElementOperation one or more times:

ADD ELEMENT ElementName
  { DATASTORE | 
  { TABLE [Owner.]TableName [CheckConflicts] } | 
    SEQUENCE [Owner.]SequenceName }
  { MASTER | PROPAGATOR } FullStoreName
  { SUBSCRIBER FullStoreName [,... ] [ReturnServiceAttribute] } [ ... ] 
  { INCLUDE | EXCLUDE } { TABLE [[Owner.]TableName[,...]] | 
    SEQUENCE [[Owner.]SequenceName[,...]] } [,...]

ALTER ELEMENT { ElementName | * IN FullStoreName ]}
  ADD SUBSCRIBER FullStoreName [,...] [ReturnServiceAttribute] |
  ALTER SUBSCRIBER FullStoreName [,...]|
  SET [ReturnServiceAttribute] 
  DROP SUBSCRIBER FullStoreName [,... ]

ALTER ELEMENT * IN FullStoreName
  SET { MASTER | PROPAGATOR } FullStoreName

ALTER ELEMENT ElementName
  {SET NAME NewElementName | SET CheckConflicts}

ALTER ELEMENT ElementName 
  { INCLUDE | EXCLUDE } { TABLE [Owner.]TableName |
    SEQUENCE [Owner.]SequenceName }[,...] 

DROP ELEMENT { ElementName | * IN FullStoreName }

CheckConflicts can only be set when replicating TABLE elements. See "CHECK CONFLICTS" for syntax requirements.

Syntax for ReturnServiceAttribute is:

{ RETURN RECEIPT [BY REQUEST] | NO RETURN }

StoreOperation clauses:

ADD STORE FullStoreName [StoreAttribute [... ]]
ALTER STORE FullStoreName SET StoreAttribute [... ]

Syntax for the StoreAttribute is:

DISABLE RETURN {SUBSCRIBER | ALL} NumFailures
RETURN SERVICES {ON | OFF} WHEN [REPLICATION] STOPPED
DURABLE COMMIT {ON | OFF}
RESUME RETURN Milliseconds 
LOCAL COMMIT ACTION {NO ACTION | COMMIT}
RETURN WAIT TIME Seconds
COMPRESS TRAFFIC {ON | OFF} 
PORT PortNumber 
TIMEOUT Seconds 
FAILTHRESHOLD Value
CONFLICT REPORTING SUSPEND AT Value 
CONFLICT REPORTING RESUME AT Value 
TABLE DEFINITION CHECKING {EXACT|RELAXED}

Specify NetworkOperation one or more times:

ADD ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName
  { { MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost }
      PRIORITY Priority } [...] 

DROP ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName
  { MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost } [...] 

Parameters

Parameter Description

[Owner.]ReplicationSchemeName

Name assigned to the classic replication scheme.

ADD ELEMENT ElementName

Adds a new element to the existing classic replication scheme. ElementName is an identifier of up to 30 characters. With DATASTORE elements, the ElementName must be unique with respect to other DATASTORE element names within the first 20 characters.

If the element is a DATASTORE, all tables are included in the database. SEQUENCE elements that are part of the database do not have their return services modified by this statement.

ADD ELEMENT ElementName DATASTORE

{INCLUDE | EXCLUDE}

{TABLE [[Owner.]TableName [,...]]|

SEQUENCE [[Owner.]SequenceName[,...]]} [,...]

Adds a new DATASTORE element to the existing classic replication scheme. ElementName is an identifier of up to 30 characters. With DATASTORE elements, the ElementName must be unique with respect to other DATASTORE element names within the first 20 characters.

INCLUDE includes in the database only the tables and sequences listed. Use one INCLUDE clause for each object type (table or sequence).

EXCLUDE includes in the database all tables and sequences except the tables and sequences listed. Use one EXCLUDE clause for each object type (table or sequence).

If the element is a sequence, RETURN attributes are not applied, no conflict checking is supported and sequences that cycle return an error.

ADD SUBSCRIBER FullStoreName

Indicates an additional subscriber database. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

ALTER ELEMENT * IN FullStoreName

SET { MASTER | PROPAGATOR } FullStoreName

Makes a change to all elements for which FullStoreName is the MASTER or PROPAGATOR. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

This syntax can be used on a set of element names to:

  • Add, alter, or drop subscribers.

  • Set the MASTER or PROPAGATOR status of the element set.

SEQUENCE elements that are part of the database being altered do not have their return services modified by this statement.

ALTER ELEMENT ElementName

Name of the element to which a subscriber is to be added or dropped.

ALTER ELEMENT

ElementName1

SET NAME ElementName2

Renames ElementName1 with the name ElementName2. You can only rename elements of type TABLE.

ALTER ELEMENT ElementName

{INCLUDE|EXCLUDE}

{TABLE [Owner.]TableName |

SEQUENCE [Owner.]SequenceName} [,...]

ElementName is the name of the element to be altered.

INCLUDE adds to the database the tables and sequences listed. Use one INCLUDE clause for each object type (table or sequence).

EXCLUDE removes from the database the tables and sequences listed. Use one EXCLUDE clause for each object type (table or sequence).

If the element is a sequence, RETURN attributes are not applied, no conflict checking is supported and sequences that cycle return an error.

ALTER SUBSCRIBER FullStoreName

SET RETURN RECEIPT

[BY REQUEST]|NO RETURN

Indicates an alteration to a subscriber database to enable, disable, or change the return receipt service. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

CheckConflicts

Check for replication conflicts when simultaneously writing to bidirectionally replicating TABLE elements between databases. You cannot check for conflicts when replicating elements of type DATASTORE. See "CHECK CONFLICTS".

COMPRESS TRAFFIC {ON | OFF}

Compress replicated traffic to reduce the amount of network bandwidth. ON specifies that all replicated traffic for the database defined by STORE be compressed. OFF (the default) specifies no compression. See "Compressing replicated traffic" in Oracle TimesTen In-Memory Database Replication Guide for details.

CONFLICT REPORTING SUSPEND AT Value

Suspends conflict resolution reporting.

Value is a non-negative integer. Conflict reporting is suspended when the rate of conflict exceeds Value. The default is 0, which means that the conflict reporting is always on and will not be suspended.

This clause is valid for table level replication.

CONFLICT REPORTING RESUME AT Value

Resumes conflict resolution reporting.

Value is a non-negative integer. Conflict reporting is resumed when the rate of conflict falls below Value. The default is 1.

This clause is valid for table level replication.

DISABLE RETURN {SUBSCRIBER | ALL} NumFailures

Set the return service failure policy so that return service blocking is disabled after the number of timeouts specified by NumFailures. Selecting SUBSCRIBER applies this policy only to the subscriber that fails to acknowledge replicated updates within the set timeout period. ALL applies this policy to all subscribers should any of the subscribers fail to respond. This failure policy can be specified for either the RETURN RECEIPT or RETURN TWOSAFE service.

If DISABLE RETURN is specified but RESUME RETURN is not specified, the return services remain off until the replication agent for the database has been restarted.

DURABLE COMMIT {ON | OFF}

Overrides the DurableCommits general connection attribute setting. DURABLE COMMIT ON enables durable commits regardless of whether the replication agent is running or stopped.

DROP ELEMENT * IN FullStoreName

Deletes the replication description of all elements for which FullStoreName is the MASTER. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

DROP ELEMENT ElementName

Deletes the replication description of ElementName.

DROP SUBSCRIBER FullStoreName

Indicates that updates should no longer be sent to the specified subscriber database. This operation fails if the classic replication scheme has only one subscriber. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

FAILTHRESHOLD Value

The number of log files that can accumulate for a subscriber database. If this value is exceeded, the subscriber is set to the Failed state.

The value 0 means "No Limit." This is the default.

See "Setting the transaction log failure threshold" in Oracle TimesTen In-Memory Database Replication Guide for more information.

FullStoreName

The database, specified as one of the following:

  • SELF

  • The prefix of the database file name

For example, if the database path is directory/subdirectory/data.ds0, then data is the database name.

This is the database file name specified in the DataStore attribute of the DSN description with optional host ID in the form:

DataStoreName [ON Host]

Host can be either an IP address or a literal host name assigned to one or more IP addresses, as described in "Configuring the network" in Oracle TimesTen In-Memory Database Replication Guide. Host names containing special characters must be surrounded by double quotes. For example: "MyHost-500".

LOCAL COMMIT ACTION {NO ACTION | COMMIT}

Specifies the default action to be taken for a RETURN TWOSAFE transaction in the event of a timeout.

NO ACTION: On timeout, the commit function returns to the application, leaving the transaction in the same state it was in when it entered the commit call, with the exception that the application is not able to update any replicated tables. The application can only reissue the commit. The transaction may not be rolled back. This is the default.

COMMIT: On timeout, the commit function attempts to perform a COMMIT to end the transaction locally. No more operations are possible on the same transaction.

This setting can be overridden for specific transactions by calling the ttRepSyncSet procedure with the localAction parameter.

MASTER FullStoreName

The database on which applications update the specified element. The MASTER database sends updates to its SUBSCRIBER databases. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

NO RETURN

Specifies that no return service is to be used. This is the default.

For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

PORT PortNumber

The TCP/IP port number on which the replication agent on this database listens for connections. If not specified, the replication agent allocates a port number automatically.

All TimesTen databases that replicate to each other must use the same port number.

PROPAGATOR FullStoreName

The database that receives replicated updates and passes them on to other databases.

RESUME RETURN Milliseconds

If return service blocking has been disabled by DISABLE RETURN, this attribute sets the policy on when to re-enable return service blocking. Return service blocking is re-enabled as soon as the failed subscriber acknowledges the replicated update in a period of time that is less than the specified Milliseconds.

If DISABLE RETURN is specified but RESUME RETURN is not specified, the return services remain off until the replication agent for the database has been restarted.

RETURN RECEIPT [BY REQUEST]

Enables the return receipt service, so that applications that commit a transaction to a master database are blocked until the transaction is received by all subscribers.

RETURN RECEIPT applies the service to all transactions. If you specify RETURN RECEIPT BY REQUEST, you can use the ttRepSyncSet procedure to enable the return receipt service for selected transactions. For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

RETURN SERVICES {ON | OFF} WHEN [REPLICATION] STOPPED

Sets return services on or off when replication is disabled (stopped or paused state).

OFF disables return services when replication is disabled and is the default for RETURN RECEIPT service. ON allows return services to continue to be enabled when replication is disabled and is the default for RETURN TWOSAFE service.

RETURN TWOSAFE [BY REQUEST]

Enables the return twosafe service, so that applications that commit a transaction to a master database are blocked until the transaction is committed on all subscribers.

RETURN TWOSAFE applies the service to all transactions. If you specify RETURN TWOSAFE BY REQUEST, you can use the ttRepSyncSet procedure to enable the return receipt service for selected transactions. For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

RETURN WAIT TIME Seconds

Specifies the number of seconds to wait for return service acknowledgment. The default value is 10 seconds. A value of 0 (zero) means there is no timeout. Your application can override this timeout setting by calling the ttRepSyncSet procedure with the returnWait parameter.

SET {MASTER | PROPAGATOR} FullStoreName

Sets the given database to be the MASTER or PROPAGATOR of the given elements. The FullStoreName must the be database's file base name.

SUBSCRIBER FullStoreName

A database that receives updates from the MASTER databases. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

TABLE DEFINITION CHECKING {EXACT|RELAXED}

Specifies type of table definition checking that occurs on the subscriber:

  • EXACT - The tables must be identical on master and subscriber.

  • RELAXED - The tables must have the same key definition, number of columns and column data types.

The default is RELAXED.

Note: If you use TABLE DEFINITION CHECKING EXACT, use ttMigrate -exactUpgrade if you migrate the database. If you use TABLE DEFINITION CHECKING RELAXED, use ttMigrate -relaxedUpgrade if you migrate the database.

TIMEOUT Seconds

The maximum number of seconds the replication agent waits for a response from remote replication agents. The default is 120 seconds.

Note: For large transactions that may cause a delayed response from the remote replication agent, the agent scales the timeout to increasingly larger values, as needed, based on the size of the transaction. This scaling will not occur, and the agent may time out waiting for responses, if you set TIMEOUT to less than or equal to 60 seconds for large transactions. Also see "Setting wait timeout for response from remote replication agents" in Oracle TimesTen In-Memory Database Replication Guide.

ADD ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName

Adds NetworkOperation to replication scheme. Enables you to control the network interface that a master store uses for every outbound connection to each of its subscriber stores.

Can be specified more than once.

For FullStoreName, ON "host" must be specified.

DROP ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName

Drops NetworkOperation from the classic replication scheme.

Can be specified more than once.

For FullStoreName, ON "host" must be specified.

MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost

MasterHost and SubscriberHost are the IP addresses for the network interface on the master and subscriber stores. Specify in dot notation or canonical format or in colon notation for IPV6.

Clause can be specified more than once. Valid for both ADD and DROP ROUTE MASTER.

PRIORITY Priority

Variable expressed as an integer from 1 to 99. Denotes the priority of the IP address. Lower integral values have higher priority. An error is returned if multiple addresses with the same priority are specified. Controls the order in which multiple IP addresses are used to establish peer connections.

Required syntax of NetworkOperation clause. Follows MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost clause.

Description

  • ALTER ELEMENT DROP SUBSCRIBER deletes a subscriber for a particular replication element.

  • ALTER ELEMENT SET NAME may be used to change the name of a replication element when it conflicts with one already defined at another database. SET NAME does not admit the use of * IN FullStoreName. The FullStoreName must be the database's file base name. For example, if the database file name is data.ds0, then data is the file base name.

  • ALTER ELEMENT SET MASTER may be used to change the master database for replication elements. The * IN FullStoreName option must be used for the MASTER operation. That is, a master database must transfer ownership of all of its replication elements, thereby giving up its master role entirely. Typically, this option is used in ALTER REPLICATION statements requested at SUBSCRIBER databases after the failure of a (common) MASTER.

  • To transfer ownership of the master elements to the subscriber:

    1. Manually drop the replicated elements by executing an ALTER REPLICATION DROP ELEMENT statement for each replicated table.

    2. Use ALTER REPLICATION ADD ELEMENT to add each table back to the replication scheme, with the newly designated MASTER / SUBSCRIBER roles.

  • ALTER REPLICATION ALTER ELEMENT SET MASTER does not automatically retain the old master as a subscriber in the scheme. If this is desired, execute an ALTER REPLICATION ALTER ELEMENT ADD SUBSCRIBER statement.

    Note:

    There is no ALTER ELEMENT DROP MASTER. Each replication element must have exactly one MASTER database, and the currently designated MASTER cannot be deleted from the replication scheme.

  • Stop the replication agent before you use the NetworkOperation clause.

  • You cannot alter the following replication schemes with the ALTER REPLICATION statement:

    • Any active standby pair. Instead, use ALTER ACTIVE STANDBY PAIR.

    • A Clusterware-managed active standby pair. Instead, perform the tasks described in "Changing the schema" section of the Oracle TimesTen In-Memory Database Replication Guide.

Examples

This example sets up a classic replication scheme for an additional table westleads that is updated on database west and replicated to database east.

ALTER REPLICATION r1
   ADD ELEMENT e3 TABLE westleads
     MASTER west ON "westcoast"
     SUBSCRIBER east ON "eastcoast";

This example adds an additional subscriber (backup) to table westleads.

ALTER REPLICATION r1
   ALTER ELEMENT e3
     ADD SUBSCRIBER backup ON "backupserver";

This example changes the element name of table westleads from e3 to newelementname.

ALTER REPLICATION r1
   ALTER ELEMENT e3
     SET NAME newelementname;

This example makes newwest the master for all elements for which west currently is the master.

ALTER REPLICATION r1
   ALTER ELEMENT * IN west
     SET MASTER newwest;

This element changes the port number for east.

ALTER REPLICATION r1
   ALTER STORE east ON "eastcoast" SET PORT 22251;

This example adds my.tab1 table to the ds1 database element in my.rep1 replication scheme.

ALTER REPLICATION my.rep1
  ALTER ELEMENT ds1 DATASTORE
     INCLUDE TABLE my.tab1;

This example adds ds1 database to my.rep1 replication scheme. Include my.tab2 table in the database.

ALTER REPLICATION my.rep1
  ADD ELEMENT ds1 DATASTORE
     MASTER rep2
     SUBSCRIBER rep1, rep3
     INCLUDE TABLE my.tab2;

This example adds ds2 database to a replication scheme but excludes my.tab1 table.

ALTER REPLICATION my.rep1
  ADD ELEMENT ds2 DATASTORE
     MASTER rep2
     SUBSCRIBER rep1
     EXCLUDE TABLE my.tab1;

Add NetworkOperation clause:

ALTER REPLICATION r
ADD ROUTE MASTER rep1 ON "machine1" SUBSCRIBER rep2 ON "machine2"
MASTERIP "1.1.1.1" PRIORITY 1 SUBSCRIBERIP "2.2.2.2"
    PRIORITY 1
MASTERIP "3.3.3.3" PRIORITY 2 SUBSCRIBERIP "4.4.4.4" PRIORITY 2;

Drop NetworkOperation clause:

ALTER REPLICATION r
DROP ROUTE MASTER repl ON "machine1" SUBSCRIBER rep2 ON "machine2"
MASTERIP "1.1.1.1" SUBSCRIBERIP "2.2.2.2"
MASTERIP "3.3.3.3" SUBSCRIBERIP "4.4.4.4";

ALTER SEQUENCE

This statement is supported in TimesTen Scaleout only.

Use the ALTER SEQUENCE statement to change the batch value of a sequence.

Required privilege

No privilege is required for the sequence owner.

ALTER ANY SEQUENCE privilege for another user's sequence.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

ALTER SEQUENCE [Owner.]SequenceName BATCH BatchValue

Parameters

Parameter Description

SEQUENCE [Owner.]SequenceName

Name of the sequence to be altered.

BATCH BatchValue

Valid with TimesTen Scaleout only. Configures the range of unique sequence values that are stored at each element of the grid. The default value is 10 million.

Description

  • Use this statement to change the batch value for a sequence in TimesTen Scaleout. The change affects future sequence numbers.

  • This statement cannot be used to alter any other values supported in the CREATE SEQUENCE statement. In this case, use the DROP SEQUENCE statement and then create a new sequence with the same name. For example, to change the MINVALUE, drop the sequence and recreate it with the same name and with the desired MINVALUE.

See "Using sequences" in Oracle TimesTen In-Memory Database Scaleout User's Guide for more information.

Examples

To change the batch value for the sequence:

ALTER SEQUENCE myseq BATCH 2000;
Sequence altered

ALTER SESSION

The ALTER SESSION statement changes session parameters dynamically. This overrides the setting of the equivalent connection attribute for the current session, as applicable.

Required privilege

None

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout. However, these parameters are not supported:

  • DDL_REPLICATION_ACTION

  • DDL_REPLICATION_LEVEL

  • REPLICATION_TRACK

SQL syntax

ALTER SESSION SET
  {COMMIT_BUFFER_SIZE_MAX = n |
   DDL_REPLICATION_ACTION={'INCLUDE'|'EXCLUDE'} | 
   DDL_REPLICATION_LEVEL={1|2|3} |
   ISOLATION_LEVEL = {SERIALIZABLE | READ COMMITTED} |
   NLS_SORT = {BINARY| SortName} |
   NLS_LENGTH_SEMANTICS = {BYTE|CHAR} |
   NLS_NCHAR_CONV_EXCP = {TRUE|FALSE} |
   PLSQL_TIMEOUT = n |
   PLSQL_OPTIMIZE_LEVEL = {0|1|2|3}|
   PLSQL_CONN_MEM_LIMIT = n |
   PLSQL_CCFLAGS = 'name1:value1, name2:value2,..., nameN:valueN' |
   PLSQL_SESSION_CACHED_CURSORS = n |
   REPLICATION_TRACK = TrackNumber |
   } 

Parameters

Parameter Description

COMMIT_BUFFER_SIZE_MAX= n

Changes the maximum size of the commit buffer when a connection is in progress. n is expressed as an integer and represents the maximum size of the commit buffer (in MB).

Change takes effect starting with the next transaction.

Call the ttConfiguration built-in procedure to see the currently configured maximum size of the commit buffer. A value of 0 means the buffer is configured with a default size. The default size is 128 KB.

For more information on the commit buffer and transaction reclaim operations, see "Transaction reclaim operations" in the Oracle TimesTen In-Memory Database Operations Guide and "CommitBufferSizeMax" in the Oracle TimesTen In-Memory Database Reference.

Note: The equivalent connection attribute is CommitBufferSizeMax.

DDL_REPLICATION_ACTION={'INCLUDE'|'EXCLUDE'}

To include a table or sequence in the active standby pair when either is created, set DDL_REPLICATION_ACTION to INCLUDE. If you do not want to include a table or sequence in the active standby pair when either is created, set DDL_REPLICATION_ACTION to EXCLUDE. The default is INCLUDE.

If set to EXCLUDE:

  • A subsequent ALTER ACTIVE STANDBY PAIR ... INCLUDE TABLE is required to be executed on the active database to add the table to the replication scheme. All tables must be empty on all active standby databases and subscribers as the table contents will be truncated when this statement is executed.

  • A subsequent ALTER ACTIVE STANDBY PAIR ... INCLUDE SEQUENCE is required to be executed on the active database to add the sequence to the replication scheme.

This attribute is valid only if DDL_REPLICATION_LEVEL is 2 or greater.

See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Note: The equivalent connection attribute is DDLReplicationAction.

DDL_REPLICATION_LEVEL={1|2|3}

Indicates whether DDL is replicated across all databases in an active standby pair. The value can be one of the following:

  • 1: Default. Add or drop a column to or from a replicated table on the active database using ALTER TABLE. The change is replicated to the table in the standby database.

  • 2: Supports replication of the creation or dropping of tables, synonyms or indexes from the active database to the standby database. This does include creating or dropping global temporary tables, but does not include CREATE TABLE AS SELECT. The statement is replicated only when the index is created on an empty table.

  • 3: Supports replication of all DDL supported by level 2 as well as replication of creation or dropping of views and sequences (not including materialized views) and changes to the cache administration user ID and password settings when you call the ttCacheUidPwdSet built-in procedure.

    Note: After you have defined cache groups, you cannot change the cache administration user ID, but can still change the cache administration password.

See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Note: The equivalent connection attribute is DDLReplicationLevel.

ISOLATION_LEVEL = {SERIALIZABLE|READ COMMITTED}

Sets isolation level. Change takes effect starting with the next transaction.

For a descriptions of the isolation levels, see "Transaction isolation levels" in the Oracle TimesTen In-Memory Database Operations Guide.

Note: The equivalent connection attribute is Isolation.

NLS_SORT={BINARY| SortName}

Indicates which collation sequence to use for linguistic comparisons.

Append _CI or _AI to either BINARY or the SortName value to do case-insensitive or accent-insensitive sorting.

If you do not specify NLS_SORT, the default is BINARY.

For a complete list of supported values for SortName, see "Linguistic sorts" in Oracle TimesTen In-Memory Database Operations Guide.

For more information on case-insensitive or accent-insensitive sorting, see "Case-insensitive and accent-insensitive linguistic sorts" in Oracle TimesTen In-Memory Database Operations Guide.

NLS_LENGTH_SEMANTICS ={BYTE|CHAR}

Sets the default length semantics configuration. BYTE indicates byte length semantics. CHAR indicates character length semantics. The default is BYTE.

For more information on length semantics, see "Length semantics and data storage" in Oracle TimesTen In-Memory Database Operations Guide.

NLS_NCHAR_CONV_EXCP = {TRUE|FALSE}

Determines whether an error should be reported when there is data loss during an implicit or explicit character type conversion between NCHAR/NVARCHAR2 data and CHAR/VARCHAR2 data. Specify TRUE to enable error reporting. Specify FALSE to not report errors. The default is FALSE.

PLSQL_TIMEOUT= n

Controls how long PL/SQL procedures run before being automatically terminated. n represents the time, in seconds. Specify 0 for no time limit or any positive integer. The default is 30.

When you modify this value, the new value impacts PL/SQL program units that are currently running as well as any other program units subsequently executed in the same connection.

See "Choose SQL and PL/SQL timeout values" in the Oracle TimesTen In-Memory Database Operations Guide for information on setting timeout values.

PLSQL_OPTIMIZE_LEVEL = {0|1|2|3}

Specifies the optimization level used to compile PL/SQL library units. The higher the setting, the more effort the compiler makes to optimize PL/SQL library units. Possible values are 0, 1, 2 or 3. The default is 2.

For more information, see "PLSQL_OPTIMIZE_LEVEL" in Oracle TimesTen In-Memory Database Reference.

PLSQL_CONN_MEM_LIMIT = n

Specifies the maximum amount of process heap memory that PL/SQL can use for this connection, where n is an integer expressed in MB. The default is 100.

For more information, see "PLSQL_CONN_MEM_LIMIT" in Oracle TimesTen In-Memory Database Reference.

PLSQL_CCFLAGS = 'name1:value1, name2:value2, ..., nameN:valueN'

Specifies inquiry directives to control conditional compilation of PL/SQL units, which enables you to customize the functionality of a PL/SQL program depending on conditions that are checked. For example, to activate debugging features:

PLSQL_CCFLAGS = 'DEBUG:TRUE'

PLSQL_SESSION_CACHED_CURSORS= n

Specifies the maximum number of session cursors to cache. The default is 50. The range of values is 1 to 65535.

The PLSQL_SESSION_CACHED_CURSORS setting in TimesTen behaves the same as the SESSION_CACHED_CURSORS setting in Oracle RDBMS.

REPLICATION_TRACK = TrackNumber

When managing track-based parallel replication, you can assign a connection to a replication track. All transactions issued by the connection are assigned to this track, unless the track is altered.

If the number specified is for a non-existent replication track X, the transaction is assigned to a track number computed as X modulo ReplicationParallelism.

You cannot change tracks in the middle of a transaction unless all preceding operations have been read operations.

For more information, see "Specifying replication tracks within an automatic parallel replication environment" in Oracle TimesTen In-Memory Database Replication Guide.

The equivalent connection attribute is ReplicationTrack.

Description

  • The ALTER SESSION statement affects commands that are subsequently executed by the session. ALTER SESSION does not do an implicit commit.

  • In cases of client failover, if an ALTER SESSION statement is issued in the failed connection, the setting is not seen or carried over to the new connection. You must re-issue the ALTER SESSION statement and re-specify the value for that parameter. For more information on client failover, in TimesTen Classic, see "Using automatic client failover" in the Oracle TimesTen In-Memory Database Operations Guide and, in TimesTen Scaleout, see "Client connection failover" in the Oracle TimesTen In-Memory Database Scaleout User's Guide.

  • Operations involving character comparisons support linguistic sensitive collating sequences. Case-insensitive sorts may affect DISTINCT value interpretation.

  • Implicit and explicit conversions between CHAR and NCHAR are supported.

  • You can use the SQL string functions with the supported character sets. For example, UPPER and LOWER functions support non-ASCII CHAR and VARCHAR2 characters as well as NCHAR and NVARCHAR2 characters.

  • Choice of character set could have an impact on memory consumption for CHAR and VARCHAR2 column data.

  • The character sets of all databases involved in a replication scheme must match.

  • To add an existing table to an active standby pair, set DDL_REPLICATION_LEVEL to 2 or greater and DDL_REPLICATION_ACTION to INCLUDE. Alternatively, you can use the ALTER ACTIVE STANDBY PAIR ... INCLUDE TABLE statement if DDL_REPLICATION_ACTION is set to EXCLUDE. In this case, the table must be empty and present on all databases before executing the ALTER ACTIVE STANDBY PAIR ... INCLUDE TABLE statement as the table contents will be truncated when this statement is executed.

  • To add an existing sequence or view to an active standby pair, set DDL_REPLICATION_LEVEL to 3. To include the sequence in the replication scheme, DDL_REPLICATION_ACTION must be set to INCLUDE. This does not apply to materialized views.

  • Objects are replicated only when the receiving database is of a TimesTen release that supports that level of replication, and is configured for an active standby pair replication scheme. For example, replication of sequences (requiring DDL_REPLICATION_LEVEL=3) to a database release prior to 11.2.2.7.0 is not supported. The receiving database must be of at least release 11.2.1.8.0 for replication of objects supported by DDL_REPLICATION_LEVEL=2.

Examples

Use the ALTER SESSION statement to change COMMIT_BUFFER_SIZE_MAX to 500 MB. First, call ttConfiguration to display the current connection setting. Use the ALTER SESSION statement to change the COMMIT_BUFFER_SIZE_MAX setting to 500. Call ttConfiguration to display the new setting.

Command>  CALL ttConfiguration ('CommitBufferSizeMax');
< CommitBufferSizeMax, 0 >
1 row found.
Command> ALTER SESSION SET COMMIT_BUFFER_SIZE_MAX = 500;
 
Session altered.
 
Command> CALL ttConfiguration ('CommitBufferSizeMax');
< CommitBufferSizeMax, 500 >
1 row found.

Use the ALTER SESSION statement to change PLSQL_TIMEOUT to 60 seconds. Use a second ALTER SESSION statement to change PLSQL_OPTIMIZE_LEVEL to 3. Then call ttConfiguration to display the new values.

Command> ALTER SESSION SET PLSQL_TIMEOUT = 60;
Session altered.
Command> ALTER SESSION SET PLSQL_OPTIMIZE_LEVEL = 3;
Session altered.

Command> CALL TTCONFIGURATION ();
< CkptFrequency, 600 >
< CkptLogVolume, 0 >
< CkptRate, 0 >
...
< PLSQL_OPTIMIZE_LEVEL, 3 >
< PLSQL_TIMEOUT, 60 >
...
47 rows found.

In this example, set PLSQL_TIMEOUT to 20 seconds. Attempt to execute a program that loops indefinitely. In 20 seconds, execution is terminated and an error is returned.

Command> ALTER SESSION SET PLSQL_TIMEOUT = 20;

Command> DECLARE v_timeout NUMBER;
         BEGIN
           LOOP
             v_timeout :=0;
             EXIT WHEN v_timeout < 0;
           END LOOP;
         END;
         /
 8509: PL/SQL execution terminated; PLSQL_TIMEOUT exceeded

The following example uses the ALTER SESSION statement to change the NLS_SORT setting from BINARY to BINARY_CI to BINARY_AI. The database and connection character sets are WE8ISO8859P1.

Command> connect "dsn=cs;ConnectionCharacterSet=WE8ISO8859P1";
Connection successful: DSN=cs;UID=user;DataStore=/datastore/user/cs;
DatabaseCharacterSet=WE8ISO8859P1;
ConnectionCharacterSet=WE8ISO8859P1;PermSize=32;
(Default setting AutoCommit=1)
Command> -- Create the Table
Command> CREATE TABLE collatingdemo (letter VARCHAR2 (10));
Command> -- Insert values
Command> INSERT INTO collatingdemo VALUES ('a');
1 row inserted.
Command> INSERT INTO collatingdemo VALUES ('A');
1 row inserted.
Command> INSERT INTO collatingdemo VALUES ('Y');
1 row inserted.
Command> INSERT INTO collatingdemo VALUES ('ä');
1 row inserted.
Command> -- SELECT
Command> SELECT * FROM collatingdemo;
< a >
< A >
< Y >
< ä >
4 rows found.
Command> --SELECT with ORDER BY
Command> SELECT * FROM collatingdemo ORDER BY letter;
< A >
< Y >
< a >
< ä >
4 rows found.
Command>-- set NLS_SORT to BINARY_CI and SELECT
Command> ALTER SESSION SET NLS_SORT = BINARY_CI;
Command> SELECT * FROM collatingdemo ORDER BY letter;
< a >
< A >
< Y >
< Ä >
< ä >
4 rows found.
Command> -- Set NLS_SORT to BINARY_AI and SELECT
Command> ALTER SESSION SET NLS_SORT = BINARY_AI;
Command> SELECT * FROM collatingdemo ORDER BY letter;
< ä >
< a >
< A >
< Y >
4 rows found.

The following example enables automatic parallel replication with disabled commit dependencies. It uses the ALTER SESSION statement to change the replication track number to 5 for the current connection. To enable automatic parallel replication with disabled commit dependencies for replication schemes, set ReplicationApplyOrdering to 2. Then, always set REPLICATION_TRACK to a number less than or equal to ReplicationParallelism. For example, the ReplicationParallelism connection attribute could be set to 6, which is higher than the value of 5 set for REPLICATION_TRACK.

Command> ALTER SESSION SET REPLICATION_TRACK = 5;
Session altered.

The following example enables replication of adding and dropping columns, tables, synonyms and indexes by setting the following on the active database in an alter standby replication pair: DDL_REPLICATON_LEVEL set to 2 and DDLReplicationAction set to 'INCLUDE'.

Command > ALTER SESSION SET DDL_REPLICATION_LEVEL=2;
Session altered.

Command > ALTER SESSION SET DDL_REPLICATION_ACTION='INCLUDE';
Session altered.

Note:

The equivalent connection attributes for DDL_REPLICATION_LEVEL and DDL_REPLICATION_ACTION are DDLReplicationLevel and DDLReplicationAction, respectively.

ALTER TABLE

The ALTER TABLE statement changes an existing table definition.

The ALTER TABLE statement is supported in TimesTen Scaleout and in TimesTen Classic. However, there are differences in syntax and semantics. For simplicity, the supported syntax, parameters, description (semantics), and examples for TimesTen Scaleout and for TimesTen Classic are separated into the usage with TimesTen Scaleout and the usage with TimesTen Classic. While there is repetition in the usages, it is presented this way in order to allow you to progress from syntax to parameters to semantics to examples for each usage.

Review the required privilege section and then see:

Required privilege

No privilege is required for the table owner.

ALTER ANY TABLE for another user's table.

For ALTER TABLE...ADD FOREIGN KEY, the owner of the altered table must have the REFERENCES privilege on the table referenced by the foreign key clause.

After reviewing this section, see:

SQL syntax for ALTER TABLE: TimesTen Scaleout

To change the distribution key in TimesTen Scaleout:

ALTER TABLE [Owner.]TableName DistributionClause

To add a primary key constraint and optionally specify a global or local index:

Note: The (CreateIndexStmt) is the clause used to represent the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must create a unique index as that is the requirement for a primary key constraint. See "CREATE INDEX" for details.

ALTER TABLE [Owner.]TableName ADD CONSTRAINT ConstraintName
  PRIMARY KEY (ColumnName [,... ]) [(UsingIndexClause2)]

UsingIndexClause2::= USING INDEX {GLOBAL|LOCAL} [USE HASH INDEX PAGES=RowPages|CURRENT]|
USING INDEX (CreateIndexStmt)

Note:

You cannot use ALTER TABLE to drop a primary key constraint. To drop the constraint, you must drop and recreate the table.

To add a unique constraint on a column and optionally specify a global or local index:

Note: The (CreateIndexStmt) is the clause used to represent the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must create a unique index as that is the requirement for a unique constraint. See CREATE INDEX for details.

ALTER TABLE Owner.]TableName
ADD UNIQUE (ColumnName)
[UsingIndexClause1]

UsingIndexClause1::= USING INDEX {GLOBAL | LOCAL}| USING INDEX (CreateIndexStmt)

To add one column:

ALTER TABLE [Owner.]TableName 
  ADD [COLUMN] ColumnName ColumnDataType
    [DEFAULT DefaultVal] [[NOT] INLINE] [UNIQUE] [NULL]
  [COMPRESS (CompressColumns [,...])]

To add multiple columns:

ALTER TABLE [Owner.]TableName 
 ADD (ColumnName ColumnDataType 
     [DEFAULT DefaultVal] [[NOT] INLINE] [UNIQUE] [NULL] [,... ] )

To add a NOT NULL column (note that the DEFAULT clause is required):

ALTER TABLE [Owner.]TableName
  ADD [COLUMN] ColumnName ColumnDataType
    NOT NULL [ENABLE] DEFAULT DefaultVal [[NOT] INLINE] [UNIQUE]

To add multiple NOT NULL columns (note that the DEFAULT clause is required):

ALTER TABLE [Owner.]TableName
  ADD (ColumnName ColumnDataType
    NOT NULL [ENABLE] DEFAULT DefaultVal [[NOT] INLINE] [UNIQUE] [,...])

To remove columns.

ALTER TABLE [Owner.]TableName 
  DROP {[COLUMN] ColumnName | (ColumnName [,... ] )}

To add a foreign key and optionally add ON DELETE CASCADE:

ALTER TABLE [Owner.]TableName 
ADD [CONSTRAINT ForeignKeyName] FOREIGN KEY
    (ColumnName [,...]) REFERENCES RefTableName
       [(ColumnName [,...])] [ON DELETE CASCADE]

To remove a foreign key:

ALTER TABLE [Owner.]TableName 
DROP CONSTRAINT ForeignKeyName

To resize a hash index:

ALTER TABLE [Owner.]TableName
SET PAGES = RowPages | CURRENT

To change the primary key to use a hash index:

ALTER TABLE [Owner.]TableName
USE HASH INDEX PAGES = RowPages | CURRENT

To change the primary key to use a range index with the USE RANGE INDEX clause:

ALTER TABLE [Owner.]TableName
USE RANGE INDEX

To change the default value of a column:

ALTER TABLE [Owner.]TableName
MODIFY (ColumnName DEFAULT DefaultVal)

To drop a unique constraint on a column:

ALTER TABLE Owner.]TableName
DROP UNIQUE (ColumnName)

To remove the default value of a column that is nullable, by changing it to NULL:

ALTER TABLE [Owner.]TableName
MODIFY (ColumnName DEFAULT NULL)

Parameters for ALTER TABLE ADD CONSTRAINT PRIMARY KEY: TimesTen Scaleout

Parameter Description
ALTER TABLE [Owner.]TableName Start of ALTER TABLE statement. Name of table required. Owner i optional.
ADD CONSTRAINT ConstraintName PRIMARY KEY Clause indicating that the table is to be altered by adding a primary key constraint. ConstraintName is the name of the constraint. Once you add the primary key constraint, you cannot drop it. You must drop the table.
(ColumnName [,…]) (ColumnName) is required and specifies the column(s) to use for the primary key constraint.
[UsingIndexClause2] UsingIndexClause2 is optional and is described in the remainder of this table. You cannot specify two USING INDEX clauses in the ALTER TABLE definition.
USING INDEX {GLOBAL|LOCAL} Part of [UsingIndexClause2]: If specified, indicates if a global or local index is to be created for the primary key.
USE HASH INDEX PAGES = RowPages|CURRENT Part of the USING INDEX {GLOBAL|LOCAL} clause and is optional. If specified, indicates a unique hash index is to be created for the primary key. If not specified, a unique range index is created. Can be used for both global and local indexes.

The PAGES clause is required and enables you to specify the expected page count value for the table.

If you specify RowPages, the number of pages is used to calculate the page count value. To determine the value for RowPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for RowPages (256000/256=1000).

The value for RowPages must be a positive constant and must be greater than 0.

If you specify CURRENT, the current number of rows in the table is used to calculate the page count value.

TimesTen recommends that you do not specify PAGES=CURRENT if there are no rows in your table.

USING INDEX (CreateIndexStmt) Part of the [UsingIndexClause2] clause. When this USING INDEX clause is specified, the (CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must create a unique index as that is the requirement for a primary key constraint. See "CREATE INDEX" for information on the CREATE INDEX statement.

Parameters for ALTER TABLE ADD UNIQUE CONSTRAINT: TimesTen Scaleout

Parameter Description
ALTER TABLE [Owner.]TableName Start of ALTER TABLE statement. Name of table required. Owner is optional.
ADD UNIQUE (ColumnName [,…]) Clause indicating that the table is to be altered by adding a unique constraint. (ColumnName) is required and specifies the column(s) to be used for the unique constraint.
[UsingIndexClause1] UsingIndexClause1 is optional and is described in the remainder of this table. You cannot specify two USING INDEX clauses in the ALTER TABLE definition. This clause enables you to define a global or local index for the PRIMARY KEY
USING INDEX {GLOBAL|LOCAL} Part of [UsingIndexClause1]. If specified, indicates if a global or local index is to be created for the unique constraint.
USING INDEX (CreateIndexStmt) Part of the [UsingIndexClause1] clause. When this USING INDEX clause is specified, the (CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must create a unique index as that is the requirement for a unique constraint. See "CREATE INDEX" for information on the CREATE INDEX statement.

Additional parameters for ALTER TABLE: TimesTen Scaleout

Parameter Description

[Owner.] TableName

Identifies the table to be altered.

DistributionClause

See "CREATE TABLE" for information on syntax.

UNIQUE

Specifies that in the column ColumnName each row must contain a unique value.

MODIFY

Specifies that an attribute of a given column is to be changed to a new value.

DEFAULT [DefaultVal |NULL]

Specifies that the column has a default value, DefaultVal. If NULL, specifies that the default value of the columns is to be dropped. If a column with a default value of SYSDATE is added, the value of the column of the existing rows only is the system date at the time the column was added. If the default value is one of the USER functions the column value is the user value of the session that executed the ALTER TABLE statement. Currently, you cannot assign a default value for the ROWID data type.

Altering the default value of a column has no impact on existing rows.

Note: To add a NOT NULL column to a table that is part of a replication scheme, DDL_REPLICATON_LEVEL must be 3 or greater.

ColumnName

Name of the column participating in the ALTER TABLE statement. A new column cannot have the same name as an existing column or another new column. If you add a NOT NULL column, you must include the DEFAULT clause.

ColumnDataType

Type of the column to be added. Some types require additional parameters. See "Data Types" for the data types that can be specified.

NOT NULL [ENABLE]

If you add a column, you can specify NOT NULL. If you specify NOT NULL, then you must include the DEFAULT clause. Optionally, you can specify ENABLE after the NOT NULL clause. Because NOT NULL constraints are always enabled, you are not required to specify ENABLE.

INLINE|NOT INLINE

By default, variable-length columns whose declared column length is > 128 bytes are stored out of line. Variable-length columns whose declared column length is <= 128 bytes are stored inline. The default behavior can be overridden during table creation through the use of the INLINE and NOT INLINE keywords.

CONSTRAINT

Specifies that a foreign key is to be dropped. Optionally specifies that an added foreign key is named by the user.

DROP UNIQUE (ColumnName) Indicates that unique constraint is to dropped. ColumnName is the name of the constraint.

ForeignKeyName

Name of the foreign key to be added or dropped. All foreign keys are assigned a default name by the system if the name was not specified by the user. Either the user-provided name or system name can be specified in the DROP FOREIGN KEY clause.

FOREIGN KEY

Specifies that a foreign key is to be added.

REFERENCES

Specifies that the foreign key references another table.

RefTableName

The name of the table that the foreign key references.

[ON DELETE CASCADE]

Enables the ON DELETE CASCADE referential action. If specified, when rows containing referenced key values are deleted from a parent table, rows in child tables with dependent foreign key values are also deleted.

USE HASH INDEX PAGES = RowPages | CURRENT

Changes primary key to use a hash index. If the primary key already uses a hash index, then this clause is equivalent to the SET PAGES clause.

USE RANGE INDEX

Changes primary key to use a range index. If the primary key already uses a range index, TimesTen ignores this clause.

SET PAGES = RowPages | CURRENT

Resizes the hash index to reflect the expected number of pages in the table. If you specify CURRENT, the current number of rows in the table is used to calculate the page count value. If you specify RowPages, the number of pages is used. To determine the value for RowPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for RowPages (256000/256=1000).

The value for RowPages must be a positive constant and must be greater than 0.

TimesTen recommends that you do not specify PAGES=CURRENT if there are no rows in your table.

If your estimate is too small, performance may be degraded. See "Column definition: TimesTen Scaleout" for more information on hash indexes.

Description for ALTER TABLE ADD PRIMARY KEY: TimesTen Scaleout

You have the option of specifying an additional clause after the PRIMARY KEY clause in your ALTER TABLE definition. This clause enables you to specify a global or local index for the primary key constraint.
  • The USING INDEX {GLOBAL | LOCAL} clause is one option that enables you to specify a global or local index for the primary key constraint. You must specify the GLOBAL or the LOCAL keyword. You can optionally specify the USE HASH INDEX clause after the USING INDEX {GLOBAL | LOCAL} clause if you want to define a hash index.

  • The USING INDEX (CreateIndexStmt) clause is your other option for specifying a global or local index. The(CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must create a unique index as that is the requirement for a primary key constraint. See "CREATE INDEX" for information on the CREATE INDEX statement.

Note:

You cannot use both the USING INDEX {GLOBAL | LOCAL} and the USING INDEX (CreateIndexStmt) in the ALTER TABLE definition. Specify one clause or the other or specify neither.

See "CREATE INDEX" for information on global and local indexes and their use in TimesTen Scaleout.

Description for ALTER TABLE ADD UNIQUE: TimesTen Scaleout

You have the option of specifying an additional clause after the UNIQUE clause in your ALTER TABLE definition. This clause enables you to specify a global or local index for the unique constraint.
  • The USING INDEX {GLOBAL | LOCAL} clause is one option that enables you to specify a global or local index for the primary key constraint. You must specify the GLOBAL or the LOCAL keyword.

  • The USING INDEX (CreateIndexStmt) clause is your other option for specifying a global or local index. The(CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must create a unique index as that is the requirement for a unique constraint. See "CREATE INDEX" for information on the CREATE INDEX statement.

Note:

You cannot use both the USING INDEX {GLOBAL | LOCAL} and the USING INDEX (CreateIndexStmt) in the ALTER TABLE definition. Specify one clause or the other or specify neither.

See CREATE INDEX for information on global and local indexes and their use in TimesTen Scaleout.

Additional ALTER TABLE information: TimesTen Scaleout

  • You can alter tables to change defaults or add and drop columns and constraints. However, you cannot change the distribution scheme unless the table is empty. In addition, you cannot drop a constraint that is named in the DISTRIBUTE BY REFERENCE clause. See "CREATE TABLE" for information on the distribution schemes. See "Altering tables" in Oracle TimesTen In-Memory Database Scaleout User's Guide for more information.

  • The ALTER TABLE statement cannot be used to alter a temporary table.

  • The ALTER TABLE ADD [COLUMN] ColumnName statement adds one or more new columns to an existing table. When you add one or more columns, the new columns are added to the end of all existing rows of the table in one new partition.

  • Columns referenced by materialized views cannot be dropped.

  • You cannot use the ALTER TABLE statement to add a column, drop a column, or add a constraint for cache group tables.

  • Only one partition is added to the table per statement regardless of the number of columns added.

  • You can ALTER a table to add a NOT NULL column with a default value. The DEFAULT clause is required. Restrictions include:

    • You cannot use the column as a primary key column. Specifically, you cannot specify the column in the statement: ALTER TABLE ADD ConstraintName PRIMARY KEY (ColumnName [,...]).

  • NULL is the initial value for all added columns, unless a default value is specified for the new column.

  • The total number of columns in the table cannot exceed 1000. In addition, the total number of partitions in a table cannot exceed 1000, one of which is used by TimesTen.

  • Do not specify the ADD CONSTRAINT ... PRIMARY KEY clause on a global temporary table.

  • As the result of an ALTER TABLE ADD statement, an additional read occurs for each new partition during queries. Therefore, altered tables may have slightly degraded performance. The performance can only by restored by dropping and recreating the table, or by using the ttMigrate create -c -relaxedUpgrade command, and restoring the table using the ttRestore -r -relaxedUpgrade command. Dropping the added column does not recover the lost performance or decrease the number of partitions.

  • When you use the ALTER TABLE DROP statement to remove one or more columns from an existing table, dropped columns are removed from all current rows of the table. Subsequent SQL statements must not attempt to make any use of the dropped columns. You cannot drop columns that are in the table's primary key. You cannot drop columns that are in any of the table's foreign keys until you have dropped all foreign keys. You cannot drop columns that are indexed until all indexes on the column have been dropped. ALTER TABLE cannot be used to drop all of the columns of a table. Use DROP TABLE instead.

  • When a column is dropped from a table, all commands referencing that table need to be recompiled. An error may result at recompilation time if a dropped column was referenced. The application must re-prepare those commands, and rebuild any parameters and result columns. When a column is added to a table, the commands that contain a SELECT * statement are invalidated. Only these commands must be re-prepared. All other commands continue to work as expected.

  • When you drop a column, the column space is not freed.

  • When you add a UNIQUE constraint, there is overhead incurred (in terms of additional space and additional time). This is because an index is created to maintain the UNIQUE constraint. You cannot use the DROP INDEX statement to drop an index used to maintain the UNIQUE constraint.

  • A UNIQUE constraint and its associated index cannot be dropped if it is being used as a unique index on a replicated table.

  • Use ALTER TABLE...USE RANGE INDEX if your application performs range queries over a table's primary key.

  • Use ALTER TABLE...USE HASH INDEX if your application performs exact match lookups on a table's primary key.

  • An error is generated if a table has no primary key and either the USE HASH INDEX clause or the USE RANGE INDEX clause is specified.

  • If ON DELETE CASCADE is specified on a foreign key constraint for a child table, a user can delete rows from a parent table for which the user has the DELETE privilege without requiring explicit DELETE privilege on the child table.

  • To change the ON DELETE CASCADE triggered action, drop then redefine the foreign key constraint.

  • ON DELETE CASCADE is supported on detail tables of a materialized view. If you have a materialized view defined over a child table, a deletion from the parent table causes cascaded deletes in the child table. This, in turn, triggers changes in the materialized view.

  • The total number of rows reported by the DELETE statement does not include rows deleted from child tables as a result of the ON DELETE CASCADE action.

  • For ON DELETE CASCADE, since different paths may lead from a parent table to a child table, the following rule is enforced:

  • Either all paths from a parent table to a child table are "delete" paths or all paths from a parent table to a child table are "do not delete" paths.

    • Specify ON DELETE CASCADE on all child tables on the "delete" path.

    • This rule does not apply to paths from one parent to different children or from different parents to the same child.

  • For ON DELETE CASCADE, a second rule is also enforced:

  • If a table is reached by a "delete" path, then all its children are also reached by a "delete" path.

  • The ALTER TABLE ADD/DROP CONSTRAINT statement has the following restrictions:

    • When a foreign key is dropped, TimesTen also drops the index associated with the foreign key. Attempting to drop an index associated with a foreign key using the regular DROP INDEX statement results in an error.

    • Foreign keys cannot be added or dropped on views or temporary tables.

    • You cannot use ALTER TABLE to drop a primary key constraint. You would have to drop and recreate the table in order to drop the constraint.

Examples: Add primary key constraint using global indexes in TimesTen Scaleout

These examples show various uses of the syntax for using global indexes with ALTER TABLE ADD PRIMARY KEY.

Create a table. Alter the table and add a primary key constraint. Specify the USING INDEX GLOBAL clause. Drop the table.
Command> CREATE TABLE mytab1 (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
           a TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> ALTER TABLE mytab1 ADD CONSTRAINT pk PRIMARY KEY (c,b) USING INDEX GLOBAL;
Command> indexes mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  PK: global unique range index on columns:
    C
    B
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab1;

Create a table. Alter the table adding a primary key constraint. Specify the USING INDEX GLOBAL with the USE HASH INDEX PAGES clause. Drop the table.

Command> CREATE TABLE mytab1 (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
           a TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> ALTER TABLE mytab1 ADD CONSTRAINT pk PRIMARY KEY (c,b) 
           USING INDEX GLOBAL USE HASH INDEX PAGES =200;
Command> INDEXES mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  PK: global unique hash index on columns:
    C
    B
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab1;

Create a table. Alter the table adding a primary key constraint. Specify the USING INDEX (CreateIndexStmt) clause. The (CreateIndexStmt) clause is the TimesTen CREATE INDEX statement. See "CREATE INDEX" for information on this statement.

Command> CREATE TABLE mytab1 (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
           a TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> ALTER TABLE mytab1 ADD CONSTRAINT pk PRIMARY KEY (c,b) 
           USING INDEX (CREATE GLOBAL UNIQUE HASH INDEX myglobalix ON mytab1 (c,b) PAGES =200);
Command> indexes mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  MYGLOBALIX: global unique hash index on columns:
    C
    B
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab1;
This example illustrates that you cannot use the USING INDEX GLOBAL|LOCAL clause with the USING INDEX (CreateIndexStmt) clause.
Command> CREATE TABLE mytab1 (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
           a TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> ALTER TABLE mytab1 ADD CONSTRAINT pk PRIMARY KEY (c,b) 
           USING INDEX GLOBAL USE HASH INDEX PAGES = 200 
           USING INDEX (CREATE GLOBAL UNIQUE HASH INDEX myglobalix ON mytab1 (c,b) PAGES =200);
 1001: Syntax error in SQL statement  before or at: "USING", character position: 102
...USING INDEX GLOBAL USE HASH INDEX PAGES = 200 USING INDEX (CREATE G...
                                                 ^^^^^
The command failed.

Examples: Add unique constraint using global indexes in TimesTen Scaleout

These examples show various uses of the syntax for using global indexes with ALTER TABLE ADD UNIQUE CONSTRAINT.

Create a table. Alter the table adding a unique constraint. Drop the table. Create the table again adding a unique constraint and specifying the USING INDEX GLOBAL clause.

Command> CREATE TABLE mytab1 (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
           a TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> ALTER TABLE mytab1 ADD UNIQUE (a);
Command> indexes mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  TTUNIQUE_6E6: unique range index on columns:
    A
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab1;
Command> CREATE TABLE mytab1 (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
           a TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> ALTER TABLE mytab1 ADD UNIQUE (a) USING INDEX GLOBAL;
Command> indexes mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  $GUA8C5B4ECE6D8: global unique range index on columns:
    A
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab1;

Create a table. Alter the table adding a unique constraint and use the USING INDEX (CreateIndexStmt) clause to create a local unique index. Alter the table a second time adding another unique constaint. Use the USING INDEX (CreateIndexStmt) clause to create a global unique index.

Command> CREATE TABLE mytab1 (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
           a TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> ALTER TABLE mytab1 ADD UNIQUE (b) USING INDEX (CREATE UNIQUE INDEX myuniqueidxB ON mytab1 (b));
Command> indexes mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  MYUNIQUEIDXB: unique range index on columns:
    B
  1 index found.

1 index found on 1 table.
Command> ALTER TABLE mytab1 ADD UNIQUE (c) USING INDEX (CREATE GLOBAL UNIQUE INDEX myuniqueidxC ON mytab1 (c));
Command> indexes mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  MYUNIQUEIDXB: unique range index on columns:
    B
  MYUNIQUEIDXC: global unique range index on columns:
    C
  2 indexes found.

2 indexes found on 1 table.
Command> DROP TABLE mytab1;

Additional examples for ALTER TABLE: TimesTen Scaleout

Table 6-6 shows the rules associated with altering tables. Supporting examples follow.

Table 6-6 ALTER TABLE rules

ALTER statement Comment
ALTER TABLE t1 ADD CONSTRAINT c1
 PRIMARY KEY (p);

The primary key constraint is added to the table. The distribution key is not changed.

CREATE TABLE t1 (c1 NUMBER, 
 c2 VARCHAR2 (10));

ALTER TABLE t1 
 DISTRIBUTE BY HASH (c1);

The operation succeeds if the table is empty. If the table is not empty, the operation fails because the distribution key cannot be changed on tables that are not empty.

ALTER TABLE t1 ADD CONSTRAINT c1
 FOREIGN KEY (f1)REFERENCES t2 (c2);

The operation succeeds. The distribution of the t1 table is not related to the c1 constraint.

CREATE TABLE t1...CONSTRAINT fk1...
 DISTRIBUTE BY REFERENCE(fk1);

ALTER TABLE t1 DROP CONSTRAINT(fk1);

The operation fails. The foreign key is used to distribute the table.

These examples support the information in the "Table 6-6" table:

Use ALTER TABLE to add a primary key constraint

This example creates the mytable table without a primary key or distribution clause. The table is distributed by hash on a hidden column. Then the ALTER TABLE statement is used to add a primary key constraint. The operation succeeds but the distribution key is not changed.

Command> CREATE TABLE mytable (col1 NUMBER NOT NULL, col2 VARCHAR2 (32));
Command> describe mytable;
 
Table SAMPLEUSER.MYTABLE:
  Columns:
    COL1                            NUMBER NOT NULL
    COL2                            VARCHAR2 (32) INLINE
  DISTRIBUTE BY HASH
 
1 table found.
(primary key columns are indicated with *)

Now alter the table to add the primary key. The operation succeeds. The distribution scheme and distribution key do not change.

Command> ALTER TABLE mytable ADD CONSTRAINT c1 PRIMARY KEY (col1);
Command> describe mytable;
 
Table SAMPLEUSER.MYTABLE:
  Columns:
   *COL1                            NUMBER NOT NULL
    COL2                            VARCHAR2 (32) INLINE
  DISTRIBUTE BY HASH
 
1 table found.
(primary key columns are indicated with *)

Add a primary key constraint on table distributed on unique column

This example creates the mytab table and distributes the data by hash on the id2 unique column. The example then alters the mytab table adding the primary key constraint on the id column. A ttIsql describe command shows the table remains distributed by hash on the id2 column.

Command> CREATE TABLE mytab (id TT_INTEGER NOT NULL, id2 TT_INTEGER UNIQUE, 
           id3 TT_INTEGER) distribute by hash (id2);
Command> ALTER TABLE mytab ADD CONSTRAINT c1 PRIMARY KEY (id);                  Command> describe mytab;
 
Table SAMPLEUSER.MYTAB:
  Columns:
   *ID                              TT_INTEGER NOT NULL
    ID2                             TT_INTEGER UNIQUE
    ID3                             TT_INTEGER
  DISTRIBUTE BY HASH (ID2)
 
1 table found.
(primary key columns are indicated with *)

Use ALTER TABLE to change the distribution key

This example shows that you can use the ALTER TABLE statement to change the distribution key, but only if the table is empty.

Command> CREATE TABLE mytable2 (col1 NUMBER NOT NULL, col2 VARCHAR2 (32))
         DISTRIBUTE BY HASH (col1,col2);
Command> describe mytable2;
 
Table SAMPLEUSER.MYTABLE2:
  Columns:
    COL1                            NUMBER NOT NULL
    COL2                            VARCHAR2 (32) INLINE
  DISTRIBUTE BY HASH (COL1, COL2)
 
1 table found.
(primary key columns are indicated with *)

Use the ALTER TABLE statement to change the distribution key to col1. The operation succeeds because the table is empty.

Command> ALTER TABLE mytable2 DISTRIBUTE BY HASH (col1);
Command> describe mytable2;
 
Table SAMPLEUSER.MYTABLE2:
  Columns:
    COL1                            NUMBER NOT NULL
    COL2                            VARCHAR2 (32) INLINE
  DISTRIBUTE BY HASH (COL1)
 
1 table found.
(primary key columns are indicated with *)

Insert a row of data and attempt to change the distribution key back to col1, col2. The operation fails because the table is not empty.

Command> INSERT INTO mytable2 VALUES (10, 'test');
1 row inserted.
Command> commit;
Command> ALTER TABLE mytable2 DISTRIBUTE BY HASH (col1,col2);
 1069: Table not empty. Alter table distribution is only permitted on empty
tables.
The command failed.

Add a foreign key constraint that is not part of the distribution key

This example first describes the accounts and accounts2 tables. The example then alters the accounts2 table, adding a foreign key constraint. Since this constraint is not part of the accounts2 table distribution, the operation succeeds.

Command> describe accounts;
 
Table SAMPLEUSER.ACCOUNTS:
  Columns:
   *ACCOUNT_ID                      NUMBER (10) NOT NULL
    PHONE                           VARCHAR2 (15) INLINE NOT NULL
    ACCOUNT_TYPE                    CHAR (1) NOT NULL
    STATUS                          NUMBER (2) NOT NULL
    CURRENT_BALANCE                 NUMBER (10,2) NOT NULL
    PREV_BALANCE                    NUMBER (10,2) NOT NULL
    DATE_CREATED                    DATE NOT NULL
    CUST_ID                         NUMBER (10) NOT NULL
  DISTRIBUTE BY REFERENCE (FK_CUSTOMER)
 
1 table found.
(primary key columns are indicated with *)

Command> describe accounts2;
 
Table SAMPLEUSER.ACCOUNTS2:
  Columns:
   *ACCOUNTS2_ID                    NUMBER (10) NOT NULL
    ACCOUNT_ORIG_ID                 NUMBER (10) NOT NULL
    STATUS                          NUMBER (2) NOT NULL
  DISTRIBUTE BY HASH (ACCOUNTS2_ID)
 
1 table found.
(primary key columns are indicated with *)

Command> ALTER TABLE accounts2 ADD CONSTRAINT accounts2_fk FOREIGN KEY 
           (account_orig_id) REFERENCES accounts (account_id);

Use the ttIsql indexes command to show the accounts2_fk constraint is created successfully.

Command> indexes accounts2;
 
Indexes on table SAMPLEUSER.ACCOUNTS2:
  ACCOUNTS2: unique range index on columns:
    ACCOUNTS2_ID
  ACCOUNTS2_FK: non-unique range index on columns:
    ACCOUNT_ORIG_ID
    (foreign key index references table SAMPLEUSER.ACCOUNTS(ACCOUNT_ID))
  2 indexes found.
 
2 indexes found on 1 table.

Attempt to drop a foreign key constraint used as a distribution key

This example attempts to drop the fk_accounts constraint. Since the constraint is used as the distribution key, the operation fails.

Command> describe transactions;
 
Table SAMPLEUSER.TRANSACTIONS:
  Columns:
   *TRANSACTION_ID                  NUMBER (10) NOT NULL
   *ACCOUNT_ID                      NUMBER (10) NOT NULL
   *TRANSACTION_TS                  TIMESTAMP (6) NOT NULL
    DESCRIPTION                     VARCHAR2 (60) INLINE
    OPTYPE                          CHAR (1) NOT NULL
    AMOUNT                          NUMBER (6,2) NOT NULL
  DISTRIBUTE BY REFERENCE (FK_ACCOUNTS)
 
1 table found.
(primary key columns are indicated with *)

Command> ALTER TABLE transactions DROP CONSTRAINT fk_accounts;
 1072: Dropping a table's reference by distribution foreign key is not allowed.
The command failed.

SQL syntax for ALTER TABLE: TimesTen Classic

To add one column:

ALTER TABLE [Owner.]TableName 
  ADD [COLUMN] ColumnName ColumnDataType
    [DEFAULT DefaultVal] [[NOT] INLINE] [UNIQUE] [NULL]
  [COMPRESS (CompressColumns [,...])]

To add multiple columns:

ALTER TABLE [Owner.]TableName 
 ADD (ColumnName ColumnDataType 
      [DEFAULT DefaultVal] [[NOT] INLINE] [UNIQUE] [NULL] [,... ] )
  [COMPRESS (CompressColumns [,...])]

To add a NOT NULL column (note that the DEFAULT clause is required):

ALTER TABLE [Owner.]TableName
  ADD [COLUMN] ColumnName ColumnDataType
    NOT NULL [ENABLE] DEFAULT DefaultVal [[NOT] INLINE] [UNIQUE]
  [COMPRESS (CompressColumns [,...])]

To add multiple NOT NULL columns (note that the DEFAULT clause is required):

ALTER TABLE [Owner.]TableName
  ADD (ColumnName ColumnDataType
       NOT NULL [ENABLE] DEFAULT DefaultVal [[NOT] INLINE] [UNIQUE] [,...])
  [COMPRESS (CompressColumns [,...])]

The CompressColumns syntax is as follows:

{ColumnDefinition | (ColumnDefinition [,...])} BY DICTIONARY 
   [MAXVALUES = CompressMax]

To remove columns.

ALTER TABLE [Owner.]TableName 
  DROP {[COLUMN] ColumnName | (ColumnName [,... ] )}

Note:

If removing columns in a compressed column group, all columns in the compressed column group must be specified.

To add a primary key constraint using a range index:

ALTER TABLE [Owner.]TableName ADD CONSTRAINT ConstraintName
  PRIMARY KEY (ColumnName [,... ])

To add a primary key constraint using a hash index:

ALTER TABLE [Owner.]TableName ADD CONSTRAINT ConstraintName
  PRIMARY KEY (ColumnName [,... ])
  USE HASH INDEX PAGES = RowPages | CURRENT

To add a foreign key and optionally add ON DELETE CASCADE:

ALTER TABLE [Owner.]TableName 
ADD [CONSTRAINT ForeignKeyName] FOREIGN KEY
    (ColumnName [,...]) REFERENCES RefTableName
       [(ColumnName [,...])] [ON DELETE CASCADE]

To remove a foreign key:

ALTER TABLE [Owner.]TableName 
DROP CONSTRAINT ForeignKeyName

Note:

You cannot use ALTER TABLE to drop a primary key constraint. To drop the constraint, drop and recreate the table.

To resize a hash index:

ALTER TABLE [Owner.]TableName
SET PAGES = RowPages | CURRENT

To change the primary key to use a hash index:

ALTER TABLE [Owner.]TableName
USE HASH INDEX PAGES = RowPages | CURRENT

To change the primary key to use a range index with the USE RANGE INDEX clause:

ALTER TABLE [Owner.]TableName
USE RANGE INDEX

To change the default value of a column:

ALTER TABLE [Owner.]TableName
MODIFY (ColumnName DEFAULT DefaultVal)

To add or drop a unique constraint on a column:

ALTER TABLE Owner.]TableName
{ADD | DROP} UNIQUE (ColumnName)

To remove the default value of a column that is nullable, by changing it to NULL:

ALTER TABLE [Owner.]TableName
MODIFY (ColumnName DEFAULT NULL)

To add LRU aging:

ALTER TABLE [Owner.]TableName
ADD AGING LRU [ON | OFF]

To add time-based aging:

ALTER TABLE [Owner.]TableName
ADD AGING USE ColumnName LIFETIME num1
    {SECOND[S] | MINUTE[S] | HOUR[S] | DAY[S]}
     [CYCLE num2 {SECOND[S] | MINUTE[S] | HOUR[S] | DAY[S] }]
    [ON | OFF]

To change the aging state:

ALTER TABLE [Owner.]TableName
SET AGING {ON | OFF}

To drop aging:

ALTER TABLE [Owner.]TableName
DROP AGING

To change the lifetime for time-based aging:

ALTER TABLE [Owner.]TableName
SET AGING LIFETIME num1 {SECOND[S] | MINUTE[S] | HOUR[S] | DAY[S]}

To change the cycle for time-based aging:

ALTER TABLE [Owner.]TableName
SET AGING CYCLE num2 {SECOND[S] | MINUTE[S] | HOUR[S] | DAY[S]}

Parameters for ALTER TABLE: TimesTen Classic

Parameter Description

[Owner.] TableName

Identifies the table to be altered.

UNIQUE

Specifies that in the column ColumnName each row must contain a unique value.

MODIFY

Specifies that an attribute of a given column is to be changed to a new value.

DEFAULT [DefaultVal |NULL]

Specifies that the column has a default value, DefaultVal. If NULL, specifies that the default value of the columns is to be dropped. If a column with a default value of SYSDATE is added, the value of the column of the existing rows only is the system date at the time the column was added. If the default value is one of the USER functions the column value is the user value of the session that executed the ALTER TABLE statement. Currently, you cannot assign a default value for the ROWID data type.

Altering the default value of a column has no impact on existing rows.

Note: To add a NOT NULL column to a table that is part of a replication scheme, DDL_REPLICATON_LEVEL must be 3 or greater.

ColumnName

Name of the column participating in the ALTER TABLE statement. A new column cannot have the same name as an existing column or another new column. If you add a NOT NULL column, you must include the DEFAULT clause.

ColumnDataType

Type of the column to be added. Some types require additional parameters. See "Data Types" for the data types that can be specified.

NOT NULL [ENABLE]

If you add a column, you can specify NOT NULL. If you specify NOT NULL, then you must include the DEFAULT clause. Optionally, you can specify ENABLE after the NOT NULL clause. Because NOT NULL constraints are always enabled, you are not required to specify ENABLE.

INLINE|NOT INLINE

By default, variable-length columns whose declared column length is > 128 bytes are stored out of line. Variable-length columns whose declared column length is <= 128 bytes are stored inline. The default behavior can be overridden during table creation through the use of the INLINE and NOT INLINE keywords.

COMPRESS (CompressColumns [,...])

Defines a compressed column group for a table that is enabled for compression. This can include one or more columns in the table.

If you define multiple columns for a compression group, you must specify the columns as INLINE. An out-of-line column cannot be in a multi-column compression group.

Each compressed column group is limited to a maximum of 16 columns.

See "Column-based compression of tables (TimesTen Classic)" for details on compression columns.

BY DICTIONARY

Defines a compression dictionary for each compressed column group.

MAXVALUES = CompressMax

CompressMax is the total number of distinct values in the table and sets the size for the compressed column group pointer column to 1, 2, or 4 bytes and sets the size for the maximum number of entries in the dictionary table.

For the dictionary table, NULL is counted as one unique value.

CompressMax takes an integer between 1 and 232-1.

The maximum size defaults to size of 232-1 if the MAXVALUES clause is omitted, which uses 4 bytes for the pointer column. An error is thrown if the value is greater than 232-1.

See "Column-based compression of tables (TimesTen Classic)" for details on maximum sizing for compression dictionaries.

ADD CONSTRAINT ConstraintName PRIMARY KEY (ColumnName

[,... ] ) [USE HASH INDEX PAGES = RowPages | CURRENT]

Adds a primary key constraint to the table. Columns of the primary key must be defined as NOT NULL.

Specify ConstraintName as the name of the index used to enforce the primary key constraint. Specify ColumnName as the name(s) of the NOT NULL column(s) used for the primary key.

Specify the USE HASH INDEX clause to use a hash index for the primary key. If not specified, a range index is used for the primary key constraint.

If you specify CURRENT, the current number of rows in the table is used to calculate the page count value. If you specify RowPages, the number of pages is used. To determine the value for RowPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for RowPages (256000/256=1000).

The value for RowPages must be a positive constant and must be greater than 0.

TimesTen recommends that you do not specify PAGES=CURRENT if there are no rows in your table.

If your estimate is too small, performance may be degraded. See "Column definition: TimesTen Classic" for more information on hash indexes.

Note: Before you use ADD CONSTRAINT to add a named primary key constraint, be aware that you cannot use ALTER TABLE to drop a primary key constraint. You would have to drop and recreate the table in order to drop the constraint.

CONSTRAINT

Specifies that a foreign key is to be dropped. Optionally specifies that an added foreign key is named by the user.

ForeignKeyName

Name of the foreign key to be added or dropped. All foreign keys are assigned a default name by the system if the name was not specified by the user. Either the user-provided name or system name can be specified in the DROP FOREIGN KEY clause.

FOREIGN KEY

Specifies that a foreign key is to be added.

REFERENCES

Specifies that the foreign key references another table.

RefTableName

The name of the table that the foreign key references.

[ON DELETE CASCADE]

Enables the ON DELETE CASCADE referential action. If specified, when rows containing referenced key values are deleted from a parent table, rows in child tables with dependent foreign key values are also deleted.

USE HASH INDEX PAGES = RowPages | CURRENT

Changes primary key to use a hash index. If the primary key already uses a hash index, then this clause is equivalent to the SET PAGES clause.

USE RANGE INDEX

Changes primary key to use a range index. If the primary key already uses a range index, TimesTen ignores this clause.

SET PAGES = RowPages | CURRENT

Resizes the hash index to reflect the expected number of pages in the table. If you specify CURRENT, the current number of rows in the table is used to calculate the page count value. If you specify RowPages, the number of pages is used. To determine the value for RowPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for RowPages (256000/256=1000).

The value for RowPages must be a positive constant and must be greater than 0.

TimesTen recommends that you do not specify PAGES=CURRENT if there are no rows in your table.

If your estimate is too small, performance may be degraded. See "Column definition: TimesTen Classic" for more information on hash indexes.

ADD AGING LRU [ON | OFF]

Adds least recently used (LRU) aging to an existing table that has no aging policy defined.

The LRU aging policy defines the type of aging (least recently used (LRU)), the aging state (ON or OFF) and the LRU aging attributes.

Set the aging state to either ON or OFF. ON indicates that the aging state is enabled and aging is done automatically. OFF indicates that the aging state is disabled and aging is not done automatically. In both cases, the aging policy is defined. The default is ON.

LRU attributes are defined by calling the ttAgingLRUConfig and/or the ttAgingTableLRUConfig built-in procedures. LRU attributes are not defined at the SQL level. See "ttAgingLRUConfig" and "ttAgingTableLRUConfig" in the Oracle TimesTen In-Memory Database Reference and "Implementing an aging policy in your tables" in the Oracle TimesTen In-Memory Database Operations Guide for more information.

ADD AGING USE ColumnName...[ON| OFF]

Adds time-based aging to an existing table that has no aging policy defined.

The time-based aging policy defines the type of aging (time-based), the aging state (ON or OFF) and the time-based aging attributes.

Set the aging state to either ON or OFF. ON indicates that the aging state is enabled and aging is done automatically. OFF indicates that the aging state is disabled and aging is not done automatically. In both cases, the aging policy is defined. The default is ON.

Time-based aging attributes are defined at the SQL level and are specified by the LIFETIME and CYCLE clauses.

Specify ColumnName as the name of the column used for time-based aging. Define the column as NOT NULL and of data type TIMESTAMP or DATE. The value of this column is subtracted from SYSDATE, truncated using the specified unit (minute, hour, day) and then compared to the LIFETIME value. If the result is greater than the LIFETIME value, then the row is a candidate for aging.

The values of the column used for aging are updated by your applications. If the value of this column is unknown for some rows, and you do not want the rows to be aged, define the column with a large default value (the column cannot be NULL).

You can define your aging column with a data type of TT_TIMESTAMP or TT_DATE. If you choose data type TT_DATE, then you must specify the LIFETIME unit as days.

For more information about time-based aging, see "Implementing an aging policy in your tables" in Oracle TimesTen In-Memory Database Operations Guide.

LIFETIME Num1 {SECOND[S] | MINUTE[S] | HOUR[S] | DAY[S]

Specify the LIFETIME clause after the ADD AGING USE ColumnName clause if you are adding the time-based aging policy to an existing table. Specify the LIFETIME clause after the SET AGING clause to change the LIFETIME setting.

The LIFETIME clause specifies the minimum amount of time data is kept in cache.

Specify Num1 as a positive integer constant to indicate the unit of time expressed in seconds, minutes, hours or days that rows should be kept in cache. Rows that exceed the LIFETIME value are aged out (deleted from the table). If you define your aging column with data type TT_DATE, then you must specify DAYS as the LIFETIME unit.

The concept of time resolution is supported. If DAYS is specified as the time resolution, then all rows whose timestamp belongs to the same day are aged out at the same time. If HOURS is specified as the time resolution, then all rows with timestamp values within that hour are aged at the same time. A LIFETIME of 3 days is different than a LIFETIME of 72 hours (3*24) or a LIFETIME of 432 minutes (3*24*60).

CYCLE Num2 {SECOND[S]| MINUTE[S]| HOUR[S]|DAY[S]}

Specify the optional CYCLE clause after the LIFETIME clause if you are adding the time-based aging policy to an existing table.

CYCLE is a time-based aging attribute.

The CYCLE clause indicates how often the system should examine rows to see if data exceeds the specified LIFETIME value and should be aged out (deleted).

Specify Num2 as a positive integer constant.

If you do not specify the CYCLE clause, then the default value is 5 minutes. If you specify 0 for Num2, then the aging thread wakes up every second.

If the aging state is OFF, then aging is not done automatically and the CYCLE clause is ignored.

Specify the CYCLE clause after the SET AGING clause to change the CYCLE setting.

SET AGING {ON|OFF}

Changes the aging state. The aging policy must be previously defined. ON enables automatic aging. OFF disables automatic aging. To control aging with an external scheduler, then disable aging and invoke the ttAgingScheduleNow built-in procedure.

DROP AGING

Drops the aging policy from the table. After you define an aging policy, you cannot alter it. Drop aging, then redefine.

SET AGING LIFETIME Num1 {SECOND[S]| MINUTE[S]|HOUR[S] |DAY[S]}

Use this clause to change the lifetime for time-based aging.

Num1 must be a positive integer constant.

If you defined your aging column with data type TT_DATE, then you must specify DAYS as the LIFETIME unit.

SET AGING CYCLE Num2 {SECOND[S]| MINUTE[S]| HOUR[S]|DAY[S]}

Use this clause to change the cycle for time-based aging.

Num2 must be a positive integer constant.

Description for ALTER TABLE: TimesTen Classic

  • The ALTER TABLE statement cannot be used to alter a temporary table.

  • The ALTER TABLE ADD [COLUMN] ColumnName statement adds one or more new columns to an existing table. When you add one or more columns, the new columns are added to the end of all existing rows of the table in one new partition.

  • The ALTER TABLE ADD or DROP COLUMN statement can be used to add or drop columns from replicated tables.

    Do not use ALTER TABLE to alter a replicated table that is part of a TWOSAFE BY REQUEST transaction.

  • Columns referenced by materialized views cannot be dropped.

  • You cannot use the ALTER TABLE statement to add a column, drop a column, or add a constraint for cache group tables.

  • Only one partition is added to the table per statement regardless of the number of columns added.

  • You can ALTER a table to add a NOT NULL column with a default value. The DEFAULT clause is required. Restrictions include:

    • You cannot use the column as a primary key column. Specifically, you cannot specify the column in the statement: ALTER TABLE ADD ConstraintName PRIMARY KEY (ColumnName [,...]).

    • You cannot use the column for time-based aging. Specifically, you cannot specify the column in the statement ALTER TABLE ADD AGING USE ColumnName.

      Note:

      To add a NOT NULL column to a table that is part of a replication scheme, DDL_REPLICATON_LEVEL must be 3 or greater.
  • NULL is the initial value for all added columns, unless a default value is specified for the new column.

  • The total number of columns in the table cannot exceed 1000. In addition, the total number of partitions in a table cannot exceed 1000, one of which is used by TimesTen.

  • Use the ADD CONSTRAINT ... PRIMARY KEY clause to add a primary key constraint to a regular table or to a detailed or materialized view table. Do not use this clause on a table that already has a primary key.

  • If you use the ADD CONSTRAINT... PRIMARY KEY clause to add a primary key constraint, and you do not specify the USE HASH INDEX clause, then a range index is used for the primary key constraint.

  • If a table is replicated and the replication agent is active, you cannot use the ADD CONSTRAINT ... PRIMARY KEY clause. Stop the replication agent first.

  • Do not specify the ADD CONSTRAINT ... PRIMARY KEY clause on a global temporary table.

  • Do not specify the ADD CONSTRAINT ... PRIMARY KEY clause on a cache group table because cache group tables defined with a primary key must be defined in the CREATE CACHE GROUP statement.

  • As the result of an ALTER TABLE ADD statement, an additional read occurs for each new partition during queries. Therefore, altered tables may have slightly degraded performance. The performance can only by restored by dropping and recreating the table, or by using the ttMigrate create -c -relaxedUpgrade command, and restoring the table using the ttRestore -r -relaxedUpgrade command. Dropping the added column does not recover the lost performance or decrease the number of partitions.

  • When you use the ALTER TABLE DROP statement to remove one or more columns from an existing table, dropped columns are removed from all current rows of the table. Subsequent SQL statements must not attempt to make any use of the dropped columns. You cannot drop columns that are in the table's primary key. You cannot drop columns that are in any of the table's foreign keys until you have dropped all foreign keys. You cannot drop columns that are indexed until all indexes on the column have been dropped. ALTER TABLE cannot be used to drop all of the columns of a table. Use DROP TABLE instead.

  • When a column is dropped from a table, all commands referencing that table need to be recompiled. An error may result at recompilation time if a dropped column was referenced. The application must re-prepare those commands, and rebuild any parameters and result columns. When a column is added to a table, the commands that contain a SELECT * statement are invalidated. Only these commands must be re-prepared. All other commands continue to work as expected.

  • When you drop a column, the column space is not freed.

  • When you add a UNIQUE constraint, there is overhead incurred (in terms of additional space and additional time). This is because an index is created to maintain the UNIQUE constraint. You cannot use the DROP INDEX statement to drop an index used to maintain the UNIQUE constraint.

  • A UNIQUE constraint and its associated index cannot be dropped if it is being used as a unique index on a replicated table.

  • Use ALTER TABLE...USE RANGE INDEX if your application performs range queries over a table's primary key.

  • Use ALTER TABLE...USE HASH INDEX if your application performs exact match lookups on a table's primary key.

  • An error is generated if a table has no primary key and either the USE HASH INDEX clause or the USE RANGE INDEX clause is specified.

  • Make sure to stop the replication agent before adding or dropping a foreign key on a replicated table.

  • If ON DELETE CASCADE is specified on a foreign key constraint for a child table, a user can delete rows from a parent table for which the user has the DELETE privilege without requiring explicit DELETE privilege on the child table.

  • To change the ON DELETE CASCADE triggered action, drop then redefine the foreign key constraint.

  • ON DELETE CASCADE is supported on detail tables of a materialized view. If you have a materialized view defined over a child table, a deletion from the parent table causes cascaded deletes in the child table. This, in turn, triggers changes in the materialized view.

  • The total number of rows reported by the DELETE statement does not include rows deleted from child tables as a result of the ON DELETE CASCADE action.

  • For ON DELETE CASCADE, since different paths may lead from a parent table to a child table, the following rule is enforced:

  • Either all paths from a parent table to a child table are "delete" paths or all paths from a parent table to a child table are "do not delete" paths.

    • Specify ON DELETE CASCADE on all child tables on the "delete" path.

    • This rule does not apply to paths from one parent to different children or from different parents to the same child.

  • For ON DELETE CASCADE, a second rule is also enforced:

  • If a table is reached by a "delete" path, then all its children are also reached by a "delete" path.

  • For ON DELETE CASCADE with replication, the following restrictions apply:

    • The foreign keys specified with ON DELETE CASCADE must match between the Master and subscriber for replicated tables. Checking is done at runtime. If there is an error, the receiver thread stops working.

    • All tables in the delete cascade tree have to be replicated if any table in the tree is replicated. This restriction is checked when the replication scheme is created or when a foreign key with ON DELETE CASCADE is added to one of the replication tables. If an error is found, the operation is aborted. You may be required to drop the replication scheme first before trying to change the foreign key constraint.

  • The ALTER TABLE ADD/DROP CONSTRAINT statement has the following restrictions:

    • When a foreign key is dropped, TimesTen also drops the index associated with the foreign key. Attempting to drop an index associated with a foreign key using the regular DROP INDEX statement results in an error.

    • Foreign keys cannot be added or dropped on tables in a cache group.

    • Foreign keys cannot be added or dropped on views or temporary tables.

    • You cannot use ALTER TABLE to drop a primary key constraint. You would have to drop and recreate the table in order to drop the constraint.

  • After you have defined an aging policy for the table, you cannot change the policy from LRU to time-based or from time-based to LRU. You must first drop aging and then alter the table to add a new aging policy.

  • The aging policy must be defined to change the aging state.

  • The following rules determine if a row is accessed or referenced for LRU aging:

    • Any rows used to build the result set of a SELECT statement.

    • Any rows used to build the result set of an INSERT ... SELECT statement.

    • Any rows that are about to be updated or deleted.

  • Compiled commands are marked invalid and need recompilation when you either drop LRU aging from or add LRU aging to tables that are referenced in the commands.

  • Call the ttAgingScheduleNow procedure to schedule the aging process right away regardless if the aging state is ON or OFF.

  • For the time-based aging policy, you cannot add or modify the aging column. This is because you cannot add or modify a NOT NULL column.

  • Aging restrictions:

    • You cannot drop the column that is used for time-based aging.

    • Tables that are related by foreign keys must have the same aging policy.

    • For LRU aging, if a child row is not a candidate for aging, neither this child row nor its parent row are deleted. ON DELETE CASCADE settings are ignored.

    • For time-based aging, if a parent row is a candidate for aging, then all child rows are deleted. ON DELETE CASCADE (whether specified or not) is ignored.

  • Restrictions for column-based compression of tables:

    • You can add compressed column groups with the ALTER TABLE statement only if the table was enabled for compression at table creation. You can add uncompressed columns to any table, including tables enabled for compression. Refer to "Column-based compression of tables (TimesTen Classic)" for more details on adding compressed column groups to a table.

    • You cannot modify columns of a compressed column group.

    • You can drop all columns within a compressed column group with the ALTER TABLE command; when removing columns in a compressed column group, all columns in the compressed column group must be specified for removal.

    • You cannot use ALTER TABLE to modify an existing uncompressed column to make it compressed. For example:

      Command> create table mytab (a varchar2 (30), b int, c int) compress ((a,b)
                 by dictionary);
      Command> alter table mytab add (d int) compress (c by dictionary);
      2246: Cannot change compression clause for already defined column C 
      The command failed.

Understanding partitions when using ALTER TABLE in TimesTen

When you create a table, an initial partition is created. If you ALTER the table, and add additional columns, secondary partitions are created. There is one secondary partition created for each ALTER TABLE statement. For a column in secondary partitions, you cannot create a primary key constraint on the column or use the column for time-based aging.

You can use ttMigrate -r -relaxedUpgrade to condense multiple partitions. This means the initial partition plus one or more secondary partitions are condensed into a single partition called the initial partition. Once you condense the partitions, you can then ALTER the table and add a primary key constraint on the column or use the column for time-based aging. This is because the columns are no longer in secondary partitions but are now in the initial partition.

If your database is involved in replication and you want to condense multiple partitions, you must use the StoreAttribute TABLE DEFINITION CHECKING RELAXED (of the CREATE REPLICATION statement). Run ttMigrate -r -relaxedUpgrade on both the master and subscriber or on either the master or subscriber by using -duplicate.

Use ttSchema to view partition numbers for columns. ttSchema displays secondary partition number 1 as partition 1, secondary partition number 2 as partition 2 and so on.

As an example, create a table MyTab with 2 columns. Then ALTER the table adding 2 columns (Col3 and Col4) with the NOT NULL DEFAULT clause.

Command> CREATE TABLE MyTab (Col1 NUMBER, Col2 VARCHAR2 (30));
Command> ALTER TABLE MyTab ADD (Col3 NUMBER NOT NULL DEFAULT 10, Col4 TIMESTAMP
            NOT NULL DEFAULT TIMESTAMP '2012-09-03 12:00:00');

Use ttSchema to verify Col3 and Col4 are in secondary partition 1.

ttschema -DSN sampledb_1122
-- Database is in Oracle type mode
create table TESTUSER.MYTAB (
        COL1 NUMBER,
        COL2 VARCHAR2(30 BYTE) INLINE,
        COL3 NUMBER NOT NULL DEFAULT 10,
        COL4 TIMESTAMP(6) NOT NULL DEFAULT TIMESTAMP '2012-09-03 12:00:00');
-- column COL3 partition 1
-- column COL4 partition 1

Attempt to add a primary key constraint on Col3 and time-based aging on Col4. You see errors because you can neither add a primary key constraint nor add time-based aging to a column that is not in the initial partition.

Command> ALTER TABLE MyTab ADD CONSTRAINT PriKey PRIMARY KEY (Col3);
 2419: All columns in a primary key constraint must be in the initial partition;
column COL3 was added by ALTER TABLE
The command failed.

Command> ALTER TABLE MyTab ADD AGING USE Col4 LIFETIME 3 DAYS;
 3023: Aging column must be in the initial partition; column COL4 was added by
ALTER TABLE
The command failed.

Use ttMigrate with the -relaxedUpgrade option to condense the partitions. Then use ttSchema to verify the partitions are condensed and there are no columns in secondary partition 1.

ttMigrate -c dsn=sampledb_1122 test.migrate
 
Saving user PUBLIC
User successfully saved.
 
Saving table TESTUSER.MYTAB
  Saving rows...
  0/0 rows saved.
Table successfully saved.

ttDestroy sampledb_1122

ttMigrate -r -relaxedUpgrade
 dsn=sampledb_1122 test.migrate
 
Restoring table TESTUSER.MYTAB
  Restoring rows...
  0/0 rows restored.
Table successfully restored.

ttSchema DSN=sampledb_1122
-- Database is in Oracle type mode
create table TESTUSER.MYTAB (
        COL1 NUMBER,
        COL2 VARCHAR2(30 BYTE) INLINE,
        COL3 NUMBER NOT NULL DEFAULT 10,
        COL4 TIMESTAMP(6) NOT NULL DEFAULT TIMESTAMP '2012-09-03 12:00:00');

Now add a primary key constraint on Col3 and time-based aging on Col4. The results are successful because Col3 and Col4 are in the initial partition as a result of ttMigrate. Use ttSchema to verify results.

Command> ALTER TABLE MyTab ADD CONSTRAINT PriKey PRIMARY KEY (Col3);
Command> ALTER TABLE MyTab ADD AGING USE Col4 LIFETIME 3 DAYS;

ttschema sampledb_1122
-- Database is in Oracle type mode
create table TESTUSER.MYTAB (
        COL1 NUMBER,
        COL2 VARCHAR2(30 BYTE) INLINE,
        COL3 NUMBER NOT NULL DEFAULT 10,
        COL4 TIMESTAMP(6) NOT NULL DEFAULT TIMESTAMP '2012-09-03 12:00:00')
    AGING USE COL4 LIFETIME 3 days CYCLE 5 minutes ON;
 
    alter table TESTUSER.MYTAB add constraint PRIKEY primary key (COL3);

Examples for ALTER TABLE: TimesTen Classic

Add returnrate column to parts table.

ALTER TABLE parts ADD COLUMN returnrate DOUBLE;

Add numsssign and prevdept columns to contractor table.

ALTER TABLE contractor
  ADD ( numassign INTEGER, prevdept CHAR(30) );

Remove addr1 and addr2 columns from employee table.

ALTER TABLE employee DROP ( addr1, addr2 );

Drop the UNIQUE title column of the books table.

ALTER TABLE books DROP UNIQUE (title);

Add the x1 column to the t1 table with a default value of 5:

ALTER TABLE t1 ADD (x1 INT DEFAULT 5);

Change the default value of column x1 to 2:

ALTER TABLE t1 MODIFY (x1 DEFAULT 2);

Alter table primarykeytest to add the primary key constraint c1. Use the ttIsql INDEXES command to show that the primary key constraint c1 is created and a range index is used:

Command> CREATE TABLE primarykeytest (col1 TT_INTEGER NOT NULL);
Command> ALTER TABLE primarykeytest ADD CONSTRAINT c1 
         PRIMARY KEY (col1);
Command> INDEXES primarykeytest;

Indexes on table SAMPLEUSER.PRIMARYKEYTEST:
  C1: unique range index on columns:
    COL1
  1 index found.

1 index found on 1 table.

Alter table prikeyhash to add the primary key constraint c2 using a hash index. Use the ttIsql INDEXES command to show that the primary key constraint c2 is created and a hash index is used:

Command> CREATE TABLE prikeyhash (col1 NUMBER (3,2) NOT NULL);
Command> ALTER TABLE prikeyhash ADD CONSTRAINT c2
           PRIMARY KEY (col1) USE HASH INDEX PAGES = 20;
Command> INDEXES prikeyhash;

Indexes on table SAMPLEUSER.PRIKEYHASH:
  C2: unique hash index on columns:
    COL1
  1 index found.

1 table found.

Attempt to add a primary key constraint on a table already defined with a primary key. You see an error:

Command> CREATE TABLE oneprikey (col1 VARCHAR2 (30) NOT NULL, 
         col2 TT_BIGINT NOT NULL, col3 CHAR (15) NOT NULL, 
         PRIMARY KEY (col1,col2));
Command> ALTER TABLE oneprikey ADD CONSTRAINT c2 
         PRIMARY KEY (col1,col2);
 2235: Table can have only one primary key
The command failed.

Attempt to add a primary key constraint on a column that is not defined as NOT NULL. You see an error:

Command> CREATE TABLE prikeynull (col1 CHAR (30));
Command> ALTER TABLE prikeynull ADD CONSTRAINT c3 
         PRIMARY KEY (col1);
 2236: Nullable column cannot be part of a primary key
The command failed.

This example illustrates the use of range and hash indexes. It creates the pkey table with col1 as the primary key. A range index is created by default. The table is then altered to change the index on col1 to a hash index. The table is altered again to change the index back to a range index.

Command> CREATE TABLE pkey (col1 TT_INTEGER PRIMARY KEY, col2 VARCHAR2 (20));
Command> INDEXES pkey;
Indexes on table SAMPLEUSER.PKEY:
   PKEY: unique range index on columns:
    COL1
 1 index found.
1 index found on 1 table.

Alter the pkey table to use a hash index:

Command> ALTER TABLE pkey USE HASH INDEX PAGES = CURRENT;
Command> INDEXES pkey;
Indexes on table SAMPLEUSER.PKEY:
  PKEY: unique hash index on columns:
    COL1
  1 index found.
1 table found.

Alter the pkey table to use a range index with the USE RANGE INDEX clause:

Command> ALTER TABLE pkey USE RANGE INDEX;
Command> INDEXES pkey;
Indexes on table SAMPLEUSER.PKEY:
  PKEY: unique range index on columns:
    COL1
  1 index found.
1 table found.

This example generates an error when attempting to alter a table to define either a range or hash index on a column without a primary key.

Command> CREATE TABLE myindex (Ccl1 CHAR (20));
Command> ALTER TABLE myindex USE RANGE INDEX;
 2810: The table has no primary key so cannot change its index type
The command failed.
Command> ALTER TABLE myindex USE HASH INDEX PAGES = CURRENT;
 2810: The table has no primary key so cannot change its index type
The command failed.

These examples show how time resolution works with aging. In this example, lifetime is three days.

  • If (SYSDATE - ColumnValue) <= 3, do not age out the row.

  • If (SYSDATE - ColumnValue) > 3, then the row is a candidate for aging.

  • If (SYSDATE - ColumnValue) = 3 days, 22 hours, then row is not aged out because lifetime was specified in days. The row would be aged out if lifetime had been specified as 72 hours.

This example alters a table by adding LRU aging. The table has no previous aging policy. The aging state is ON by default.

ALTER TABLE agingdemo3 ADD AGING LRU;
Command> DESCRIBE agingdemo3;
Table USER.AGINGDEMO3:
  Columns:
   *AGINGID                         NUMBER NOT NULL
    NAME                            VARCHAR2 (20) INLINE
  Aging lru on
1 table found.
(primary key columns are indicated with *)

This example alters a table by adding time-based aging. The table has no previous aging policy. The agingcolumn column is used for aging. LIFETIME is 2 days. CYCLE is 30 minutes.

ALTER TABLE agingdemo4
       ADD AGING USE agingcolumn LIFETIME 2 DAYS CYCLE 30 MINUTES;
Command> DESCRIBE agingdemo4;
Table USER.AGINGDEMO4:
  Columns:
   *AGINGID                         NUMBER NOT NULL
    NAME                            VARCHAR2 (20) INLINE
    AGINGCOLUMN                     TIMESTAMP (6) NOT NULL
  Aging use AGINGCOLUMN lifetime 2 days cycle 30 minutes on

This example illustrates that after you create an aging policy, you cannot change it. You must drop aging and redefine.

CREATE TABLE agingdemo5
       (agingid NUMBER NOT NULL PRIMARY KEY
       ,name VARCHAR2 (20)
       ,agingcolumn TIMESTAMP NOT NULL
       )
       AGING USE agingcolumn LIFETIME 3 DAYS OFF;
ALTER TABLE agingdemo5
      ADD AGING LRU;
 2980: Cannot add aging policy to a table with an existing aging policy. Have to
 drop the old aging first
The command failed.

Drop aging on the table and redefine with LRU aging.

ALTER TABLE agingdemo5
      DROP AGING;
ALTER TABLE agingdemo5
       ADD AGING LRU;
Command> DESCRIBE agingdemo5;
Table USER.AGINGDEMO5:
  Columns:
   *AGINGID                         NUMBER NOT NULL
    NAME                            VARCHAR2 (20) INLINE
    AGINGCOLUMN                     TIMESTAMP (6) NOT NULL
  Aging lru on
1 table found.
(primary key columns are indicated with *)

This example alters a table by setting the aging state to OFF. The table has been defined with a time-based aging policy. If you set the aging state to OFF, aging is not done automatically. This is useful to use an external scheduler to control the aging process. Set aging state to OFF and then call the ttAgingScheduleNow procedure to start the aging process.

Command> DESCRIBE agingdemo4;
Table USER.AGINGDEMO4:
  Columns:
   *AGINGID                         NUMBER NOT NULL
    NAME                            VARCHAR2 (20) INLINE
    AGINGCOLUMN                     TIMESTAMP (6) NOT NULL
  Aging use AGINGCOLUMN lifetime 2 days cycle 30 minutes on

ALTER TABLE AgingDemo4
       SET AGING OFF;

Note that when you describe agingdemo4, the aging policy is defined and the aging state is set to OFF.

Command> DESCRIBE agingdemo4;
Table USER.AGINGDEMO4:
  Columns:
   *AGINGID                         NUMBER NOT NULL
    NAME                            VARCHAR2 (20) INLINE
    AGINGCOLUMN                     TIMESTAMP (6) NOT NULL
  Aging use AGINGCOLUMN lifetime 2 days cycle 30 minutes off
1 table found.
(primary key columns are indicated with *)

Call ttAgingScheduleNow to invoke aging with an external scheduler:

Command> CALL ttAgingScheduleNow ('agingdemo4');

Attempt to alter a table adding the aging column and then use that column for time-based aging. An error is generated.

Command> DESCRIBE x;
Table USER1.X:
  Columns:
   *ID                              TT_INTEGER NOT NULL
1 table found.
(primary key columns are indicated with *)
Command> ALTER TABLE x ADD COLUMN t TIMESTAMP;
Command> ALTER TABLE x ADD AGING USE t LIFETIME 2 DAYS;
 2993: Aging column cannot be nullable
The command failed.

Attempt to alter the LIFETIME clause for a table defined with time-based aging. The aging column is defined with data type TT_DATE. An error is generated because the LIFETIME unit is not expressed in DAYS.

Command> CREATE TABLE aging1 (col1 TT_DATE NOT NULL) AGING USE 
         col1 LIFETIME 2 DAYS;
Command> ALTER TABLE aging1 SET AGING LIFETIME 2 HOURS;
 2977: Only DAY lifetime unit is allowed with a TT_DATE column
The command failed.

Alter the employees table to add a new compressed column of state, which contains the full name of the state. Note that the employees table already has a compressed column group consisting of job_id and manager_id.

Command> ALTER TABLE employees 
         ADD COLUMN state VARCHAR2(20) 
         COMPRESS (state BY DICTIONARY);

Command> DESCRIBE employees; 
Table MYSCHEMA.EMPLOYEES:
  Columns:
   *EMPLOYEE_ID                     NUMBER (6) NOT NULL
    FIRST_NAME                      VARCHAR2 (20) INLINE
    LAST_NAME                       VARCHAR2 (25) INLINE NOT NULL
    EMAIL                           VARCHAR2 (25) INLINE NOT NULL
    PHONE_NUMBER                    VARCHAR2 (20) INLINE
    HIRE_DATE                       DATE NOT NULL
    JOB_ID                          VARCHAR2 (10) INLINE NOT NULL
    SALARY                          NUMBER (8,2)
    COMMISSION_PCT                  NUMBER (2,2)
    MANAGER_ID                      NUMBER (6)
    DEPARTMENT_ID                   NUMBER (4)
    STATE                           VARCHAR2 (20) INLINE
  COMPRESS ( ( JOB_ID, MANAGER_ID ) BY DICTIONARY,
             STATE BY DICTIONARY )
 
1 table found.
(primary key columns are indicated with *)

The following example drops the compressed column state from the employees table:

Command> ALTER TABLE employees
 DROP state;
Command> DESCRIBE employees; 
Table MYSCHEMA.EMPLOYEES:
  Columns:
   *EMPLOYEE_ID                     NUMBER (6) NOT NULL
    FIRST_NAME                      VARCHAR2 (20) INLINE
    LAST_NAME                       VARCHAR2 (25) INLINE NOT NULL
    EMAIL                           VARCHAR2 (25) INLINE NOT NULL
    PHONE_NUMBER                    VARCHAR2 (20) INLINE
    HIRE_DATE                       DATE NOT NULL
    JOB_ID                          VARCHAR2 (10) INLINE NOT NULL
    SALARY                          NUMBER (8,2)
    COMMISSION_PCT                  NUMBER (2,2)
    MANAGER_ID                      NUMBER (6)
    DEPARTMENT_ID                   NUMBER (4)
 COMPRESS ( ( JOB_ID, MANAGER_ID ) BY DICTIONARY )
 
1 table found.
(primary key columns are indicated with *)

ALTER USER

The ALTER USER statement enables you to change a user's password. It also enables you to change the profile for the user, to lock or unlock the user's account, and to expire the user's password. A user with the ADMIN privilege can perform these operations.

This statement also enables you to change a user from internal to external or from external to internal.

Required privilege

No privilege is required to change the user's own password.

ADMIN privilege is required for all other operations.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

This is the syntax for ALTER USER...IDENTIFIED BY. Ensure to specify at least one of these clauses: IDENTIFIED BY, PROFILE, ACCOUNT, or PASSWORD EXPIRE.

ALTER USER user [IDENTIFIED BY {password | "password"}]
  [PROFILE profile] [ACCOUNT {LOCK|UNLOCK}] [PASSWORD EXPIRE]

This is the syntax for ALTER USER...IDENTIFIED EXTERNALLY. Ensure to specify at least one of these clauses: IDENTIFIED EXTERNALLY, PROFILE, or ACCOUNT.

ALTER USER user [IDENTIFIED EXTERNALLY]
  [PROFILE profile] [ACCOUNT {LOCK|UNLOCK}]

Parameters

Parameter Description

user

Name of the user to alter.

IDENTIFIED BY password|"password"

Specifies an internal user and the password for the internal user.

The password you can specify is dependent on the profile assigned to the user. Specifically, the value of the PASSWORD_COMPLEXITY_CHECKER password parameter determines the complexity of the password. If the value is TT_VERIFY_FUNCTION, TT_STRONG_VERIFY_FUNCTION, or TT_STIG_VERIFY_FUNCTION, the password must meet specific password verification requirements. For example, if the value is TT_VERIFY_FUNCTION, the password cannot contain the name of the database. See "About password complexity checker verification" for details.

IDENTIFIED EXTERNALLY

Specifies the user is an external user.

PROFILE profile

Use the PROFILE clause to specify the name of the profile (designated by profile) that you want to assign to the user. The profile sets the limits for the password parameters for the user. See "CREATE PROFILE" for information on these password parameters. You can specify a PROFILE clause for external users, but the password parameters have no effect for these users.

ACCOUNT [LOCK|UNLOCK]

Specify ACCOUNT LOCK to lock the user's account and disable connections to the database. Specify ACCOUNT UNLOCK to unlock the user's account and enable connections to the database. The default is ACCOUNT UNLOCK.

PASSWORD EXPIRE

Specify PASSWORD EXPIRE if you want the user's password to expire. This setting forces a user with ADMIN privileges to change the password before the user can connect to the database. This clause is not valid for an externally identified user (as denoted by the IDENTIFIED EXTERNALLY clause).

Description

  • Database users can be internal or external.

    • Internal users are defined for a TimesTen database.

    • External users are defined by the operating system. External users cannot be assigned a TimesTen password.

  • Password requirements:
    • Cannot exceed 30 characters.
    • Is case-sensitive.
    • Must start with a letter. A password cannot start with a digit or a special character unless the password is enclosed in double quotation marks.
    • If a special character is used, the password must be contained in double quotation marks. The exceptions are the # and the @ special characters. A password that contains the # or the @ special character does not need to be enclosed in double quotation marks.
    • Cannot contain a semi-colon (;) or a double quotation mark (").
  • Use the PROFILE clause to change the profile for a user. See "CREATE PROFILE" for details.

  • Use the ACCOUNT LOCK or ACCOUNT UNLOCK to change the lock settings for the user account.

  • Use the PASSWORD EXPIRE clause to expire the user's password and force a password change before the user can connect to the database.

  • You can alter a user over a client/sever connection if the connection is encrypted with TLS. See "Transport Layer Security for TimesTen Client/Server" in the Oracle TimesTen In-Memory Database Security Guide for details.

  • When replication is configured, this statement is replicated.

Examples

Illustrate password verification when altering user

This example creates the myprofile_strongpw profile and specifies a value of TT_STRONG_VERIFY_FUNCTION for the PASSWORD_COMPLEXITY_CHECKER password parameter. The example then creates the sampleuser_pwchange user and assigns the myprofile_strongpw profile to this user. The specified password meets the requirements of the TT_STRONG_VERIFY_FUNCTION function and the user is created. See "TT_STRONG_VERIFY_FUNCTION" for more information on the TT_STRONG_VERIFY_FUNCTION function.

Command> CREATE PROFILE myprofile_strongpw LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_STRONG_VERIFY_FUNCTION;

Profile created.

Command> CREATE USER sampleuser_pwchange 
           IDENTIFIED BY "5&AbbN*60" PROFILE myprofile_strongpw;

User created.

Now alter the myprofile_strongpw profile, changing the value of the PASSWORD_COMPLEXITY_CHECKER password parameter to TT_STIG_VERIFY_FUNCTION. Use the ALTER USER statement to expire the password for the sampleuser_pwchange user. Attempt to connect to the database as the sampleuser_pwchange user. The connection fails, as the password is expired.

Command> ALTER PROFILE myprofile_strongpw LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_STIG_VERIFY_FUNCTION;

Profile altered.

Command> ALTER USER sampleuser_pwchange PASSWORD EXPIRE;

User altered.

Command> GRANT CONNECT TO sampleuser_pwchange;
Command> connect adding "UID=sampleuser_pwchange;PWD=5&AbbN*60" as sampleuser;
15180: the password has expired
The command failed.

Use the ALTER USER statement to change the password for the sampleuser_pwchange user. The ALTER USER statement succeeds, as the password meets the requirements of the TT_STIG_VERIFY_FUNCTION function. Attempt to connect to the database as the sampleuser_pwchange user. The connection is successful. See "TT_STIG_VERIFY_FUNCTION" for more information on the TT_STIG_VERIFY_FUNCTION function.

access1: Command> ALTER USER sampleuser_pwchange 
           IDENTIFIED BY "bd@<!BCvvKASn67";

User altered.

Command> connect adding "UID=sampleuser_pwchange;PWD=bd@<!BCvvKASn67" 
           as sampleuser;
Connection successful: DSN=access1;UID=sampleuser_pwchange;
DataStore=/scratch/sampleuser/mydatabase1;DatabaseCharacterSet=AL32UTF8;
ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Change the user's profile

This example creates the user1 user and assigns the user1 user the profile1 profile. The example then uses the ALTER USER statement to change the user1 user's profile to profile2.

Command> CREATE USER user1 IDENTIFIED BY user1 PROFILE profile1;
 
User created.
 
Command> ALTER USER user1 PROFILE profile2;
 
User altered.

Query the dba_users system view to verify the user1 profile has been changed to profile2.

Command> SELECT profile FROM dba_users WHERE username = 'USER1';
< PROFILE2 >
1 row found.

Lock and unlock a user's account

This example creates the user2 user. It then uses the ALTER USER statement to lock and then unlock the user2 user's account.

Command> CREATE USER user2 IDENTIFIED BY user2 PROFILE profile1;
 
User created.
 
Command> ALTER USER user2 ACCOUNT LOCK;
 
User altered.
 

Grant the CONNECT privilege to user2;

Command> GRANT CONNECT TO user2;

Attempt to connect to the database as user2. The user2 account is locked so the connection fails.

Command> connect adding "UID=user2;PWD=user2" as user2;
15179: the account is locked
The command failed.

As the instance administrator, reconnect to the database and use the ALTER USER statement to unlock the user2 account.

none: Command> use database1
database1: Command> ALTER USER user2 ACCOUNT UNLOCK;
 
User altered.
 

Attempt to connect to the database as the user2 user. The connection succeeds.

database1: Command> connect adding "UID=user2;PWD=user2" as user2;
Connection successful: DSN=database1;UID=user3;DataStore=/scratch/database1;
DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Expire a user's password

This example uses the ALTER USER statement to change the user2 user's account to expire the password. A user with ADMIN privilege must change the user2 password before user2 can connect to the database.

Command> ALTER USER user2 PASSWORD EXPIRE;
 
User altered.
 

Attempt to connect to the database as user2. The user2 password must be changed before the user2 user can connect to the database.

Command>  connect adding "UID=user2;PWD=user2" as user2;
15180: the password has expired
The command failed.

As the instance administrator, reconnect to the database and use the ALTER USER statement to change the user2 password.

none: Command> use database1
database1: Command>  ALTER USER user2 IDENTIFIED BY newuser2password;
 
User altered.
 

Attempt to connect to the database a the user2 user. The connection succeeds.

database1: Command> connect adding "UID=user2;PWD=newuser2password" as user2;
Connection successful: DSN=database1;UID=user4;DataStore=/scratch/database1;
DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Change a user from external to internal and internal to external

This example uses the ALTER USER statement to change the user2 internal user to an external user and then back to an internal user.

Command> ALTER USER user2 IDENTIFIED EXTERNALLY;
 
User altered.

Use the ALTER USER statement to change the user2 external user back to an internal user.

Command> ALTER USER user2 IDENTIFIED BY user2_password_change;
 
User altered.

CALL

Use the CALL statement to execute a TimesTen built-in procedure or to execute a PL/SQL procedure or function that is standalone or part of a package from within SQL.

Required privilege

The privileges required for executing each TimesTen built-in procedure are listed in the description of each procedure in the "Built-In Procedures" section in the Oracle TimesTen In-Memory Database Reference.

No privileges are required for an owner calling its own PL/SQL procedure or function that is standalone or part of a package using the CALL statement. For all other users, the EXECUTE privilege on the procedure or function or on the package in which it is defined is required.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

To call a TimesTen built-in procedure:

CALL TimesTenBuiltIn [( arguments )]

When calling PL/SQL procedures or functions that are standalone or part of a package, you can either call these by name or as the result of an expression.

To call a PL/SQL procedure:

CALL [Owner.][Package.]ProcedureName [( arguments )]

To call a PL/SQL function that returns a parameter, one of the following are appropriate:

CALL [Owner.][Package.]FunctionName [( arguments )] INTO :return_param

Note:

A user's own PL/SQL procedure or function takes precedence over a TimesTen built-in procedure with the same name.

Parameters

Parameter Description

TimesTenBuiltIn

Name of the TimesTen built-in procedure. For a full list of TimesTen built-in procedures, see "Built-In Procedures" in the Oracle TimesTen In-Memory Database Reference.

[Owner.]ProcedureName

Name of the PL/SQL procedure. You can optionally specify the owner of the procedure.

[Owner.]FunctionName

Name of the PL/SQL function. You can optionally specify the owner of the function.

arguments

Specify 0 or more arguments for the PL/SQL procedure or function.

INTO

If the routine is a function, the INTO clause is required.

return_param

Specify the host variable that stores the return value of the function.

Description

Detailed information on how to execute PL/SQL procedures or functions with the CALL statement in TimesTen is provided in "Executing procedures and functions" in the Oracle TimesTen In-Memory Database PL/SQL Developer's Guide, "Using CALL to execute procedures and functions" in the Oracle TimesTen In-Memory Database C Developer's Guide, or "Using CALL to execute procedures and functions" in the Oracle TimesTen In-Memory Database Java Developer's Guide.

Examples

The following is the definition of the mytest function:

create or replace function mytest return number is
begin
  return 1;
end;
/

Perform the following to execute the mytest function in a CALL statement:

Command> variable n number;
Command> call mytest() into :n;
Command> print n;
N                    : 1

The following example creates a function that returns the salary of the employee whose employee ID is specified as input, then calls the function and displays the result that was returned.

Command> CREATE OR REPLACE FUNCTION get_sal
          (p_id employees.employee_id%TYPE) RETURN NUMBER IS
            v_sal employees.salary%TYPE := 0;
         BEGIN
           SELECT salary INTO v_sal FROM employees
             WHERE employee_id = p_id;
           RETURN v_sal;
         END get_sal;
         /
 
Function created.
 
Command> variable n number;
Command> call get_sal(100) into :n;
Command> print n;
N                    : 24000

COMMIT

The COMMIT statement ends the current transaction and makes permanent all changes performed in the transaction.

Required privilege

None

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

COMMIT [WORK]

Parameters

The COMMIT statement enables the following optional keyword:

Parameter Description

[WORK]

Optional clause supported for compliance with the SQL standard. COMMIT and COMMIT WORK are equivalent.

Description

  • Until you commit a transaction:

    • You can see any changes you have made during the transaction but other users cannot see the changes. After you commit the transaction, the changes are visible to other users' statements that execute after the commit.

    • You can roll back (undo) changes made during the transaction with the ROLLBACK statement.

  • This statement releases transaction locks.

  • For passthrough, the Oracle Database transaction will also be committed.

  • A commit closes all open cursors.

Examples

Insert a row into regions table of the HR schema and commit transaction. First set autocommit to 0:

Command> SET AUTOCOMMIT 0;
Command> INSERT INTO regions VALUES (5,'Australia');
1 row inserted.
Command> COMMIT;
Command> SELECT * FROM regions;
< 1, Europe >
< 2, Americas >
< 3, Asia >
< 4, Middle East and Africa >
< 5, Australia >
5 rows found.

See also

ROLLBACK

CREATE ACTIVE STANDBY PAIR

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

This statement creates an active standby pair. It includes an active master database, a standby master database, and may also include one or more read-only subscribers. The active master database replicates updates to the standby master database, which propagates the updates to the subscribers.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

CREATE ACTIVE STANDBY PAIR
  FullStoreName, FullStoreName [ReturnServiceAttribute]
    [SUBSCRIBER FullStoreName [,...]]
    [STORE FullStoreName [StoreAttribute [...]]]
    [NetworkOperation [...] ]
    [{ INCLUDE | EXCLUDE }{TABLE [[Owner.]TableName [,...]]|
         CACHE GROUP [[Owner.]CacheGroupName [,...]]|
         SEQUENCE [[Owner.]SequenceName [,...]]} [,...]]

Syntax for ReturnServiceAttribute:

{ RETURN RECEIPT [BY REQUEST] |
  RETURN TWOSAFE [BY REQUEST] |
  NO RETURN }

Syntax for StoreAttribute:

  DISABLE RETURN {SUBSCRIBER | ALL} NumFailures 
  RETURN SERVICES {ON | OFF} WHEN [REPLICATION] STOPPED 
  DURABLE COMMIT {ON | OFF}
  RESUME RETURN Milliseconds 
  LOCAL COMMIT ACTION {NO ACTION | COMMIT} 
  RETURN WAIT TIME Seconds 
  COMPRESS TRAFFIC {ON | OFF}
  PORT PortNumber 
  TIMEOUT Seconds 
  FAILTHRESHOLD Value 
  TABLE DEFINITION CHECKING {RELAXED|EXACT}

Syntax for NetworkOperation:

ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName
  { { MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost }
      PRIORITY Priority } [...]

Parameters

Parameter Description

FullStoreName

The database, specified as one of the following:

  • SELF

  • The prefix of the database file name

For example, if the database path is directory/subdirectory/data.ds0, then data is the database name that should be used.

This is the database file name specified in the DataStore attribute of the DSN description with optional host ID in the form:

DataStoreName [ON Host]

Host can be either an IP address or a literal host name assigned to one or more IP addresses, as described in "Configuring the network" in Oracle TimesTen In-Memory Database Replication Guide. Host names containing special characters must be surrounded by double quotes. For example: "MyHost-500".

RETURN RECEIPT [BY REQUEST]

Enables the return receipt service, so that applications that commit a transaction to an active master database are blocked until the transaction is received by the standby master database.

Specifying RETURN RECEIPT applies the service to all transactions. If you specify RETURN RECEIPT BY REQUEST, you can use the ttRepSyncSet procedure to enable the return receipt service for selected transactions. For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

RETURN TWOSAFE [BY REQUEST]

Enables the return twosafe service, so that applications that commit a transaction to an active master database are blocked until the transaction is committed on the standby master database.

Specifying RETURN TWOSAFE applies the service to all transactions. If you specify RETURN TWOSAFE BY REQUEST, you can use the ttRepSyncSet procedure to enable the return receipt service for selected transactions.

For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

DISABLE RETURN {SUBSCRIBER | ALL} NumFailures

Set the return service failure policy so that return service blocking is disabled after the number of timeouts specified by NumFailures.

Specifying SUBSCRIBER is the same as specifying ALL. Both settings refer to the standby master database.

This failure policy can be specified for either the RETURN RECEIPT or RETURN TWOSAFE service.

RETURN SERVICES {ON | OFF} WHEN [REPLICATION] STOPPED

Sets return services on or off when replication is disabled (stopped or paused state).

OFF disables return services when replication is disabled and is the default for RETURN RECEIPT service. ON allows return services to continue to be enabled when replication is disabled and is the default for RETURN TWOSAFE service.

See "Establishing return service failure/recovery policies" in Oracle TimesTen In-Memory Database Replication Guide.

RESUME RETURN Milliseconds

If DISABLE RETURN has disabled return service blocking, this attribute sets the policy for when to re-enable the return service.

NO RETURN

Specifies that no return service is to be used. This is the default.

For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

RETURN WAIT TIME Seconds

Specifies the number of seconds to wait for return service acknowledgment. A value of 0 (zero) means that there is no waiting. The default value is 10 seconds.

The application can override this timeout setting by using the returnWait parameter in the ttRepSyncSet built-in procedure.

SUBSCRIBER FullStoreName [,...]]

A database that receives updates from a master database. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

STORE FullStoreName [StoreAttribute [...]]

Defines the attributes for the specified database. Attributes include PORT, TIMEOUT and FAILTHRESHOLD. FullStoreName is the database file name specified in the DataStore attribute of the DSN description.

TABLE DEFINITION CHECKING {EXACT|RELAXED}

StoreAttribute clause.

Specifies type of table definition checking that occurs on the subscriber:

  • EXACT - The tables must be identical on master and subscriber.

  • RELAXED - The tables must have the same key definition, number of columns and column data types.

The default is RELAXED.

Note: If you use TABLE DEFINITION CHECKING EXACT, use ttMigrate -exactUpgrade if you migrate the database. If you use TABLE DEFINITION CHECKING RELAXED, use ttMigrate -relaxedUpgrade if you migrate the database.

{INCLUDE | EXCLUDE}

{TABLE [[Owner.]TableName[,...]]|

CACHE GROUP

[[Owner.]CacheGroupName

[,...]]|

SEQUENCE

[[Owner.]SequenceName

[,...]]}

[,...]

An active standby pair replicates an entire database by default.

INCLUDE includes only the listed tables, sequences or cache groups for the replication scheme. Use one INCLUDE clause for each object type (table, sequence or cache group).

EXCLUDE removes tables or sequences or cache groups from the replication scheme. Use one EXCLUDE clause for each object type (table, sequence or cache group).

Do not use the EXCLUDE clause for AWT cache groups.

COMPRESS TRAFFIC {ON | OFF}

Compress replicated traffic to reduce the amount of network bandwidth. ON specifies that all replicated traffic for the database defined by STORE be compressed. OFF (the default) specifies no compression. See "Compressing replicated traffic" in Oracle TimesTen In-Memory Database Replication Guide for details.

DURABLE COMMIT {ON | OFF}

Overrides the DurableCommits general connection attribute setting. DURABLE COMMIT ON enables durable commits regardless of whether the replication agent is running or stopped. It also enables durable commits when the ttRepStateSave built-in procedure has marked the standby database as failed.

FAILTHRESHOLD Value

The number of log files that can accumulate for a subscriber database. If this value is exceeded, the subscriber is set to the Failed state.The value 0 means "No Limit." This is the default.

See "Setting the transaction log failure threshold" in Oracle TimesTen In-Memory Database Replication Guide for more information.

LOCAL COMMIT ACTION {NO ACTION | COMMIT}

Specifies the default action to be taken for a return twosafe transaction in the event of a timeout.

Note: This attribute is valid only when the RETURN TWOSAFE or RETURN TWOSAFE BY REQUEST attribute is set in the SUBSCRIBER clause.

NO ACTION: On timeout, the commit function returns to the application, leaving the transaction in the same state it was in when it entered the commit call, with the exception that the application is not able to update any replicated tables. The application can only reissue the commit. The transaction may not be rolled back. This is the default.

COMMIT: On timeout, the commit function attempts to perform a COMMIT to end the transaction locally. No more operations are possible on the same transaction.

This setting can be overridden for specific transactions by calling the localAction parameter in the ttRepSyncSet procedure.

MASTER FullStoreName

The database on which applications update the specified element. The MASTER database sends updates to its SUBSCRIBER databases. The FullStoreName must be the database specified in the DataStore attribute of the DSN description.

PORT PortNumber

The TCP/IP port number on which the replication agent for the database listens for connections. If not specified, the replication agent automatically allocates a port number.

In an active standby pair, the standby master database listens for updates from the active master database. Read-only subscribers listen for updates from the standby master database.

ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName

Denotes the NetworkOperation clause. If specified, enables you to control the network interface that a master store uses for every outbound connection to each of its subscriber stores. In the context of the ROUTE clause, you can define the following:

  • A route for the active database to the standby database and for the standby database to the active database for when failover occurs

  • A route for a read-only subscriber to the active and standby databases

When using active standby pairs, ROUTE should be specified at least twice for an active standby pair with no read only subscribers. Then, ROUTE should be specified twice more for each read only subscriber on the active standby pair.

For FullStoreName, ON "host" must be specified.

MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost

MasterHost and SubscriberHost are the IP addresses for the network interface on the master and subscriber stores. Specify in dot notation or canonical format or in colon notation for IPV6.

Clause can be specified more than once.

PRIORITY Priority

Variable expressed as an integer from 1 to 99. Denotes the priority of the IP address. Lower integral values have higher priority. An error is returned if multiple addresses with the same priority are specified. Controls the order in which multiple IP addresses are used to establish peer connections.

Required syntax of NetworkOperation clause. Follows MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost clause.

TIMEOUT Seconds

The maximum number of seconds the replication agent waits for a response from remote replication agents. The default is 120 seconds.

In an active standby pair, the active master database sends messages to the standby master database. The standby master database sends messages to the read-only subscribers.

Note: For large transactions that may cause a delayed response from the remote replication agent, the agent scales the timeout based on the size of the transaction. This scaling is disabled if you set TIMEOUT to less than or equal to 60 seconds. Also see "Setting wait timeout for response from remote replication agents" in Oracle TimesTen In-Memory Database Replication Guide.

Description

  • After you create an active standby pair, make one of your databases the active database. To accomplish this, call ttRepStateSet ('ACTIVE'). Then use ttRepAdmin to duplicate the active database to the second database. When the operation is successful, the second database becomes the standby database. For more information, see "Setting up an active standby pair with no cache groups" in Oracle TimesTen In-Memory Database Replication Guide.

  • The SUBSCRIBER clause lists one or more read-only subscriber databases. You can designate up to 127 subscriber databases.

  • Replication between the active master database and the standby master database can be RETURN TWOSAFE, RETURN RECEIPT, or asynchronous. RETURN TWOSAFE ensures no transaction loss.

  • Use the INCLUDE and EXCLUDE clauses to exclude the listed tables, sequences and cache groups from replication, or to include only the listed tables, sequences and cache groups, excluding all others.

  • If the active standby pair has the RETURN TWOSAFE attribute and replicates a cache group, a transaction may fail if:

    • The transaction that is being replicated contains an ALTER TABLE statement or an ALTER CACHE GROUP statement.

    • The transaction contains an INSERT, UPDATE or DELETE statement on a replicated table, replicated cache group or an asynchronous writethrough cache group.

  • You can use an active standby pair to replicate read-only cache groups and asynchronous writethrough (AWT) cache groups. You cannot use an active standby pair to replicate synchronous writethrough (SWT) cache groups or user managed cache groups.

  • You cannot use the EXCLUDE clause for AWT cache groups.

  • You cannot execute the CREATE ACTIVE STANDBY PAIR statement when Oracle Clusterware is used with TimesTen.

Examples

This example creates an active standby pair whose master databases are rep1 and rep2. There is one subscriber, rep3. The type of replication is RETURN RECEIPT. The statement also sets PORT and TIMEOUT attributes for the master databases.

CREATE ACTIVE STANDBY PAIR rep1, rep2 RETURN RECEIPT
  SUBSCRIBER rep3
  STORE rep1 PORT 21000 TIMEOUT 30
  STORE rep2 PORT 22000 TIMEOUT 30;

Specify NetworkOperation clause to control network interface:

CREATE ACTIVE STANDBY PAIR rep1,rep2
ROUTE MASTER rep1 ON "machine1" SUBSCRIBER rep2 ON "machine2"
MASTERIP "1.1.1.1" PRIORITY 1 SUBSCRIBERIP "2.2.2.2" PRIORITY 1;
ROUTE MASTER rep2 ON "machine2" SUBSCRIBER rep1 ON "machine1"
MASTERIP "2.2.2.2" PRIORITY 1 SUBSCRIBERIP "1.1.1.1" PRIORITY 1;

CREATE CACHE GROUP

The CREATE CACHE GROUP statement:

  • Creates the table defined by the cache group.

  • Loads all new information associated with the cache group in the appropriate system tables.

A cache group is a set of tables related through foreign keys that cache data from tables in an Oracle database. There is one root table that does not reference any of the other tables. All other cache tables in the cache group reference exactly one other table in the cache group. In other words, the foreign key relationships form a tree.

A cache table is a set of rows satisfying the conditions:

  • The rows constitute a subset of the rows of a vertical partition of an Oracle database table.

  • The rows are stored in a TimesTen table with the same name as the Oracle database table.

If a database has more than one cache group, the cache groups must correspond to different Oracle database (and TimesTen) tables.

Cache group instance refers to a row in the root table and all the child table rows related directly or indirectly to the root table rows.

A cache group can be either system managed or user managed.

A system managed cache group is fully managed by TimesTen and has fixed properties. System managed cache group types include:

  • Read-only cache groups are updated in the Oracle database, and the updates are propagated from the Oracle database to the cache.

  • Asynchronous writethrough (AWT) cache groups are updated in the cache and the updates are propagated to the Oracle database. Transactions continue executing on the cache without waiting for a commit on the Oracle database.

  • Synchronous writethrough (SWT) cache groups are updated in the cache and the updates are propagated to the Oracle database. Transactions are committed on the cache after notification that a commit has occurred on the Oracle database.

Because TimesTen manages system managed cache groups, including loading and unloading the cache group, certain statements and clauses cannot be used in the definition of these cache groups, including:

  • WHERE clauses in AWT and SWT cache group definitions

  • READONLY, PROPAGATE and NOT PROPAGATE in cache table definitions

  • AUTOREFRESH in AWT and SWT cache group definitions

The FLUSH CACHE GROUP and REFRESH CACHE GROUP operations are not allowed for AWT and SWT cache groups.

You must stop the replication agent before creating an AWT cache group.

A user managed cache group must be managed by the application or user. PROPAGATE in a user managed cache group is synchronous. The table-level READONLY keyword can only be used for user managed cache groups.

In addition, both TimesTen and Oracle Database must be able to parse all WHERE clauses.

Cache groups can be explicitly or dynamically loaded.

In cache groups that are explicitly loaded, new cache instances are loaded manually into the TimesTen cache tables from the Oracle database tables using a LOAD CACHE GROUP or REFRESH CACHE GROUP statement or automatically using an autorefresh operation.

In a dynamic cache group, new cache instances can be loaded manually into the TimesTen cache tables by using a LOAD CACHE GROUP or on demand using a dynamic load operation. In a dynamic load operation, data is automatically loaded into the TimesTen cache tables from the cached Oracle database tables when a SELECT, UPDATE, DELETE or INSERT statement is issued on one of the cache tables, where the data is not present in the cache table but does exist in the cached Oracle database table. A manual refresh or automatic refresh operation on a dynamic cache group can result in the updating or deleting of existing cache instances, but not in the loading of new cache instances.

Any cache group type (read-only, asynchronous writethrough, synchronous writethrough, user managed) can be defined as an explicitly loaded cache group.

Any cache group type can be defined as a dynamic cache group except a user managed cache group that has both the AUTOREFRESH cache group attribute and the PROPAGATE cache table attribute.

Data in a dynamic cache group is aged out because LRU aging is defined by default. Use the ttAgingLRUConfig and/or the ttAgingTableLRUConfig built-in procedures to override the space usage thresholds for LRU aging. You can also define time-based aging on a dynamic cache group to override LRU aging.

For more information on static and dynamic cache groups, see "Cache groups and cache tables" in Oracle TimesTen In-Memory Database Cache Guide.

Required privilege

CREATE CACHE GROUP or CREATE ANY CACHE GROUP and CREATE TABLE (if all tables in the cache group are owned by the current user) or CREATE ANY TABLE (if at least one of the tables in the cache group is not owned by the current user).

Usage with TimesTen Scaleout

Static read-only cache groups with incremental autorefresh are supported.

SQL syntax: TimesTen Scaleout

For static read-only cache groups in TimesTen Scaleout:
CREATE READONLY CACHE GROUP [Owner.]GroupName
 [AUTOREFRESH 
    [MODE INCREMENTAL]
    [INTERVAL IntervalValue {MINUTE[S] | SECOND[S] | MILLISECOND[S]}]
    [STATE {ON|OFF|PAUSED}]
  ]
  FROM
    [Owner.]TableName (ColumnDefinition[,...][,PRIMARY KEY(ColumnName[,...])])
  [UNIQUE HASH ON (HashColumnName[,...]) PAGES=PrimaryPages]
  [ParentDistributionClause]
    [WHERE ExternalSearchCondition]
  [,[Owner.]TableName (ColumnDefinition[,...]
    [,PRIMARY KEY(ColumnName[,...])]
    [,FOREIGN KEY(ColumnName[,...])
        REFERENCES RefTableName (ColumnName [,...])[ON DELETE CASCADE]])
 [UNIQUE HASH ON (HashColumnName[,...]) PAGES=PrimaryPages]
 [ChildDistributionClause]
   [WHERE ExternalSearchCondition]
 [,...]
 ]

The syntax for the distribution clause for a parent:
ParentDistributionClause::= DISTRIBUTE BY HASH [(ColumnName [,...])] | DUPLICATE

The syntax for the distribution clause for a child:
ChildDistributionClause::= DISTRIBUTE BY HASH [(ColumnName [,...])] |
 DISTRIBUTE BY REFERENCE [(ForeignKeyConstraint)] | DUPLICATE

SQL syntax: TimesTen Classic

For read-only cache groups:

CREATE [DYNAMIC] [HYBRID] READONLY CACHE GROUP [Owner.]GroupName
 [AUTOREFRESH
  [MODE {INCREMENTAL | FULL}]
  [INTERVAL IntervalValue {MINUTE[S] | SECOND[S] | MILLISECOND[S] }]
  [STATE {ON|OFF|PAUSED}]
 ]
 FROM
  [Owner.]TableName (
    {ColumnDefinition[,...]}
    [,PRIMARY KEY(ColumnName[,...])]
    [,FOREIGN KEY(ColumnName [,...])
            REFERENCES RefTableName (ColumnName [,...])]
                    [ON DELETE CASCADE])
 [UNIQUE HASH ON (HashColumnName[,...]) PAGES=PrimaryPages]
 [AGING {LRU|
         USE ColumnName
         LIFETIME Num1 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}
         [CYCLE Num2 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}]
        }[ON|OFF]
 ]
 [WHERE ExternalSearchCondition]
} [,...]

For asynchronous writethrough cache groups:

CREATE [DYNAMIC] [ASYNCHRONOUS] WRITETHROUGH CACHE GROUP   [Owner.]GroupName
  FROM
   {[Owner.]TableName (
     {ColumnDefinition[,...]}
    [,PRIMARY KEY(ColumnName[,...])]
     [FOREIGN KEY(ColumnName [,...])
          REFERENCES RefTableName (ColumnName [,...])]
      [ON DELETE CASCADE])
 UNIQUE HASH ON (HashColumnName[,...]) PAGES=PrimaryPages]
 [AGING {LRU|
     USE ColumnName
          LIFETIME Num1 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}
         [CYCLE Num2 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}]
     }[ON|OFF]
 ]
} [,...]

For synchronous writethrough cache groups:

CREATE [DYNAMIC] SYNCHRONOUS WRITETHROUGH
CACHE GROUP [Owner.]GroupName
 FROM 
   {[Owner.]TableName (
     {ColumnDefinition[,...]}
    [,PRIMARY KEY(ColumnName[,...])]
     [FOREIGN KEY(ColumnName [,...])
            REFERENCES RefTableName (ColumnName [,...])]
   [ON DELETE CASCADE])
 [UNIQUE HASH ON (HashColumnName[,...]) PAGES=PrimaryPages]
 [AGING {LRU|
     USE ColumnName
         LIFETIME Num1 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}
         [CYCLE Num2 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}]
     }[ON|OFF]
 ]
} [,...]

For user managed cache groups:

CREATE [DYNAMIC][USERMANAGED] CACHE GROUP [Owner.]GroupName
 [AUTOREFRESH
   [MODE {INCREMENTAL | FULL}]
   [INTERVAL IntervalValue {MINUTE[S] | SECOND[S] | MILLISECOND[S] }]
   [STATE {ON|OFF|PAUSED}]
 ]
  FROM 
   {[Owner.]TableName (
    {ColumnDefinition[,...]}
    [,PRIMARY KEY(ColumnName[,...])]
    [FOREIGN KEY(ColumnName[,...])
          REFERENCES RefTableName (ColumnName [,...])]
    [ON DELETE CASCADE]
    [,{READONLY | PROPAGATE | NOT PROPAGATE}])
 [UNIQUE HASH ON (HashColumnName[,...]) PAGES=PrimaryPages]
 [AGING {LRU|
         USE ColumnName
             LIFETIME Num1 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}
             [CYCLE Num2 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}]
        }[ON|OFF]
 ]
 [WHERE ExternalSearchCondition]
} [,...]

Parameters

Following are the parameters for the cache group definition before the FROM keyword:

Parameter Description

[Owner.]GroupName

Owner and name assigned to the new cache group.

DYNAMIC

Supported in TimesTen Classic only.

If specified, a dynamic cache group is created.

HYBRID

Supported in TimesTen Classic only.

If specified, a dynamic read-only cache group where the root table does not exist in the Oracle database.

AUTOREFRESH

The AUTOREFRESH parameter automatically propagates changes from the Oracle database to the cache group.

MODE [INCREMENTAL | FULL]

Determines which rows in the cache are updated during an autorefresh. If the INCREMENTAL clause is specified, TimesTen refreshes only rows that have been changed on the Oracle database since the last propagation. If the FULL clause is specified, TimesTen updates all rows in the cache with each autorefresh. The default autorefresh mode is INCREMENTAL.

In TimesTen Scaleout, MODE INCREMENTAL is supported.

INTERVAL IntervalValue

Indicates the interval at which autorefresh should occur in units of minutes, seconds or milliseconds. IntervalValue is an integer value that specifies how often autorefresh should be scheduled, in minutes, seconds, or milliseconds. The default IntervalValue value is 5 minutes. An autorefresh interval set to 0 milliseconds enables continuous autorefresh, where the next autorefresh cycle is scheduled immediately after the last autorefresh cycle has ended. See Automatically refreshing a cache group in the Oracle TimesTen In-Memory Database Cache Guide for more information.

If the specified interval is not long enough for an autorefresh to complete, a runtime warning is generated and the next autorefresh waits until the current one finishes. An informational message is generated in the support log if the wait queue reaches 10.

STATE [ON | OFF | PAUSED]

Specifies whether autorefresh should be ON or OFF or PAUSED when the cache group is created. You can alter this setting later by using the ALTER CACHE GROUP statement. By default, the AUTOREFRESH state is PAUSED.

FROM

Designates one or more table definitions for the cache group.

Everything after the FROM keyword comprises the definitions of the Oracle database tables cached in the cache group. The syntax for each table definition is similar to that of a CREATE TABLE statement. However, primary key constraints are required for the cache group table.

Table definitions have the following parameters.

Parameter Description

[Owner.]TableName

Owner and name to be assigned to the new table. If you do not specify the owner name, your login becomes the owner name for the new table.

ColumnDefinition

Name of an individual column in a table, its data type and whether it is nullable. Each table must have at least one column.

PRIMARY KEY (ColumnName[,...])

Specifies that the table has a primary key. Primary key constraints are required for a cache group. ColumnName is the name of the column that forms the primary key for the table to be created. Up to 16 columns can be specified for the primary key. Cannot be specified with UNIQUE in one specification.

FOREIGN KEY (ColumnName[,...])

Specifies that the table has a foreign key. ColumnName is the name of the column that forms the foreign key for the table to be created.

REFERENCES RefTableName (ColumnName[,...])

Specifies the table which the foreign key is associated with. RefTableName is the name of the referenced table and ColumnName is the name of the column referenced in the table.

[ON DELETE CASCADE]

Enables the ON DELETE CASCADE referential action. If specified, when rows containing referenced key values are deleted from a parent table, rows in child tables with dependent foreign key values are also deleted.

READONLY

Specifies that changes cannot be made on the cached table.

PROPAGATE|NOT PROPAGATE

Supported in TimesTen Classic only.

Specifies whether changes to the cached table are automatically propagate to the corresponding Oracle database table at commit time.

UNIQUE HASH ON (HashColumnName)

Specifies that a hash index is created on this table. HashColumnName identifies the column that is to participate in the hash key of this table. The columns specified in the hash index must be identical to the columns in the primary key.

PAGES = PrimaryPages

Sizes the hash index to reflect the expected number of pages in your table. To determine the value for PrimaryPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for PrimaryPages (256000/256=1000).

The value for PrimaryPages must be a positive constant and must be greater than 0.

If your estimate for PrimaryPages is too small, performance may be degraded.

For more information on hash indexes, see "CREATE TABLE".

WHERE ExternalSearchCondition

The WHERE clause evaluated by the Oracle database for the cache group table. This WHERE clause is added to every LOAD and REFRESH operation on the cache group. It may not directly reference other tables. It is parsed by both TimesTen and Oracle Database. See "Using a WHERE clause" in Oracle TimesTen In-Memory Database Cache Guide.

ParentDistributionClause

In TimesTen Scaleout, distribution clause for a parent table in a static read-only cache group with incremental autorefresh. These distribution schemes are supported for parent tables:

  • DISTRIBUTE BY HASH [(ColumnName [,...])]

  • DUPLICATE

ChildDistributionClause

In TimesTen Scaleout, distribution clause for a child table in a static read-only cache group with incremental autorefresh. These distribution schemes are supported for child tables:

  • DISTRIBUTE BY HASH [(ColumnName [,...])]

  • DISTRIBUTE BY REFERENCE [(ForeignKeyConstraint)]

  • DUPLICATE

AGING LRU [ON | OFF]

Supported in TimesTen Classic only.

If specified, defines the LRU aging policy on the root table. The LRU aging policy applies to all tables in the cache group. The LRU aging policy defines the type of aging (least recently used (LRU)), the aging state (ON or OFF) and the LRU aging attributes.

Set the aging state to either ON or OFF. ON indicates that the aging state is enabled and aging is done automatically. OFF indicates that the aging state is disabled and aging is not done automatically. In both cases, the aging policy is defined. The default is ON.

In dynamic cache groups, LRU aging is ON by default. However, you can specify time-based aging or set LRU aging to OFF at the syntax level.

LRU aging cannot be specified on a cache group with the autorefresh attribute, unless the cache group is dynamic.

LRU attributes are defined by calling the ttAgingLRUConfig and/or the ttAgingTableLRUConfig built-in procedures. LRU attributes are not defined at the SQL level. See "ttAgingLRUConfig" and "ttAgingTableLRUConfig" in the Oracle TimesTen In-Memory Database Reference and "Implementing an aging policy in your tables" in the Oracle TimesTen In-Memory Database Operations Guide for more information.

AGING USE ColumnName...[ON|OFF]

Supported in TimesTen Classic only.

If specified, defines the time-based aging policy on the root table. The time-based aging policy applies to all tables in the cache group. The time-based aging policy defines the type of aging (time-based), the aging state (ON or OFF) and the time-based aging attributes.

Set the aging state to either ON or OFF. ON indicates that the aging state is enabled and aging is done automatically. OFF indicates that the aging state is disabled and aging is not done automatically. In both cases, the aging policy is defined. The default is ON.

Time-based aging attributes are defined at the SQL level and are specified by the LIFETIME and CYCLE clauses.

Specify ColumnName as the name of the column used for time-based aging. Define the column as NOT NULL and of data type TIMESTAMP or DATE. The value of this column is subtracted from SYSDATE, truncated using the specified unit (second, minute, hour, day) and then compared to the LIFETIME value. If the result is greater than the LIFETIME value, then the row is a candidate for aging.

The values of the column used for aging are updated by your applications. If the value of this column is unknown for some rows, and you do not want the rows to be aged, define the column with a large default value (the column cannot be NULL).

For more information about time-based aging, see "Implementing aging in a cache group for TimesTen Classic" in Oracle TimesTen In-Memory Database Cache Guide.

LIFETIME Num1 {SECOND[S]|MINUTE[S]|HOUR[S]DAY[S]}

Supported in TimesTen Classic only.

LIFETIME is a time-based aging attribute and is a required clause.

Specify the LIFETIME clause after the AGING USE ColumnName clause.

The LIFETIME clause specifies the minimum amount of time data is kept in cache.

Specify Num1 as a positive integer constant to indicate the unit of time expressed in seconds, minutes, hours or days that rows should be kept in cache. Rows that exceed the LIFETIME value are aged out (deleted from the table).

The concept of time resolution is supported. If DAYS is specified as the time resolution, then all rows whose timestamp belongs to the same day are aged out at the same time. If HOURS is specified as the time resolution, then all rows with timestamp values within that hour are aged at the same time. A LIFETIME of 3 days is different than a LIFETIME of 72 hours (3*24) or a LIFETIME of 432 minutes (3*24*60).

[CYCLE Num2 {SECOND[S] | MINUTE[S] |HOUR[S]|DAY[S]}]

Supported in TimesTen Classic only.

CYCLE is a time-based aging attribute and is optional. Specify the CYCLE clause after the LIFETIME clause.

The CYCLE clause indicates how often the system should examine rows to see if data exceeds the specified LIFETIME value and should be aged out (deleted).

Specify Num2 as a positive integer constant.

If you do not specify the CYCLE clause, then the default value is 5 minutes. If you specify 0 for Num2, then the aging thread wakes up every second.

If the aging state is OFF, then aging is not done automatically and the CYCLE clause is ignored.

Cache groups in TimesTen Scaleout

TimesTen Scaleout supports static read-only cache groups with incremental autorefresh. You can specify a distribution scheme on a parent table and on one or more child tables. The distribution scheme specifies how data is distributed across the elements of the database.

A distribution scheme is denoted by the DISTRIBUTE BY clause:
  • For a single table cache group, the default distribution scheme is HASH.
  • If you do not specify a column in the DISTRIBUTE BY clause, the primary key columns are used as the key columns for the distribution scheme.
  • For a multiple table cache group, you can specify either the HASH or the DUPLICATE distribution scheme on the parent table. If you define the DUPLICATE distribution scheme, you can only specify HASH or DUPLICATE on the child tables.
  • For a multiple table cache group, HASH is the default distribution scheme for the parent table and all child tables default to the REFERENCE distribution scheme. If you specify DUPLICATE on the parent table, and do not specify a distribution scheme for the child tables, the default distribution scheme for the child tables is DUPLICATE.
  • If the foreign key on a child table is identical to the primary key on the parent table, the HASH distribution scheme is used for the child table as an optimization.
  • It is best practice to distribute child tables by reference.

See "Distribution schemes for TimesTen Cache in TimesTen Scaleout" in the Oracle TimesTen In-Memory Database Scaleout User's Guide for more information on distribution schemes.

The following are not supported:
  • Full autorefresh mode
  • Aging
  • Materialized views
  • Global indexes

For more information on static read-only cache groups in TimesTen Scaleout, see "Using Cache Groups in TimesTen Scaleout" in the Oracle TimesTen In-Memory Database Scaleout User's Guide.

Cache groups in TimesTen Classic

Dynamic hybrid read-only cache groups

A dynamic hybrid read-only cache group is a dynamic read-only cache group where the root table does not exist in the Oracle database. The root table is automatically created in the TimesTen database from the cache group definition. The cache group definition includes the description of this root table, as if it existed in the Oracle database.

Description and restrictions:
  • The root table must not exist in the Oracle database.
  • The root table in the TimesTen database must have a primary key.
  • The root table can only contain columns in the primary key. The primary key must be referenced by at least one child table.
  • For dynamic load triggering, you can use a derived table in the FROM clause of the SELECT statement. You can also specify more than one table of the same hybrid cache group in the SELECT query.
  • If you issue a SELECT query on the root table in the TimesTen database, this SELECT operation does not trigger a dynamic load.
  • You cannot specify time-based aging on this type of cache group. LRU aging is enabled by default.
  • The WHERE clause in the cache group definition is not supported.
  • You cannot use the LOAD CACHE GROUP statement to manually load the cache group.
  • The UNLOAD CACHE GROUP ...WITH ID statement is not supported.

See "Hybrid cache group" in the Oracle TimesTen In-Memory Database Cache Guide for more information on dynamic hybrid read-only cache groups.

Description of cache groups

  • Two cache groups cannot have the same owner name and group name. If you do not specify the owner name, your schema becomes the owner name for the new cache group.

  • Neither a cache table name nor a cache group name can contain #.

  • Dynamic parameters are not allowed in the WHERE clause.

  • Oracle Database temporary tables cannot be cached.

  • Each table must correspond to a table in the Oracle database.

  • In the Oracle database, you can define a parent/child relationship and then insert a null value into the foreign key column of the child table. This means this row in the child table references a null parent. You can then create a cache group and cache the parent/child relationship of the Oracle database tables. However, if you load data from the Oracle database tables into the cache group, the row that contains the null value of the foreign key column is not loaded. TimesTen recommends that you do not create cache groups if the tables you cache define a parent/child relationship in which the foreign key represents a null parent.

  • You cannot use lowercase delimited identifiers to name your cache tables. Table names in TimesTen are case-insensitive and are stored as uppercase. The name of the cache table must be the same as the Oracle database table name. Uppercase table names on TimesTen will not match mixed case table names on the Oracle database. As a workaround, create a synonym for your table in the Oracle database and use that synonym as the table name for the cache group. This workaround is not available for read-only cache groups or cache groups with the AUTOREFRESH parameter set.

  • Each column in the cache table must match each column in the Oracle database table, both in name and in data type. See "Mappings between Oracle Database and TimesTen data types" in Oracle TimesTen In-Memory Database Cache Guide. In addition, each column name must be fully qualified with an owner and table name when referenced in a WHERE clause.

  • The WHERE clause can only directly refer to the cache group table. Tables that are not in the cache group can only be referenced with a subquery.

  • Generally, you do not have to fully qualify the column names in the WHERE clause of the CREATE CACHE GROUP, LOAD CACHE GROUP, UNLOAD CACHE GROUP, REFRESH CACHE GROUP or FLUSH CACHE GROUP statements. However, since TimesTen automatically generates queries that join multiple tables in the same cache group, a column must be fully qualified if there is more than one table in the cache group that contains columns with the same name.

  • By default, a range index is created to enforce the primary key for a cache group table. Use the UNIQUE HASH clause if you want to specify a hash index for the primary key.

    • If your application performs range queries over a cache group table's primary key, then choose a range index for that cache group table by omitting the UNIQUE HASH clause.

    • If, however, your application performs only exact match lookups on the primary key, then a hash index may offer better response time and throughput. In such a case, specify the UNIQUE HASH clause. See "CREATE TABLE" for more information on the UNIQUE HASH clause.

    • Use ALTER TABLE to change the representation of the primary key index for a table.

  • For cache group tables with the PROPAGATE attribute and for tables of SWT and AWT cache groups, foreign keys specified with ON DELETE CASCADE must be a proper subset of foreign keys with ON DELETE CASCADE in the Oracle database tables.

  • You cannot execute the CREATE CACHE GROUP statement when performed under the serializable isolation level. An error message is returned when attempted.

The AUTOREFRESH parameter automatically propagates changes from the Oracle database to TimesTen cache groups. For static cache groups, deletes, updates and inserts are automatically propagated from the Oracle database to the cache group. For dynamic cache groups, only deletes and updates are propagated. Inserts to the specified Oracle database tables are not propagated to dynamic cache groups. They are dynamically loaded into TimesTen Cache when referenced by the application. They can also be explicitly loaded by the application.

To use autorefresh with a cache group, you must specify AUTOREFRESH when you create the cache group. You can change the MODE, STATE and INTERVAL AUTOREFRESH settings after a cache group has been created by using the ALTER CACHE GROUP statement. Once a cache group has been specified as either AUTOREFRESH or PROPAGATE, you cannot change these attributes. If you are creating a read-only cache group, you do not need to specify the autorefresh clause. A read-only cache group defaults to incremental autorefresh.

TimesTen supports FULL or INCREMENTAL AUTOREFRESH. In FULL mode, the entire cache is periodically unloaded and then reloaded. In INCREMENTAL mode, TimesTen installs triggers in the Oracle database to track changes and periodically updates only the rows that have changed in the specified Oracle database tables. The first incremental refresh is always a full refresh, unless the autorefresh state is PAUSED. The default mode is INCREMENTAL.

FULL AUTOREFRESH is more efficient when most of the Oracle database table rows have been changed. INCREMENTAL AUTOREFRESH is more efficient when there are fewer changes.

TimesTen schedules an autorefresh operation when the transaction that contains a statement with AUTOREFRESH specified is committed. The statement types that cause autorefresh to be scheduled are:

  • A CREATE CACHE GROUP statement in which AUTOREFRESH is specified, and the AUTOREFRESH state is specified as ON.

  • An ALTER CACHE GROUP statement in which the AUTOREFRESH state has been changed to ON.

  • A LOAD CACHE GROUP statement on an empty cache group whose autorefresh state is PAUSED.

The specified interval determines how often autorefresh occurs.

The current STATE of AUTOREFRESH can be ON, OFF or PAUSED. By default, the autorefresh state is PAUSED.

The NOT PROPAGATE attribute cannot be used with the AUTOREFRESH attribute.

Aging in cache groups:

  • You can implement sliding windows with time-based aging. See "Configuring a sliding window in TimesTen Classic" in Oracle TimesTen In-Memory Database Cache Guide.

  • After you have defined an aging policy for the table, you cannot change the policy from LRU to time-based or from time-based to LRU. You must first drop aging and then alter the table to add a new aging policy.

  • The aging policy must be defined to change the aging state.

  • LRU and time-based aging can be combined in one system. If you use only LRU aging, the aging thread wakes up based on the cycle specified for the whole database. If you use only time-based aging, the aging thread wakes up based on an optimal frequency. This frequency is determined by the values specified in the CYCLE clause for the database. If you use both LRU and time-based aging, then the thread wakes up based on a combined consideration of both types.

  • Call the ttAgingScheduleNow procedure to schedule the aging process right away regardless if the aging state is ON or OFF.

  • The following rules determine if a row is accessed or referenced for LRU aging:

    • Any rows used to build the result set of a SELECT statement.

    • Any rows used to build the result set of an INSERT...SELECT statement.

    • Any rows that are about to be updated or deleted.

  • Compiled commands are marked invalid and need recompilation when you either drop LRU aging from or add LRU aging to tables that are referenced in the commands.

  • use

    For LRU aging, if a child row is not a candidate for aging, then neither this child row nor its parent row are deleted. ON DELETE CASCADE settings are ignored.

  • For time-based aging, if a parent row is a candidate for aging, then all child rows are deleted. ON DELETE CASCADE (whether specified or not) is ignored.

  • Specify either the LRU aging or time-based aging policy on the root table. The policy applies to all tables in the cache group.

  • For the time-based aging policy, you cannot add or modify the aging column. This is because you cannot add or modify a NOT NULL column.

  • Restrictions on defining aging for a cache group:

    • LRU aging is not supported on a cache group defined with the autorefresh attribute, unless it is a dynamic cache group.

    • The aging policy cannot be added, altered, or dropped for read-only cache groups or cache groups with the AUTOREFRESH attribute while the cache agent is active. Stop the cache agent first.

    • You cannot drop the column that is used for time-based aging.

Examples: TimesTen Classic

These examples are specific to TimesTen Classic. For information and examples on using cache groups in TimesTen Scaleout, see "Using Cache Groups in TimesTen Scaleout" in the Oracle TimesTen In-Memory Database Scaleout User's Guide.

Create a read-only cache group:

CREATE READONLY CACHE GROUP customerorders
AUTOREFRESH INTERVAL 10 MINUTES
FROM
customer (custid INT NOT NULL,
       name CHAR(100) NOT NULL,
       addr CHAR(100),
       zip INT,
       region CHAR(10),
       PRIMARY KEY(custid)),
ordertab (orderid INT NOT NULL,
       custid INT NOT NULL,
       PRIMARY KEY (orderid),
       FOREIGN KEY (custid) REFERENCES customer(custid));

Create an asynchronous writethrough cache group:

CREATE ASYNCHRONOUS WRITETHROUGH CACHE GROUP cstomers
FROM
customer (custid INT NOT NULL,
       name CHAR(100) NOT NULL,
       addr CHAR(100),
       zip INT,
       PRIMARY KEY(custid));

Create a synchronous writethrough cache group:

CREATE SYNCHRONOUS WRITETHROUGH CACHE GROUP customers
FROM
customer (custid INT NOT NULL,
       name CHAR(100) NOT NULL,
       addr CHAR(100),
       zip INT,
       PRIMARY KEY(custid));

Create a user managed cache group:

CREATE USERMANAGED CACHE GROUP updateanywherecustomers
AUTOREFRESH
       MODE INCREMENTAL
       INTERVAL 30 SECONDS
       STATE ON
FROM
customer (custid INT NOT NULL,
       name CHAR(100) NOT NULL,
       addr CHAR(100),
       zip INT,
       PRIMARY KEY(custid),
       PROPAGATE);

Create a cache group with time-based aging. Specify agetimestamp as the column for aging. Specify LIFETIME 2 hours, CYCLE 30 minutes. Aging state is not specified, so the default setting (ON) is used.

CREATE READONLY CACHE GROUP agingcachegroup
AUTOREFRESH
       MODE INCREMENTAL
       INTERVAL 5 MINUTES
       STATE PAUSED
FROM
customer (customerid NUMBER NOT NULL,
       agetimestamp TIMESTAMP NOT NULL,
       PRIMARY KEY (customerid))
       AGING USE agetimestamp LIFETIME 2 HOURS CYCLE 30 MINUTES;

Command> DESCRIBE customer;
Table USER.CUSTOMER:
  Columns:
   *CUSTOMERID                      NUMBER NOT NULL
    AGETIMESTAMP                    TIMESTAMP (6) NOT NULL
  AGING USE AgeTimestamp LIFETIME 2 HOURS CYCLE 30 MINUTES ON
1 table found.
(primary key columns are indicated with *)

Use a synonym for a mixed case delimited identifier table name in the Oracle database so the mixed case table name can be cached in TimesTen. First attempt to cache the mixed case Oracle database table name. You see the error "Could not find 'NameofTable' in Oracle":

Command> AUTOCOMMIT 0;
Command> PASSTHROUGH 3;
Command> CREATE TABLE "MixedCase" (col1 NUMBER PRIMARY KEY NOT NULL);
Command> INSERT INTO "MixedCase" VALUES (1);
1 row inserted.
Command> COMMIT;
Command> CREATE CACHE GROUP MixedCase1 from "MixedCase" 
        (col1 NUMBER PRIMARY KEY NOT NULL);
 5140: Could not find SAMPLEUSER.MIXEDCASE in Oracle.  May not have privileges.
The command failed.

Now, using the PassThrough attribute, create the synonym "MIXEDCASE" in the Oracle database and use that synonym as the table name.

Command> AUTOCOMMIT 0;
Command> PASSTHROUGH 3;
Command> CREATE SYNONYM "MIXEDCASE" FOR "MixedCase";
Command> COMMIT;
Command> CREATE CACHE GROUP MixedCase2 FROM "MIXEDCASE" 
         (col1 NUMBER PRIMARY KEY NOT NULL);
Warning  5147: Cache group contains synonyms
Command> COMMIT;

Attempt to use a synonym name with a read-only cache group or a cache group with the AUTOREFRESH attribute. You see an error:

Command> AUTOCOMMIT 0;
Command> PASSTHROUGH 3;
Command> CREATE SYNONYM "MIXEDCASE_AUTO" FOR "MixedCase";
Command> COMMIT;
Command> CREATE READONLY CACHE GROUP MixedCase3 AUTOREFRESH MODE
         INCREMENTAL INTERVAL 10 MINUTES FROM "MIXEDCASE_AUTO" 
         (Col1 NUMBER PRIMARY KEY NOT NULL);
 5142: Autorefresh is not allowed on cache groups with Oracle synonyms
The command failed.

CREATE FUNCTION

The CREATE FUNCTION statement creates a standalone stored function.

Required privilege

CREATE PROCEDURE (if owner) or CREATE ANY PROCEDURE (if not owner).

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

CREATE [OR REPLACE] FUNCTION [Owner.]FunctionName 
     [(arguments [IN|OUT|IN OUT][NOCOPY] DataType [DEFAULT expr][,...])]
     RETURN DataType 
     [InvokerRightsClause][AccessibleByClause][DETERMINISTIC]
     {IS|AS} PlsqlFunctionBody

InvokerRightsClause::=
AUTHID {CURRENT_USER|DEFINER}

AccessibleByClause::=
ACCESSIBLE BY (accessor[,...])

accessor::= 
[UnitKind][Owner.]UnitName

You can specify InvokerRightsClause, AccessibleByClause, or DETERMINISTIC in any order.

Parameters

Parameter Description

OR REPLACE

Specify OR REPLACE to recreate the function if it already exists. Use this clause to change the definition of an existing function without dropping and recreating it. When you recreate a function, TimesTen recompiles it.

FunctionName

Name of function.

arguments

Name of argument or parameter. You can specify 0 or more parameters for the function. If you specify a parameter, you must specify a data type for the parameter. The data type must be a PL/SQL data type.

IN|OUT|IN OUT

Parameter modes.

IN is a read-only parameter. You can pass the parameter's value into the function but the function cannot pass the parameter's value out of the function and back to the calling PL/SQL block. The value of the parameter cannot be changed.

OUT is a write-only parameter. Use an OUT parameter to pass a value back from the function to the calling PL/SQL block. You can assign a value to the parameter.

IN OUT is a read/write parameter. You can pass values into the function and return a value back to the calling program (either the original, unchanged value or a new value set within the function.

IN is the default.

NOCOPY

Specify NOCOPY to instruct TimesTen to pass the parameter as fast as possible. You can enhance performance when passing a large value such as a record, an index-by-table, or a varray to an OUT or IN OUT parameter. IN parameters are always passed NOCOPY.

DEFAULT expr

Use this clause to specify a default value for the parameter. You can specify := in place of the keyword DEFAULT.

RETURN DataType

Required clause. A function must return a value. You must specify the data type of the return value of the function.

Do not specify a length, precision, or scale for the data type.

The data type is a PL/SQL data type.

InvokerRightsClause

Lets you specify whether the SQL statements in PL/SQL functions or procedures execute with definer's or invoker's rights. The AUTHID setting affects the name resolution and privilege checking of SQL statements that a PL/SQL procedure or function issues at runtime, as follows:

  • Specify DEFINER so that SQL name resolution and privilege checking operate as though the owner of the procedure or function (the definer, in whose schema it resides) is running it. DEFINER is the default.

  • Specify CURRENT_USER so that SQL name resolution and privilege checking operate as though the current user (the invoker) is running it.

For more information, see "Definer's rights and invoker's rights (AUTHID clause)" in the Oracle TimesTen In-Memory Database Security Guide.

AccessibleByClause

Use this clause to specify one or more accessors (PL/SQL units) that can invoke the function directly. The list of accessors that can access the function is called a white list. A white list gives you the ability to add an extra layer of security to your PL/SQL objects. Specifically, you can restrict access to the function to only those objects on the white list.

AccessibleByClause can appear only once in the CREATE FUNCTION statement.

Syntax: ACCESSIBLE BY (accessor [,...])

accessor

Used in AccessibleByClause. An accessor is a PL/SQL unit that can invoke the function.

An accessor can appear more than once in the AccessibleByClause clause.

Syntax: [UnitKind][Owner.]UnitName

UnitKind

Used in the accessor clause (which is part of the AccessibleByClause clause). Specifies the kind of PL/SQL unit that can invoke the function.

  • UnitKind is optional, but if specified, valid options are:
  • FUNCTION

  • PROCEDURE

  • PACKAGE

[Owner.]UnitName

Used in the accessor clause (which is part of the AccessibleByClause clause). Specifies the name of the PL/SQL unit that can invoke the function. If you specify UnitKind, then UnitName must be a name of a unit of that kind. For example, if you specify PROCEDURE for UnitKind, then UnitName must be the name of a procedure. UnitName is required.

You can optionally specify Owner. If you specify Owner, then UnitName must reside in that owner's schema. If you do not specify Owner, UnitName must be in the schema that contains the function.

DETERMINISTIC

Specify DETERMINISTIC to indicate that the function returns the same result value whenever it is called with the same values for its parameters.

IS|AS

Specify either IS or AS to declare the body of the function.

plsql_function_spec

Specifies the function body.

Description

  • AccessibleByClause:

    • The compiler checks the validity of the syntax of the ACCESSIBLE BY clause, but does not check that the accessor exists. Therefore, you can define an accessor that does yet exist in the owner's schema.

    • When you invoke the function, the compiler first does the normal permission checks on the invocation. If any check fails, the invocation fails, even if the invoker is an accessor. If all normal permission checks on the invocation succeed, and the function has no ACCESSIBLE BY clause, the invocation succeeds. If the function has an ACCESSIBLE BY clause, the invocation succeeds only if the invoker is an accessor.

  • When you create or replace a function, the privileges granted on the function remain the same. If you drop and recreate the object, the object privileges that were granted on the original object are revoked.

  • In a replication environment, the CREATE FUNCTION statement is not replicated. For more information, see "Creating a new PL/SQL object in an existing active standby pair" and "Adding a PL/SQL object to an existing classic replication scheme" in the Oracle TimesTen In-Memory Database Replication Guide.

  • TimesTen does not support:

    • parallel_enable_clause

      You can specify this clause, but it has no effect.

    • call_spec clause

    • AS EXTERNAL clause

Examples

Using the AccessibleByClause

This example creates the ProtectedFunction function. The ACCESSIBLE BY clause is used to restrict the invocation of the function to the CallingProc1 and CallingProc2 procedures. Note that for CallingProc1, the type of PL/SQL unit is not specified and for CallingProc2, the type of PL/SQL unit is specified (PROCEDURE).

Command> CREATE OR REPLACE FUNCTION ProtectedFunction (a IN NUMBER)
           RETURN NUMBER
         ACCESSIBLE BY (CallingProc1, PROCEDURE CallingProc2)
         AS
         BEGIN
           RETURN a * 1;
         END;
         /
         
Function created.

Create the CallingProc1 and CallingProc2 procedures.

Command> CREATE OR REPLACE PROCEDURE CallingProc1 AS
           a NUMBER:=1;
         BEGIN
           a:=ProtectedFunction(a);
           DBMS_OUTPUT.PUT_LINE ('Calling Procedure: '|| a);
         END;
         /
         
Procedure created.

Command> CREATE OR REPLACE PROCEDURE CallingProc2
         AS
           a NUMBER:=2;
         BEGIN
           a:=ProtectedFunction(a);
           DBMS_OUTPUT.PUT_LINE ('Calling Procedure: '|| a);
         END;
         /
         
Procedure created.

Call the procedures. CallingProc1 and CallingProc2 are in the white list, resulting in successful execution.

Command> SET SERVEROUTPUT ON
Command> exec CallingProc1;
Calling Procedure: 1
 
PL/SQL procedure successfully completed.
 
Command> exec CallingProc2;
Calling Procedure: 2
 
PL/SQL procedure successfully completed.

Illustrating the syntax for creating a PL/SQL function

Create function get_sal with one input parameter. Return salary as type NUMBER.

Command> CREATE OR REPLACE FUNCTION get_sal
           (p_id employees.employee_id%TYPE) RETURN NUMBER IS
           v_sal employees.salary%TYPE := 0;
         BEGIN
           SELECT salary INTO v_sal FROM employees
             WHERE employee_id = p_id;
           RETURN v_sal;
         END get_sal;
         /
 
Function created.

CREATE INDEX

The CREATE INDEX statement creates an index on one or more columns of a table or a materialized view.

Required privilege

For a global index (supported in TimesTen Scaleout) and for a local index (supported in TimesTen Scaleout and in TimesTen Classic:
  • If the owner, no privilege is required.
  • If not the owner, the CREATE ANY INDEX system privilege or the INDEX object privilege is required.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout. You can create a global or a local index.

SQL syntax

The syntax to create a range index follows. Use the GLOBAL keyword, the optional INCLUDE clause, and the optional IndexDistributionClause clause to create a global index. A global index is supported only in TimesTen Scaleout:

CREATE [GLOBAL][UNIQUE] INDEX [Owner.]IndexName ON
[Owner.]TableName (ColumnName [ASC | DESC][,... ])
[INCLUDE (ColumnName[,…])] 
[IndexDistributionClause]

The syntax to create a hash index follows. Use the GLOBAL keyword to create a global index. The optional INCLUDE clause, and the optional IndexDistributionClause clause can only be used with a global index. A global index is supported only in TimesTen Scaleout:

CREATE [GLOBAL][UNIQUE] HASH INDEX [Owner.]IndexName ON
[Owner.]TableName (ColumnName [ASC | DESC][,... ] ) 
[INCLUDE (ColumnName [,…])]
[ PAGES = RowPages | CURRENT ]
[IndexDistributionClause]

The syntax for IndexDistributionClause can only be used for a global index (supported in TimesTen Scaleout):

IndexDistributionClause::=
DISTRIBUTE BY HASH [(ColumnName [,…])]

Parameters

Parameter Description

GLOBAL

The GLOBAL keyword is only supported in TimesTen Scaleout.

In TimesTen Scaleout:
  • Specify GLOBAL to create a global index. A global index maps all rows in the table to a hash distribution scheme. When you create a global index, TimesTen Scaleout creates a materialized view with a local index and a hash distribution scheme to the index key columns.
  • If you do not specify GLOBAL, a local index is created in each database element. The index in this database element maps to rows in the table in the database element.

UNIQUE

You can specify UNIQUE for both range and hash indexes. If you specify UNIQUE each possible combination of index key column values can occur in only one row of the table.

HASH

Specify HASH to create a hash index. Specify UNIQUE with HASH to create a unique hash index.

[Owner.] IndexName

Name to be assigned to the new index. A table cannot have two indexes with the same name. If the owner is specified, it must be the same as the owner of the table.

[Owner.] TableName

Designates the table or materialized view for which an index is to be created.

ColumnName

Name of a column to be used as an index key. You can specify up to 32 columns in order from major index key to minor index key.

[ASC|DESC]

Specifies the order of the index to be either ascending (the default) or descending. In TimesTen, this clause is currently ignored.

INCLUDE (ColumnName [,…])

The INCLUDE clause is only supported in TimesTen Scaleout and can only be used when defining a global index. Use the INCLUDE clause to add non-index columns to the index definition. These non-index columns can be used to satisfy some queries without accessing the base table. This is especially useful if such columns are accessed frequently in queries (both for equality and for range conditions).

PAGES = {RowPages | CURRENT}

The PAGES clause sizes the hash index to reflect the expected number of pages in the table. If you do not specify the PAGES clause when defining a hash index, the default is CURRENT.

If you specify CURRENT, the current number of rows in the table is used to calculate the page count value. If you specify RowPages, the number of pages is used. To determine the value for RowPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for RowPages (256000/256=1000).

The value for RowPages must be a positive constant and must be greater than 0.

Do not specify PAGES=CURRENT if there are no rows in your table. This is because when rows are added to the table, the hash index does not perform optimally.

IndexDistributionClause :: = DISTRIBUTE BY HASH [( ColumnName [,…])]

You can specify the DISTRIBUTE BY HASH clause only if you are defining a global index. This clause is optional.

If you do not specify this clause, the column(s) defined in the global index definition form the distribution key.

If you do specify this clause, you can optionally specify the ColumnName clause:
  • If specified, you must then specify one or more columns for the distribution key. These columns must include one or more of the columns specified in the global index definition.

  • If not specified, the column(s) defined in the global index definition form the distribution key.

    Note:

    The distribution key of the global index cannot be the same as the distribution key of the table.

Indexes in TimesTen Scaleout

TimesTen Scaleout supports global and local indexes:
  • Global index: Maps all rows in the table to a hash distribution scheme. When you create a global index, TimesTen Scaleout creates a materialized view with a local index and a hash distribution scheme to the index key columns. The local index that is created on the materialized view further optimizes query performance.
  • Local index: Is created in each database element. The index in this database element maps to rows in the table in this database element. Queries on index columns that do not include the distribution key columns on the table require communication with an element in every replica set.

See "Understanding indexes" in the Oracle TimesTen In-Memory Database Scaleout User's Guide for information on using indexes in TimesTen Scaleout.

Choosing a global or a local index in TimesTen Scaleout

Consider the following when deciding whether to use a global or a local index:
  • Use a global index for:

    • Unique columns: A global unique index optimizes query execution by performing unique constraint checks more efficiently. TimesTen Scaleout uses the distribution key columns for uniqueness verification instead of accessing all replica sets. However, if the distribution key is a subset of the index key, create a local index.
    • Queries that have an equality predicate that do not include all of the columns in the distribution key of the table.
    • A group of columns that are frequently joined in queries with primary key columns.
    • Non-index columns that are frequently used in queries: Define a global index with the INCLUDE clause to include those non-index columns. In such a case, the table does not need to be accessed to satisfy the query.
    • An index where the index key is a prefix of the distribution key of the table.
  • Use a local index for:
    • Non-unique columns: If the index key consists of only non-unique columns, create a local non-unique index.
    • An index key that has the same columns as the distribution key for the table.
    • The situation where the distribution key of the table is a prefix of the index key.
    • Queries that have an equality predicate that includes all columns in the distribution key of the table.

See "Understanding indexes" in the Oracle TimesTen In-Memory Database Scaleout User's Guide for more information.

Description of global indexes

Global index usage:
  • You must specify the GLOBAL keyword to create a global index. An index is local by default.
  • Global indexes by default are distributed by hash and can only be distributed by hash. Local indexes are not distributed.
  • When you create a global index, TimesTen Scaleout internally creates its own materialized view and its own local index on that materialized view.
  • Global indexes result in more efficient query execution with joins. However DML operations are slower due to the maintenance of the internal materialized view (that is created when you define a global index).
  • When a new element is added in the grid, the schema is replicated on the new element. In addition, the rows are redistributed, and the indexes are rebuilt. This includes the global indexes. Similarly, when an element is removed from the grid, the rows are redistributed and the indexes are rebuilt.
Distribution scheme of table:
  • You can define a global index on a table distributed by hash and on a table distributed by reference. Global indexes on parent and child (first-level reference) tables are supported. However, you cannot define a global index on grandchild tables or any tables that are not first-level reference tables.
  • You cannot define a global index on a table distributed by duplicate.

Restrictions on global indexes:

  • The column list for the distribution key cannot contain the ROWID pseudocolumn or a column of type ROWID.
  • Not supported on a global temporary table.
  • Not supported on readonly cache groups.
  • Not supported on a materialized view.

Syntax and semantic rules for global indexes

You must specify the GLOBAL keyword to create a global index. If you do not specify the GLOBAL keyword, a local index is created. Global indexes are distributed by hash on index key columns.

If you specify the GLOBAL keyword, you can optionally specify these clauses that are specific to global indexes:
  • INCLUDE clause: Optional clause that enables you to include non-key columns in the index. If such columns are frequently accessed by queries that use the index, this may improve performance.
  • IndexDistributionClause: Optional clause that enables you to specify what columns to use for the hash distribution. If you do not specify this clause, then the index columns form the distribution key. The distribution key of the index cannot be the same as the distribution key of the table.

  • Examples:

    Global range index:

    Command> CREATE GLOBAL INDEX globalindex1 ON mytab (a) INCLUDE (b,c) DISTRIBUTE BY HASH (a);
    Command> indexes mytab;
    
    Indexes on table SAMPLEUSER.MYTAB:
      MYTAB: unique range index on columns:
        C
        B
      GLOBALINDEX1: global non-unique range index on columns:
        A
        Included columns:
          B
          C
      2 indexes found.
    
    2 indexes found on 1 table.
    Command> drop table mytab;

    Global hash index:

    Command> CREATE TABLE mytab (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
               a TT_INTEGER NOT NULL, PRIMARY KEY (c,b)) DISTRIBUTE BY HASH (a,b);
    Command> CREATE GLOBAL HASH INDEX globalhashindex1 ON mytab(a) INCLUDE (b,c) PAGES=200 DISTRIBUTE BY HASH (a);
    Command> indexes MYTAB;
    
    Indexes on table SAMPLEUSER.MYTAB:
      MYTAB: unique range index on columns:
        C
        B
      GLOBALHASHINDEX1: global non-unique hash index on columns:
        A
        Included columns:
          B
          C
      2 indexes found.
    
    2 indexes found on 1 table.
    

See "Examples: TimesTen Scaleout" for additional examples.

General description of indexes in TimesTen Scaleout

  • TimesTen creates a nonunique range index by default. Specify CREATE UNIQUE INDEX to create a unique range index.

  • To create a nonunique hash index, specify CREATE HASH INDEX. To create a unique hash index, specify CREATE UNIQUE HASH INDEX.

  • If UNIQUE is specified, all existing rows must have unique values in the indexed column(s).

  • The new index is maintained automatically until the index is deleted by a DROP INDEX statement or until the table associated with it is dropped.

  • Any prepared statements that reference the table with the new index are automatically prepared again the next time they are executed. Then the statements can take advantage, if possible, of the new index.

  • An index on a temporary table cannot be created by a connection if any other connection has a non-empty instance of the table.

  • If you are using linguistic comparisons, you can create a linguistic index. A linguistic index uses sort key values and storage is required for these values. Only one unique value for NLS_SORT is allowed for an index. For more information on linguistic indexes and linguistic comparisons, see "Using linguistic indexes" in Oracle TimesTen In-Memory Database Operations Guide.

  • If you create indexes that are redundant, TimesTen generates warnings or errors. Call ttRedundantIndexCheck to see the list of redundant indexes for your tables.

  • To change the size or type of a hash index, drop the hash index and create a new index.

  • A hash index is created with a fixed size that remains constant for the life of the table. To resize the hash index, drop and recreate the index. If the hash index has insufficient pages it results in hash collisions which slows down the index look-up. Hash key comparison is a fast operation, so a small number of hash collisions should not cause a performance problem for TimesTen.

    To ensure that your hash index is sized correctly, your application must indicate the expected size of your table with the value of the RowPages parameter of the SET PAGES clause. Compute this value by dividing the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for the value of RowPages (256000/256=1000).

  • The maximum number of columns that can be specified for an index is 32.

Using indexes in query processing

Proper indexes can improve query performance. Some queries can benefit from the use of indexes and some queries do not benefit from the use of indexes. Additionally, the choice of indexes for your queries is important.

A range index is ideal for processing range searches and exact matches, especially if most of the values in the index columns are unique. For example, if a range index is defined on columns (C1,C2), the index can be used to process the following types of predicates. ConstantOrParam refers to a constant value or dynamic parameter and range refers to the operators >,<,>=, or <=:

  • C1 = ConstantOrParam AND C2 = ConstantOrParam

  • C1 = ConstantOrParam AND C2 range ConstantOrParam

  • C1 = ConstantOrParam

  • C1 range ConstantOrParam

A range index efficiently processes equality and range predicates and efficiently processes sort and group operations. Use range indexes on index columns with many unique values. The order of columns you specify in a range index is relevant. The order of expressions in the predicate of a query that uses the range index is not relevant. When your query is processed, only one range index is used for each scan of your table even if you have defined multiple range indexes on your table.

A hash index efficiently processes equality predicates. You must size your hash index correctly for optimal performance. Use the PAGES parameter to size your hash index. If you specify a PAGES value that is too small, a large number of hash collisions may result, leading to performance degradation for statements that access the hash index. The order of columns specified in the hash index is not relevant and the order of expressions in the predicate of the query that uses the hash index is not relevant. If either a hash index or a range index can be used to process a particular equality predicate, the hash index is chosen because a lookup in a hash index is faster than a scan of a range index.

You can influence the indexes used by the optimizer by setting statement level or transaction level optimizer hints. see "Statement level optimizer hints" for information on statement level optimizer hints. For more information on transaction level optimizer hints, see "ttOptSetFlag", ttOptSetOrder", or "ttOptUseIndex" in the Oracle TimesTen In-Memory Database Reference.

Examples: TimesTen Scaleout

These examples illustrate the syntax requirements for creating a global index. You must specify the GLOBAL keyword to create a global index.

Illustrate global index syntax

This example illustrates the supported syntax for a global index.

Create a table with three columns (c,b,a) and define a primary key on two of those columns (c,b). Distribute the table by hash on columns (a,b).

Create a global unique range index on columns (c,b).
Command> CREATE TABLE mytab1 (c TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, a TT_INTEGER NOT NULL, 
PRIMARY KEY (c,b)) DISTRIBUTE BY HASH (a,b);
Command> CREATE GLOBAL UNIQUE INDEX mygix1 ON mytab1 (c,b);
Command> indexes mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  MYGIX1: global unique range index on columns:
    C
    B
  MYTAB1: unique range index on columns:
    C
    B
  2 indexes found.

2 indexes found on 1 table.
Create a second global range index and specify the INCLUDE clause.

Command> CREATE GLOBAL INDEX mygix2 ON MYTAB1(b) include (a);
Command> indexes mytab1

Indexes on table SAMPLEUSER.MYTAB1:
  MYGIX1: global unique range index on columns:
    C
    B
  MYTAB1: unique range index on columns:
    C
    B
  MYGIX2: global non-unique range index on columns:
    B
    Included columns:
      A
  3 indexes found.

3 indexes found on 1 table.
Drop the second index and recreate it. Distribute the index by hash on column b.
Command> DROP INDEX mygix2;
Command> CREATE GLOBAL INDEX mygix2 ON MYTAB1(b) INCLUDE (a) DISTRIBUTE BY HASH(b);
Command> INDEXES mytab1

Indexes on table SAMPLEUSER.MYTAB1:
  MYGIX1: global unique range index on columns:
    C
    B
  MYTAB1: unique range index on columns:
    C
    B
  MYGIX2: global non-unique range index on columns:
    B
    Included columns:
      A
  3 indexes found.

3 indexes found on 1 table.

Create a global hash index.

Command> CREATE GLOBAL HASH INDEX mygix3 ON mytab1(a) PAGES =200;
Command> indexes mytab1;

Indexes on table SAMPLEUSER.MYTAB1:
  MYGIX1: global unique range index on columns:
    C
    B
  MYTAB1: unique range index on columns:
    C
    B
  MYGIX3: global non-unique hash index on columns:
    A
  MYGIX2: global non-unique range index on columns:
    B
    Included columns:
      A
  4 indexes found.

4 indexes found on 1 table.

Distribution key of global index is same as distribution key of table

This example illustrates that you cannot create a global index whose distribution key is the same as the distribution key of the table. In this example, the mytab1 table is distributed by hash on columns (a,b). An attempt to create a global index, with columns (a,b) as the distribution key, results in an error.

Command> CREATE TABLE mytab1 (a TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL,
           c TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> CREATE GLOBAL INDEX gix1 ON mytab1(a,b,c) DISTRIBUTE BY HASH (a,b);
 2253: Distribution key for global index cannot be same as that of the table or
 other global index. Consider creating a local index.
The command failed.

Global index creates its own materialized view and its own local index

This example illustrates that when you create a global index, TimesTen Scaleout creates its own internal materialized view and its own local index. Create the mytab2 table distributed by hash on columns (a,b). Create a global non-unique range index distributed by hash on columns (b,a). Run the ttIsql indexes command to show the gix2 global index is created as well as an internal local index on the internal materialized view. Then, run the ttIsql views command to show an internal materialized view is also created as a result of creating the global index. Run the ttIsql describe command to show the internal materialized view. Note that you cannot explicitly drop the internal materialized view or the internal local index.

Command> CREATE TABLE mytab2 (a TT_INTEGER NOT NULL, b TT_INTEGER NOT NULL, 
           c TT_INTEGER NOT NULL) DISTRIBUTE BY HASH (a,b);
Command> CREATE GLOBAL INDEX gix2 ON mytab2(a,b,c) DISTRIBUTE BY HASH (b,a);
Command> indexes;

Indexes on materialized view SAMPLEUSER.$GV9B55D3955D52:
  $GVI9B55D3955D52: non-unique range index on columns:
    A
    B
    C
  1 index found.

Indexes on table SAMPLEUSER.MYTAB2:
  GIX2: global non-unique range index on columns:
    A
    B
    C
  1 index found.

2 indexes found on 2 tables.
Command> views;
  SAMPLEUSER.$GV9B55D3955D52
1 view found.
Command> describe SAMPLEUSER.$GV9B55D3955D52;

Materialized view SAMPLEUSER.$GV9B55D3955D52:
  Global index: GIX2 (table: MYTAB2)
  Columns:
    A                               TT_INTEGER NOT NULL
    B                               TT_INTEGER NOT NULL
    C                               TT_INTEGER NOT NULL
  DISTRIBUTE BY HASH (B, A)

1 view found.

Examples: TimesTen Classic

Create a table and then create a unique hash index on col2. Do not specify the PAGES clause. If PAGES is not specified, the current table page count is used for the size of the hash table. Use INDEXES to verify the index was created. Insert a row in the table, set SHOWPLAN to 1 and then verify the optimizer uses the hash index.

Command> CREATE TABLE tab (col1 NUMBER PRIMARY KEY NOT NULL, col2 VARCHAR2 (30));
Command> CREATE UNIQUE HASH INDEX hash1 ON tab (col2);
Command> INDEXES;
 
Indexes on table TESTUSER.TAB:
  HASH1: unique hash index on columns:
    COL2
  TAB: unique range index on columns:
    COL1
  2 indexes found.
 
2 indexes found on 1 table.
Command> INSERT INTO tab VALUES (10, 'ABC');
Command> SHOWPLAN 1;
Command> SELECT * FROM tab where col2 = 'ABC';
 
Query Optimizer Plan:
 
  STEP:                1
  LEVEL:               1
  OPERATION:           RowLkHashScan
  TBLNAME:             TAB
  IXNAME:              HASH1
  INDEXED CONDITION:   TAB.COL2 = 'ABC'
  NOT INDEXED:         <NULL>
 
< 10, ABC >
1 row found.

Create a table and create a nonunique hash index on col1. Use PAGES = CURRENT to use the current table page count to size the hash index. Use INDEXES to verify the nonunique hash index is created.

Command> CREATE TABLE tab2 (col1 NUMBER);
Command> CREATE HASH INDEX hash_index ON tab2 (col1) PAGES = CURRENT;
Command> INDEXES;
 
Indexes on table TESTUSER.TAB2:
  HASH_INDEX: non-unique hash index on columns:
    COL1
  1 index found.
 
1 index found on 1 table.

Create table and create unique hash index on col3. Use PAGES = 100 to specify a page count of 100 for the size of the hash table. Use INDEXES to verify the unique hash index is created.

Command> CREATE TABLE tab3 (col1 NUMBER, col2 NUMBER, col3 TT_INTEGER);
Command> CREATE UNIQUE HASH INDEX unique_hash1 on tab3 (col3) PAGES = 100;
Command> INDEXES;
 
Indexes on table TESTUSER.TAB3:
  UNIQUE_HASH1: unique hash index on columns:
    COL3
  1 index found.
 
1 index found on 1 table.

The regions table in the HR schema creates a unique index on region_id. Issue the ttIsql INDEXES command on table regions. You see the unique range index regions.

Command> INDEXES REGIONS;

Indexes on table SAMPLEUSER.REGIONS:
  REGIONS: unique range index on columns:
    REGION_ID
    (referenced by foreign key index COUNTR_REG_FK on table SAMPLEUSER.COUNTRIES)
  1 index found.

1 index found on 1 table.

Attempt to create a unique index i on table regions indexing on column region_id. You see a warning message.

Command> CREATE UNIQUE INDEX i ON regions (region_id);
Warning  2232: New index I is identical to existing index REGIONS; 
consider dropping index I

Call ttRedundantIndexCheck to see warning message for this index:

Command> CALL ttRedundantIndexCheck ('regions');
< Index SAMPLEUSER.REGIONS.I is identical to index SAMPLEUSER.REGIONS.REGIONS;
consider dropping index SAMPLEUSER.REGIONS.I >
1 row found.

Create table redundancy and define columns co11 and col2. Create two user indexes on col1 and col2. You see an error message when you attempt to create the second index r2. Index r1 is created. Index r2 is not created.

Command> CREATE TABLE redundancy (col1 CHAR (30), col2 VARCHAR2 (30));
Command> CREATE INDEX r1 ON redundancy (col1, col2);
Command> CREATE INDEX r2 ON redundancy (col1, col2);
 2231: New index R2 would be identical to existing index R1
The command failed.

Issue the ttIsql command INDEXES on table redundancy to show that only index r1 is created:

Command> INDEXES redundancy;

Indexes on table SAMPLEUSER.REDUNDANCY:
  R1: non-unique range index on columns:
    COL1
    COL2
  1 index found.

1 index found on 1 table.

This unique index ensures that all part numbers are unique.

CREATE UNIQUE INDEX purchasing.partnumindex
ON purchasing.parts (partnumber);

Create a linguistic index named german_index on table employees1. To have more than one linguistic sort, create a second linguistic index.

Command> CREATE TABLE employees1 (id CHARACTER (21),
id2 character (21));
Command> CREATE INDEX german_index ON employees1
           (NLSSORT(id, 'NLS_SORT=GERMAN'));
Command> CREATE INDEX german_index2 ON employees1
           (NLSSORT(id2, 'nls_sort=german_ci'));
Command> indexes employees1;
Indexes on table SAMPLEUSER.EMPLOYEES1:
  GERMAN_INDEX: non-unique range index on columns:
    NLSSORT(ID,'NLS_SORT=GERMAN')
  GERMAN_INDEX2: non-unique range index on columns:
    NLSSORT(ID2,'nls_sort=german_ci')
  2 indexes found.
1 table found.

See also

DROP INDEX

CREATE MATERIALIZED VIEW

The CREATE MATERIALIZED VIEW statement creates a view of the table specified in the SelectQuery clause. The original tables used to create a view are referred to as detail tables. The view is refreshed synchronously with regard to changes in the detail tables.

Required privileges

User executing the statement must have CREATE MATERIALIZED VIEW (if owner) or CREATE ANY MATERIALIZED VIEW (if not owner) privilege.

Owner of the materialized view must have:
  • SELECT privilege on the detail tables.
  • CREATE TABLE privilege.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout. You must specify the DISTRIBUTE BY HASH clause and you must define a distribution key. The DISTRIBUTE BY REFERENCE and DUPLICATE clauses are not supported.

SQL syntax: TimesTen Scaleout

CREATE MATERIALIZED VIEW [Owner.]ViewName
DISTRIBUTE BY HASH (ColumnName [,...])
AS SelectQuery
[PRIMARY KEY (ColumnName [,...])] 
[UNIQUE HASH ON (HashColumnName [,...]) PAGES = PrimaryPages]

SQL syntax: TimesTen Classic

CREATE MATERIALIZED VIEW [Owner.]ViewName
AS SelectQuery
[PRIMARY KEY (ColumnName [,...])] 
[UNIQUE HASH ON (HashColumnName [,...]) PAGES = PrimaryPages]

Parameters

Parameter Description

[Owner.]ViewName

Name assigned to the new view.

DISTRIBUTE BY HASH (ColumnName [,...])

TimesTen Scaleout only. You must specify the DISTRIBUTE BY HASH clause and you must specify one or more columns for the distribution key (even if you have specified a primary key).

The detail table must be distributed by hash.

DISTRIBUTE BY REFERENCE or DUPLICATE clauses are not supported.

This clause must appear before the AS SelectQuery clause.

SelectQuery

Select column from the detail tables to be used in the view.

ColumnName

Name of the column(s) that forms the primary key for the view to be created. Up to 32 columns can be specified for the primary key. Each result column name of a viewed table must be unique. The column name definition cannot contain the table or owner component.

UNIQUE HASH ON

Hash index for the table. Only unique hash indexes are created. This parameter is used for equality predicates. UNIQUE HASH ON requires that a primary key be defined.

HashColumnName

Column defined in the view that is to participate in the hash key of this table. The columns specified in the hash index must be identical to the columns in the primary key.

PAGES = PrimaryPages

Sizes the hash index to reflect the expected number of pages in your table. To determine the value for PrimaryPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for PrimaryPages (256000/256=1000).

The value for PrimaryPages must be a positive constant and must be greater than 0.

If your estimate for PrimaryPages is too small, performance may be degraded.

See "CREATE TABLE" for information on hash indexes.

Description and restrictions for CREATE MATERIALIZED VIEW: TimesTen Scaleout

Description and restrictions include:

  • The SQL optimizer may re-write a query against a base table to use an available materialized view if the use of the materialized view is expected to improve the execution time of the query.

  • You must specify the DISTRIBUTE BY HASH clause and you must specify it with a distribution key (even if you have specified a primary key and intend to use the primary key as the distribution key).

  • You must specify the DISTRIBUTE BY HASH clause before the AS SelectQuery clause.

  • You can only specify the DISTRIBUTE BY HASH clause. The DISTRIBUTE BY REFERENCE and DUPLICATE clauses are not supported.

  • The SelectQuery must be restricted to single table SELECT statements.

  • You cannot specify the GROUP BY or the WHERE clause in the SelectQuery.

  • You cannot use SQL functions in the SelectQuery.

  • You cannot use an expression in the SelectQuery.

  • The detail table of the materialized view cannot have a foreign key with a cascade delete clause.

  • The distribution key columns must be in the project list of the SelectQuery.

  • There are no DDL rewrites. For example, if you create a unique index on the detail table, a corresponding index on the materialized view (which is distributed on the unique column) is not created.

Description: TimesTen Scaleout and TimesTen Classic

The restrictions and requirements on the defining query include:

  • Each expression in the select list must have a unique name.

  • Do not use non-materialized views to define a materialized view.

  • Do not define CLOB, BLOB, or NCLOB data types for columns in the select list of the materialized view query.

  • The detail tables cannot belong to a cache group and the detail tables cannot have compression.

  • Do not use SELECT FOR UPDATE.

  • Do not reference system tables or views.

  • Do not use nested definitions for a materialized view.

  • Do not use dynamic parameters.

  • Do not use ROWNUM.

  • Do not use analytic functions.

  • Do not use GROUPING SETS, ROLLUP, or CUBE.

  • Do not use the SYSDATE function.

  • Do not use the functions SYSTEM_USER, USER, CURRENT_USER, or SESSION_USER.

  • Do not use NEXTVAL or CURRVAL.

  • Outer joins are allowed but the select list must project at least one non-nullable column from each of the inner tables specified in the outer join.

  • Do not use the WITH subquery clause.

The restrictions (not on the defining query) include:

  • Do not have a hash-based primary key that contains any aggregate columns of the materialized view.

  • A materialized view cannot be replicated directly using TimesTen replication. You can replicate the detail tables. You must define the same materialized view on both sides of replication. TimesTen automatically updates the corresponding materialized views.

  • You cannot define a foreign key if the referencing or referenced table is a materialized view.

The following restrictions and requirements on the defining query are:

  • The view definition must include all columns in the group by list in the select list.

  • An aggregate view must include a COUNT (*) or COUNT (non-nullable column) in the select list.

  • Do not use derived tables or JOIN tables.

  • Do not use SELECT DISTINCT or an aggregate distinct function.

  • Do not use the set operators UNION, MINUS, or INTERSECT.

  • Do not use SUM of nullable expressions.

  • Use only simple columns as group by columns.

  • Group by columns cannot belong to self join tables.

  • Do not use these clauses:

    • HAVING

    • ORDER BY

    • DISTINCT

    • FIRST

    • JOIN

  • Do not use the TT_HASH function.

  • You can use SUM and COUNT but do not use expressions involving SUM and COUNT. Do not use AVG, which is treated as SUM/COUNT.

  • Do not specify MIN or MAX functions in the select list.

  • For joins:

    • Join predicates cannot have an OR.

    • Do not specify Cartesian product joins (joins with no join predicate).

    • For outer joins, outer join each inner table with at most one table.

Additional considerations include:

  • A materialized view is read-only and cannot be updated directly. A materialized view is updated only when changes are made to the associated detail tables. Therefore a materialized view cannot be the target of a DELETE, UPDATE or INSERT statement.

  • By default, a range index is created to enforce the primary key for a materialized view. Alternatively, use the UNIQUE HASH clause to specify a hash index for the primary key.

    • If your application performs range queries over a materialized view's primary key, then choose a range index for that view by omitting the UNIQUE HASH clause.

    • If your application performs only exact match lookups on the primary key, then a hash index may offer better response time and throughput. In such a case, specify the UNIQUE HASH clause. See "CREATE TABLE" for more information about the UNIQUE HASH clause.

  • You can use ALTER TABLE to change the representation of the primary key index or resize a hash index of a materialized view.

  • You cannot add or drop columns in the materialized view with the ALTER TABLE statement. To change the structure of the materialized view, drop and recreate the view.

  • You can create indexes on the materialized view with the CREATE INDEX SQL statement.

The owner of a materialized view must have the SELECT privilege on its detail tables. The SELECT privilege is implied by the SELECT ANY TABLE and ADMIN system privileges. When the SELECT privilege or a higher-level system privilege on the detail tables is revoked from the owner of the materialized view, the materialized view becomes invalid.

Selecting from an invalid materialized view fails with an error. Updates to the detail tables of an invalid materialized view do not update the materialized view.

You can identify invalid materialized views by using the ttIsql describe command and by inspecting the STATUS column of the SYS.DBA_OBJECTS, SYS.ALL_OBJECTS or SYS.USER_OBJECTS system tables. See Oracle TimesTen In-Memory Database System Tables and Views Reference.

If the revoked privilege is restored, you can make an invalid materialized view valid again by dropping and recreating the materialized view.

For more information, see "Object privileges for materialized views" in Oracle TimesTen In-Memory Database Security Guide.

Examples for CREATE MATERIALIZED VIEW: TimesTen Scaleout

Syntax example:

Command> CREATE MATERIALIZED VIEW mv
             DISTRIBUTE BY HASH (phone)
             AS SELECT phone FROM accounts;
1010 rows materialized.

Examples: TimesTen Classic

Create a materialized view of columns from the customer and bookorder tables.

CREATE MATERIALIZED VIEW custorder AS
  SELECT custno, custname, ordno, book
  FROM customer, bookorder
  WHERE customer.custno=bookorder.custno;

Create a materialized view of columns x1 and y1 from the t1 table.

CREATE MATERIALIZED VIEW v1 AS SELECT x1, y1 FROM t1
  PRIMARY KEY (x1) UNIQUE HASH ON (x1) PAGES=100;

Create a materialized view from an outer join of columns x1 and y1 from the t1 and t2 tables.

CREATE MATERIALIZED VIEW v2 AS SELECT x1, y1 FROM t1, t2
  WHERE x1=x2(+);

The following example creates a materialized view empmatview2 based on selected columns employee_id and email from table employees. After the materialized view is created, create an index on the materialized view column mvemp_id of the materialized view empmatview2.

CREATE MATERIALIZED VIEW empmatview2
   AS SELECT employee_id mvemp_id, email mvemail 
         FROM employees;
107 rows materialized. 

CREATE INDEX empmvindex ON empmatview2 (mvemp_id);

CREATE PACKAGE

The CREATE PACKAGE statement creates the specification for a standalone package, which is an encapsulated collection of related procedures, functions, and other program objects stored together in your database. The package specification declares these objects. The package body defines these objects.

Required privilege

CREATE PROCEDURE (if owner) or CREATE ANY PROCEDURE (if not owner).

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

CREATE [OR REPLACE] PACKAGE [Owner.]PackageName
      [InvokerRightsClause] [AccessibleByClause]
      {IS|AS}
      PlsqlPackageSpec

InvokerRightsClause::=
AUTHID {CURRENT_USER | DEFINER}

AccessibleByClause::=
ACCESSIBLE BY (accessor[,...])

accessor::= 
[UnitKind][Owner.]UnitName

You can specify InvokerRightsClause or AccessibleByClause in any order.

Parameters

Parameter Description

OR REPLACE

Specify OR REPLACE to recreate the package specification if it already exists. Use this clause to change the specification of an existing package without dropping and recreating the package. When you change a package specification, TimesTen recompiles it.

PackageName

Name of the package.

InvokerRightsClause

Lets you specify whether the SQL statements in PL/SQL functions or procedures execute with definer's or invoker's rights. The AUTHID setting affects the name resolution and privilege checking of SQL statements that a PL/SQL procedure or function issues at runtime, as follows:

  • Specify DEFINER so that SQL name resolution and privilege checking operate as though the owner of the procedure or function (the definer, in whose schema it resides) is running it. DEFINER is the default.

  • Specify CURRENT_USER so that SQL name resolution and privilege checking operate as though the current user (the invoker) is running it.

For more information, see "Definer's rights and invoker's rights (AUTHID clause)" in the Oracle TimesTen In-Memory Database Security Guide.

AccessibleByClause

Use this clause to specify one or more accessors (PL/SQL units) that can invoke the package directly. The list of accessors that can access the package is called a white list. A white list gives you the ability to add an extra layer of security to your PL/SQL objects. Specifically, you can restrict access to the package to only those objects on the white list.

AccessibleByClause can appear only once in the CREATE PACKAGE statement.

Syntax: ACCESSIBLE BY (accessor [,...])

accessor

Used in AccessibleByClause. An accessor is a PL/SQL unit that can invoke the package.

An accessor can appear more than once in the AccessibleByClause clause.

Syntax: [UnitKind][Owner.]UnitName

UnitKind

Used in the accessor clause (which is part of the AccessibleByClause clause). Specifies the kind of PL/SQL unit that can invoke the package.

  • UnitKind is optional, but if specified, valid options are:
  • FUNCTION

  • PROCEDURE

  • PACKAGE

[Owner.]UnitName

Used in the accessor clause (which is part of the AccessibleByClause clause). Specifies the name of the PL/SQL unit that can invoke the package. If you specify UnitKind, then UnitName must be a name of a unit of that kind. For example, if you specify PROCEDURE for UnitKind, then UnitName must be the name of a procedure. UnitName is required.

You can optionally specify Owner. If you specify Owner, then UnitName must reside in that owner's schema. If you do not specify Owner, UnitName must be in the schema that contains the package.

IS|AS

Specify either IS or AS to declare the body of the function.

PlsqlPackageSpec

Specifies the package specification. Can include type definitions, cursor declarations, variable declarations, constant declarations, exception declarations and PL/SQL subprogram declarations.

Description

  • AccessibleByClause:

    • AccessibleByClause is valid at the top-level package definition. You cannot specify AccessibleByClause in the individual procedures or functions within the package. In addition, you cannot specify AccessibleByClause in the CREATE PACKAGE BODY statement.

    • You can use this clause to restrict access to helper packages. For example, assume your PL/SQL package defines an API for a given functionality and that functionality is implemented using a set of helper procedures and functions. You want to limit applications to only be able to call the API procedure or function that is defined in your package, and to not be able to call the helper procedures and functions directly. You can use the ACCESSIBLE BY clause to achieve this.

    • The compiler checks the validity of the syntax of the ACCESSIBLE BY clause, but does not check that the accessor exists. Therefore, you can define an accessor that does yet exist in the owner's schema.

    • When you invoke the package, the compiler first does the normal permission checks on the invocation. If any check fails, the invocation fails, even if the invoker is an accessor. If all normal permission checks on the invocation succeed, and the package has no ACCESSIBLE BY clause, the invocation succeeds. If the package has an ACCESSIBLE BY clause, the invocation succeeds only if the invoker is an accessor.

  • When you create or replace a package, the privileges granted on the package remain the same. If you drop and recreate the object, the object privileges that were granted on the original object are revoked.

  • In a replicated environment, the CREATE PACKAGE statement is not replicated. For more information, see "Creating a new PL/SQL object in an existing active standby pair" and "Adding a PL/SQL object to an existing classic replication scheme" in the Oracle TimesTen In-Memory Database Replication Guide.

Examples

Illustrating the correct usage of the AccessibleByClause

This example illustrates the correct usage of the AccessibleByClause. The clause is specified at the top-level of the CREATE PACKAGE statement. Note that the CallingProc procedure does not need to exist.

Command> CREATE OR REPLACE PACKAGE ProtectedPkg
           ACCESSIBLE BY (PROCEDURE CallingProc)
         AS
           PROCEDURE ProtectedProc;
         END;
         /
 
Package created.

Illustrating the incorrect usage of the AccessibleByClause

These examples show the incorrect use of the AccessibleByClause. The first example attempts to use AccessibleByClause in the packaged procedure, resulting in a compilation error. The second example attempts to use AccessibleByClause in the CREATE PACKAGE BODY statement, resulting in a compilation error.

This example uses the ACCESSIBLE BY clause in the packaged procedure.

Command> CREATE OR REPLACE PACKAGE ProtectedPkg1
         AS
           PROCEDURE ProtectedProc1 
           ACCESSIBLE BY (PROCEDURE CallingProc)
         END;
         /
         
Warning: Package created with compilation errors.
 
Command> SHOW ERRORS
Errors for PACKAGE PROTECTEDPKG1:
 
LINE/COL ERROR
-------- -----------------------------------------------------------------
0/0      PLS-00157: Only schema-level programs allow ACCESSIBLE BY

This example uses the ACCESSIBLE BY clause in the CREATE PACKAGE BODY statement.

Command> CREATE OR REPLACE PACKAGE ProtectedPkg3
           ACCESSIBLE BY (PROCEDURE CallingProc3)
         AS
           PROCEDURE ProtectedProc3;
         END;
         /
        
Package created.

Command> CREATE OR REPLACE PACKAGE BODY ProtectedPkg3
           ACCESSIBLE BY (PROCEDURE CallingProc3)
         AS
           PROCEDURE ProtectedProc3 AS
           BEGIN
             NULL;
           END;
           ;
         /
 
Warning: Package body created with compilation errors.
 
Command> SHOW ERRORS
Errors for PACKAGE BODY PROTECTEDPKG3:
 
LINE/COL ERROR
-------- -----------------------------------------------------------------
2/1      PLS-00103: Encountered the symbol "ACCESSIBLE" when expecting one of the
following:
 
  is as compress compiled wrapped

Ensuring only the API can access the helper package

This example walks through a series of steps to illustrate the use of the AccessibleByClause. The example creates the SampleAPI package and the SampleHelper package. The ACCESSIBLE BY clause is specified on the SampleHelper to ensure that only the SampleAPI package can access the SampleHelper package.

Steps:

  1. Create the SampleHelper package. Specify the ACCESSIBLE BY clause, giving the SampleAPI package access to the SampleHelper package. The SampleAPI package is in the white list.

    Command> CREATE OR REPLACE PACKAGE SampleHelper
               ACCESSIBLE BY (SampleAPI)
             AS
               PROCEDURE SampleH1;
               PROCEDURE SampleH2;
             END;
             /
     
    Package created.
    
  2. Create the SampleHelper package body.

    Command> CREATE OR REPLACE PACKAGE BODY SampleHelper
             AS
               PROCEDURE SampleH1 AS
               BEGIN
                 DBMS_OUTPUT.PUT_LINE('Sample helper procedure SampleH1');
               END;
               PROCEDURE SampleH2 AS
               BEGIN
                 DBMS_OUTPUT.PUT_LINE('Sample helper procedure SampleH2');
               END;
             END;
             /
     
    Package body created.
    
  3. Create the SampleAPI package.

    Command> CREATE OR REPLACE PACKAGE SampleAPI
             AS
               PROCEDURE p1;
               PROCEDURE p2;
             END;
             /
     
    Package created.
    
  4. Create the SampleAPI package body. The p1 procedure references the SampleHelper.SampleH1 procedure. The p2 procedure references the SampleHelper.SampleH2 procedure.

    Command> CREATE OR REPLACE PACKAGE BODY SampleAPI
             AS
               PROCEDURE p1 AS
               BEGIN
                 DBMS_OUTPUT.PUT_LINE('SampleAPI procedure p1');
                 SampleHelper.SampleH1;
               END;
               PROCEDURE p2 AS
               BEGIN
                 DBMS_OUTPUT.PUT_LINE('SampleAPI procedure p2');
                 SampleHelper.SampleH2;
               END;
             END;
             /
     
    Package body created.
    
  5. Call the SampleAPI.p1 and the SampleAPI.p2 procedures. The SampleAPI package is in the white list of the SampleHelper package, resulting in successful execution.

    Command> SET SERVEROUTPUT ON
    Command> BEGIN
               SampleAPI.p1;
               SampleAPI.p2;
             END;
             /
    SampleAPI procedure p1
    Sample helper procedure SampleH1
    SampleAPI procedure p2
    Sample helper procedure SampleH2
     
    PL/SQL procedure successfully completed.
    
  6. Call the SampleHelper.SampleH1 procedure directly. An error is returned due to insufficient access privileges.

    Command> BEGIN
               SampleHelper.SampleH1;
             END;
            /
     8503: ORA-06550: line 2, column 3:
    PLS-00904: insufficient privilege to access object SAMPLEHELPER
     8503: ORA-06550: line 2, column 3:
    PL/SQL: Statement ignored
    The command failed.

CREATE PACKAGE BODY

The CREATE PACKAGE BODY statement creates the body of a standalone package. A package is an encapsulated collection of related procedures, functions, and other program objects stored together in your database. A package specification declares these objects. A package body defines these objects.

Required privilege

CREATE PROCEDURE (if owner) or CREATE ANY PROCEDURE (if not owner).

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

CREATE [OR REPLACE] PACKAGE BODY [Owner.]PackageBody 
      {IS|AS} plsql_package_body

Parameters

Parameter Description

OR REPLACE

Specify OR REPLACE to recreate the package body if it already exists. Use this clause to change the body of an existing package without dropping and recreating it. When you change a package body, TimesTen recompiles it.

PackageBody

Name of the package body.

IS|AS

Specify either IS or AS to declare the body of the function.

plsql_package_body

Specifies the package body which consists of PL/SQL subprograms.

Description

In a replicated environment, the CREATE PACKAGE BODY statement is not replicated. For more information, see "Creating a new PL/SQL object in an existing active standby pair" and "Adding a PL/SQL object to an existing classic replication scheme" in the Oracle TimesTen In-Memory Database Replication Guide.

When you create or replace a package body, the privileges granted on the package body remain the same. If you drop and recreate the object, the object privileges that were granted on the original object are revoked.

CREATE PROCEDURE

The CREATE PROCEDURE statement creates a standalone stored procedure.

Required privilege

CREATE PROCEDURE (if owner) or CREATE ANY PROCEDURE (if not owner).

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

CREATE [OR REPLACE] PROCEDURE [Owner.]ProcedureName 
     [(arguments [IN|OUT|IN OUT][NOCOPY] DataType [DEFAULT expr][,...])]
     [InvokerRightsClause][AccessibleByClause] [DETERMINISTIC]
     {IS|AS} plsql_procedure_body

InvokerRightsClause::=
AUTHID {CURRENT_USER|DEFINER}

AccessibleByClause::=
ACCESSIBLE BY(accessor[,...])

accessor::= 
[UnitKind][Owner.]UnitName

You can specify InvokerRightsClause, AccessibleByClause, or DETERMINISTIC in any order.

Parameters

Parameter Description

OR REPLACE

Specify OR REPLACE to recreate the procedure if it already exists. Use this clause to change the definition of an existing procedure without dropping and recreating it. When you recreate a procedure, TimesTen recompiles it.

ProcedureName

Name of procedure.

arguments

Name of argument/parameter. You can specify 0 or more parameters for the procedure. If you specify a parameter, you must specify a data type for the parameter. The data type must be a PL/SQL data type.

[IN|OUT|IN OUT]

Parameter modes.

IN is a read-only parameter. You can pass the parameter's value into the procedure but the procedure cannot pass the parameter's value out of the procedure and back to the calling PL/SQL block.The value of the parameter cannot be changed.

OUT is a write-only parameter. Use an OUT parameter to pass a value back from the procedure to the calling PL/SQL block. You can assign a value to the parameter.

IN OUT is a read/write parameter. You can pass values into the procedure and return a value back to the calling program (either the original, unchanged value or a new value set within the procedure.

IN is the default.

NOCOPY

Specify NOCOPY to instruct TimesTen to pass the parameter as fast as possible. Can enhance performance when passing a large value such as a record, an index-by-table, or a varray to an OUT or IN OUT parameter. IN parameters are always passed NOCOPY.

DEFAULT expr

Use this clause to specify a DEFAULT value for the parameter. You can specify := in place of the keyword DEFAULT.

InvokerRightsClause

Lets you specify whether the SQL statements in PL/SQL functions or procedures execute with definer's or invoker's rights. The AUTHID setting affects the name resolution and privilege checking of SQL statements that a PL/SQL procedure or function issues at runtime, as follows:

  • Specify DEFINER so that SQL name resolution and privilege checking operate as though the owner of the procedure or function (the definer, in whose schema it resides) is running it. DEFINER is the default.

  • Specify CURRENT_USER so that SQL name resolution and privilege checking operate as though the current user (the invoker) is running it.

For more information, see "Definer's rights and invoker's rights (AUTHID clause)" in the Oracle TimesTen In-Memory Database Security Guide.

AccessibleByClause

Use this clause to specify one or more accessors (PL/SQL units) that can invoke the procedure directly. The list of accessors that can access the procedure is called a white list. A white list gives you the ability to add an extra layer of security to your PL/SQL objects. Specifically, you can restrict access to the procedure to only those objects on the white list.

The AccessibleByClause can appear only once in the CREATE PROCEDURE statement.

Syntax: ACCESSIBLE BY (accessor [,...])

accessor

Used in the AccessibleByClause. An accessor is a PL/SQL unit that can invoke the procedure.

An accessor can appear more than once in the AccessibleByClause.

Syntax: [UnitKind][Owner.]UnitName

UnitKind

Used in the accessor clause (which is part of the AccessibleByClause clause). Specifies the kind of PL/SQL unit that can invoke the procedure.

  • UnitKind is optional, but if specified, valid options are:
  • FUNCTION

  • PROCEDURE

  • PACKAGE

[Owner.]UnitName

Used in the accessor clause (which is part of the AccessibleByClause clause). Specifies the name of the PL/SQL unit that can invoke the procedure. If you specify UnitKind, then UnitName must be a name of a unit of that kind. For example, if you specify PROCEDURE for UnitKind, then UnitName must be the name of a procedure. UnitName is required.

You can optionally specify Owner. If you specify Owner, then UnitName must reside in that owner's schema. If you do not specify Owner, UnitName must be in the schema that contains the procedure.

DETERMINISTIC

Specify DETERMINISTIC to indicate that the procedure returns the same result value whenever it is called with the same values for its parameters.

IS|AS

Specify either IS or AS to declare the body of the procedure.

plsql_procedure_body

Specifies the procedure body.

Description

  • AccessibleByClause:

    • The compiler checks the validity of the syntax of the AccessibleByClause, but does not check that the accessor exists. Therefore, you can define an accessor that does yet exist in the owner's schema.

    • When you invoke the procedure, the compiler first does the normal permission checks on the invocation. If any check fails, the invocation fails, even if the invoker is an accessor. If all normal permission checks on the invocation succeed, and the procedure has no AccessibleByClause, the invocation succeeds. If the procedure has an AccessibleByClause, the invocation succeeds only if the invoker is an accessor.

  • When you create or replace a procedure, the privileges granted on the procedure remain the same. If you drop and recreate the object, the object privileges that were granted on the original object are revoked.

  • The namespace for PL/SQL procedures is distinct from the TimesTen built-in procedures. You can create a PL/SQL procedure with the same name as a TimesTen built-in procedure.

  • TimesTen does not support:

    • call_spec clause

    • AS EXTERNAL clause

  • In a replicated environment, the CREATE PROCEDURE statement is not replicated. For more information, see "Creating a new PL/SQL object in an existing active standby pair" and "Adding a PL/SQL object to an existing classic replication scheme" in the Oracle TimesTen In-Memory Database Replication Guide.

Examples

Using the AccessibleByClause

This example creates the ProtectedProc procedure and uses the ACCESSIBLE BY clause to restrict access to the CallingProc procedure. The CallingProc procedure does not yet exist. The example then creates the CallingProc procedure, which calls the ProtectedProc procedure. The CallingProc procedure is successfully created, as it is specified in the ACCESSIBLE BY clause. The example then attempts to call the ProtectedProc procedure directly, resulting in an error. It concludes with attempting to create the AnotherCallingProc procedure that references the ProtectedProc procedure, but the AnotherCallingProc procedure is not in the white list. A compilation error results.

Steps to illustrate the example:

  1. Create the ProtectedProc procedure, specifying the ACCESSIBLE BY clause. The CallingProc procedure is in the white list. It does not yet exist.

    Command> CREATE OR REPLACE PROCEDURE ProtectedProc
               ACCESSIBLE BY (CallingProc)
             AS
             BEGIN
               DBMS_OUTPUT.PUT_LINE ('ProtectedProc');
             END;
             /
              
    Procedure created.
    
  2. Create the CallingProc procedure, referencing the ProtectedProc procedure.

    Command> CREATE OR REPLACE PROCEDURE CallingProc
             AS
             BEGIN
               DBMS_OUTPUT.PUT_LINE ('CallingProc');
               ProtectedProc;
             END;
             /
             
    Procedure created.
    
  3. Call the CallingProc procedure. The procedure is successfully executed.

    Command> SET SERVEROUTPUT ON
    Command> exec CallingProc;
    CallingProc
    ProtectedProc
     
    PL/SQL procedure successfully completed.
    
  4. Attempt to call the ProtectedProc procedure directly. An error is thrown due to insufficient access privileges.

    Command> exec ProtectedProc;
     8503: ORA-06550: line 1, column 7:
    PLS-00904: insufficient privilege to access object PROTECTEDPROC
     8503: ORA-06550: line 1, column 7:
    PL/SQL: Statement ignored
    The command failed.
    
  5. Create the AnotherCallingProc procedure that references the ProtectedProc procedure. The AnotherCallingProc is not in the white list (not listed in the ACCESSIBLE BY clause of ProtectedProc), resulting in a compilation error.

    Command> CREATE OR REPLACE PROCEDURE AnotherCallingProc
             AS
             BEGIN
               DBMS_OUTPUT.PUT_LINE ('AnotherCallingProc');
               ProtectedProc;
             END;
             /
     
    Warning: Procedure created with compilation errors.
     
    Command> SHOW ERRORS
    Errors for PROCEDURE ANOTHERCALLINGPROC:
     
    LINE/COL ERROR
    -------- -----------------------------------------------------------------
    5/1      PL/SQL: Statement ignored
    5/1      PLS-00904: insufficient privilege to access object PROTECTEDPROC
    

Using the accessor clause

This example illustrates the uses of the accessor clause through a sequence of steps.

  1. Create the SampleUser1 and SampleUser2 users and grant ADMIN privileges to both users.

    Command> CREATE USER SampleUser1 IDENTIFIED BY SampleUser1;
     
    User created.
     
    Command> CREATE USER SampleUser2 IDENTIFIED BY SampleUser2;
     
    User created.
    Command> GRANT ADMIN TO SampleUser1, SampleUser2;
    
  2. Create the SampleUser1.ProtectedProc procedure, specifying the ACCESSIBLE BY clause. The CallingProc procedure is specified in the white list without an owner. The owner of the CallingProc procedure is assumed to be in the same schema as the owner of the procedure with the ACCESSIBLE BY clause. Thus, CallingProc is assumed to be in the SampleUser1 schema.

    Command> CREATE OR REPLACE PROCEDURE SampleUser1.ProtectedProc
             ACCESSIBLE BY (CallingProc)
             AS
             BEGIN
               DBMS_OUTPUT.PUT_LINE ('SampleUser1 ProtectedProc');
             END;
             /
     
    Procedure created.
    
  3. Connect as SampleUser1. Create the CallingProc procedure, referencing the SampleUser1.ProtectedProc procedure.

    Command> Connect adding "uid=SampleUser1;pwd=SampleUser1PW" as SampleUser1;
    Connection successful:
    DSN=database1;UID=SampleUser1;DataStore=/scratch/sampleuser1/database1;
    DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;
    PermSize=128;
    (Default setting AutoCommit=1)
    
    sampleuser1: Command> CREATE OR REPLACE PROCEDURE CallingProc
                          AS
                          BEGIN
                            DBMS_OUTPUT.PUT_LINE ('SampleUser1 CallingProc');
                            ProtectedProc;
                          END;
                          /
     
    Procedure created.
    
  4. From the SampleUser1 connection, call the CallingProc procedure. The call succeeds.

    sampleuser1: Command> SET SERVEROUTPUT ON
    sampleuser1: Command> exec CallingProc;
    SampleUser1 CallingProc
    SampleUser1 ProtectedProc
     
    PL/SQL procedure successfully completed.
    
  5. Connect to SampleUser2. Create the CallingProc procedure, referencing the SampleUser1.ProtectedProc procedure. A compilation error results.

    SampleUser1: Command> connect adding "uid=Sampleuser2;pwd=SampleUser2PW"
                          as SampleUser2;
    Connection successful:
    DSN=database1;UID=Sampleuser2;DataStore=/scratch/sampleuser2/database1;
    DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;
    PermSize=128;
    (Default setting AutoCommit=1)
    
    sampleuser2: Command> CREATE OR REPLACE PROCEDURE CallingProc
                          AS
                          BEGIN
                            DBMS_OUTPUT.PUT_LINE ('SampleUser2 CallingProc');
                            SampleUser1.ProtectedProc;
                          END;
                          /
     
    Warning: Procedure created with compilation errors.
     
    sampleuser2: Command> SHOW ERRORS
    Errors for PROCEDURE CALLINGPROC:
     
    LINE/COL ERROR
    -------- -----------------------------------------------------------------
    5/1      PL/SQL: Statement ignored
    5/1      PLS-00904: insufficient privilege to access object PROTECTEDPROC
    
  6. Switch to the SampleUser1 connection. Recreate the ProtectedProc procedure.

    sampleuser2: Command> use SampleUser1
    sampleuser1: Command> CREATE OR REPLACE PROCEDURE ProtectedProc
                          ACCESSIBLE BY (CallingProc, SampleUser2.CallingProc)
                          AS
                          BEGIN
                            DBMS_OUTPUT.PUT_LINE ('SampleUser1 ProtectedProc');
                          END;
                          /
     
    Procedure created.
    
  7. From the SampleUser2 connection, call the CallingProc procedure. The SampleUser2.CallingProc is in the white list of the SampleUser1.ProtectedProc procedure, resulting in successful execution.

    sampleuser1: Command> use SampleUser2;
    sampleuser2: Command> SET SERVEROUTPUT ON
    sampleuser2: Command> exec CallingProc
    SampleUser2 CallingProc
    SampleUser1 ProtectedProc
     
    PL/SQL procedure successfully completed.

Using the CREATE PROCEDURE statement to retrieve information

Create a procedure query_emp to retrieve information about an employee. Pass the employee_id 171 to the procedure and retrieve the last_name and salary into two OUT parameters.

Command> CREATE OR REPLACE PROCEDURE query_emp
                   (p_id IN employees.employee_id%TYPE,
                    p_name  OUT employees.last_name%TYPE,
                    p_salary OUT employees.salary%TYPE) IS
                 BEGIN
                   SELECT last_name, salary INTO p_name, p_salary
                   FROM employees
                   WHERE employee_id = p_id;
                 END query_emp;
                 /
 
Procedure created.

CREATE PROFILE

The CREATE PROFILE statement creates a profile, which is a set of limits on the database resources. If you assign a profile to a user, that user cannot exceed the limits specified in the profile.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

CREATE PROFILE profile LIMIT password_parameters

password_parameters::=
[FAILED_LOGIN_ATTEMPTS password_parameter_options]
[PASSWORD_LIFE_TIME password_parameter_options]
[PASSWORD_REUSE_TIME password_parameter_options]
[PASSWORD_REUSE_MAX password_parameter_options]
[PASSWORD_LOCK_TIME password_parameter_options]
[PASSWORD_GRACE_TIME password_parameter_options]
[{PASSWORD_COMPLEXITY_CHECKER|PASSWORD_VERIFY_FUNCTION} password_checker_options]

password_parameter_options::=
UNLIMITED|DEFAULT|constant

password_checker_options::=
function|NULL|DEFAULT

function::
TT_VERIFY_FUNCTION|TT_STRONG_VERIFY_FUNCTION|TT_STIG_VERIFY_FUNCTION

Parameters

Parameter Description

profile

Name of the profile.

LIMIT password_parameters

The LIMIT clause sets the limits for the password parameters. The LIMIT keyword is required.

The password parameters consist of the name of the password parameter and the value (or limit) for the password parameter. This includes the password complexity checker functions. All the parameters (with the exception of FAILED_LOGIN_ATTEMPTS and PASSWORD_REUSE_MAX) set lengths of time and are interpreted in number of days. You can use a decimal value (for example, you can use .0833 to denote approximately one hour). The minimum value is 1 second. The maximum value is 106,751,991 days. The constant value must be expressed in days. For example, to set a value of 5 minutes, specify the constant value of 0.0034722222222222 (5/1440 days). For FAILED_LOGIN_ATTEMPTS and PASSWORD_REUSE_MAX, you must specify an integer.

If you do not specify a password parameter after the LIMIT clause, the limit for that password parameter is based on the limit defined in the DEFAULT profile. In addition, if you only specify the LIMIT keyword with no additional parameters, the limits for the profile are based on the limits of the DEFAULT profile.

FAILED_LOGIN_ATTEMPTS

Specifies the number of consecutive failed attempts to connect to the database by a user before that user's account is locked.

PASSWORD_LIFE_TIME

Specifies the number of days that a user can use the same password for authentication. If you also set a value for PASSWORD_GRACE_TIME, then the password expires if it is not changed within the grace period. In such a situation, future connections to the database are rejected.

PASSWORD_REUSE_TIME and PASSWORD_REUSE_MAX

These two parameters must be used together.

  • PASSWORD_REUSE_TIME specifies the number of days that must pass before a user can reuse a password. For example, if you specify a value of 30, then after 30 days the user can reuse a previous password.

  • PASSWORD_REUSE_MAX specifies the number of password changes that are required before the current password can be reused.

You must specify a value for both parameters for them to have any effect. Specifically:

  • If you specify a value for both parameters: A user cannot reuse a password until the password has been changed the number of times specified for PASSWORD_REUSE_MAX during the number of days specified for PASSWORD_REUSE_TIME. For example, if you specify a value of 30 for PASSWORD_REUSE_TIME and a value of 10 for PASSWORD_REUSE_MAX, then the user can reuse the password after 30 days if the password has been changed 10 times.

  • If you specify a value for one parameter and specify a value of UNLIMITED for the second parameter, then the user can never reuse a password.

  • If you specify a value of UNLIMITED for both parameters, then TimesTen ignores both values, indicating that the password can be reused.

PASSWORD_LOCK_TIME

Specifies the number of days the user account is locked after the specified number of consecutive failed connection attempts.

PASSWORD_GRACE_TIME

Specifies the number of days after the grace period begins during which TimesTen issues a warning, but allows the connection to the database. If the password is not changed during the grace period, the password expires. This parameter is associated with the PASSWORD_LIFE_TIME parameter.

UNLIMITED

Indicates that there is no limit for the password parameter. If you specify UNLIMITED, it must follow the password parameter. For example, FAILED_LOGIN_ATTEMPTS UNLIMITED.

DEFAULT

Indicates that you want to omit a limit for the password parameter in this profile. A user that is assigned this profile is subject to the limit defined in the DEFAULT profile for this password parameter.

If you specify DEFAULT, it must follow the password parameter. For example, FAILED_LOGIN_ATTEMPTS DEFAULT.

constant

Indicates the value of the password parameter if you do not specify UNLIMITED or DEFAULT. If specified, it must follow the password parameter. For example, FAILED_LOGIN_ATTEMPTS 3.

{PASSWORD_COMPLEXITY_CHECKER| PASSWORD_VERIFY_FUNCTION} {function|NULL|DEFAULT}

Indicates if password verification is done on passwords and, if so, the function used for verification. You can specify either the PASSWORD_COMPLEXITY_CHECKER or the PASSWORD_VERIFY_FUNCTION password parameter. They are synonymous.

function refers to one of the three supported password complexity checker functions. Specify one of these functions to direct TimesTen to perform password verification. Valid values:
  • TT_VERIFY_FUNCTION
  • TT_STRONG_VERIFY_FUNCTION
  • TT_STIG_VERIFY_FUNCTION

NULL indicates that there is not a password verification function assigned for the profile.

DEFAULT indicates that the user is subject to the limits defined by the DEFAULT profile. The DEFAULT profile initially has a value of NULL.

If you do not specify the PASSWORD_COMPLEXITY_CHECKER password parameter, the value defaults to the limits defined for the DEFAULT profile.

Description: PROFILE statement

  • Use the CREATE PROFILE statement to create a profile for the password resources, which is a set of limits for the password parameters. If you assign the profile to a user, the user cannot exceed the limits specified for the profile. If you do not assign a profile to a user, TimesTen assigns the DEFAULT profile. See "Password management" in the Oracle TimesTen In-Memory Database Security Guide for more information on password management and profiles.

  • To specify the password parameter limits for a user, do the following:

    • Use the CREATE PROFILE statement to create a profile that defines the password parameter limits.

    • Use the CREATE USER or ALTER USER statement to assign the profile to the user.

  • There is a DEFAULT profile that defines a limit for each of the password parameters. This profile initially defines UNLIMITED for these parameters (which indicates that no limit has been set for the parameter). The exceptions are:

    • FAILED_LOGIN_ATTEMPTS: Set to 10.

    • PASSWORD_LOCK_TIME: Set to 0.0034722222222222 days (equal to 5 minutes, 5/1440 days)

    • PASSWORD_COMPLEXITY_CHECKER: Set to NULL.

    You can change these limits by using the ALTER PROFILE statement and specifying "DEFAULT" for the profile name. (Note that DEFAULT must be enclosed in double quotation marks.) See "ALTER PROFILE" for information.

  • If a user is not assigned a profile, the user is subject to the limits defined in the DEFAULT profile. If a user is assigned a profile and that profile omits a limit on the password parameter or specifies DEFAULT for the password parameter, then the user is subject to the limits on those password parameters as defined by the DEFAULT profile.

  • The instance administrator is assigned a system profile. You cannot alter or drop the profile of an instance administrator.

About password complexity checker verification

Password complexity checker verification ensures the password for a user is complex enough to deter intruders who try to guess passwords. When you specify a password complexity checker function in the CREATE PROFILE statement, and then assign this profile to a user, the user must create a password that meets the requirements defined for the password complexity checker function. These requirements depend on the specific password complexity checker function that you specify.

TimesTen provides the TT_VERIFY_FUNCTION, the TT_STRONG_VERIFY_FUNCTION, and the TT_STIG_VERIFY_FUNCTION password complexity checker functions to manage the complexity of the passwords. These functions are stored in the SYS schema.

The characters of interest for the password complexity checker functions are letters, digits, and special characters:
  • letter: Uppercase and lowercase letters
  • digit: 0-9 numbers
  • special: A character that is neither a letter nor a digit.
    `~!@#$%^&*()_-+={}[]\/<>,.?':|(space)

    Note:

    • If you use one or more of the special characters, the entire password must be enclosed in double quotation marks ("). The exceptions are the # and the @ special characters. (A password that contains the # or the @ does not need to be enclosed in double quotation marks.)

    • The password cannot contain a semicolon (;) or a double quotation mark (").

    • The password must begin with a letter unless you enclose the entire password in double quotation marks.

You cannot define your own function for password complexity checker verification. The complexity of the password is checked when you use the IDENTIFIED BY clause in the CREATE USER or ALTER USER statements.

Here are the password complexity checker functions:

TT_VERIFY_FUNCTION

The TT_VERIFY_FUNCTION does the following password complexity checker verification:
  • A password:
    • Must have at least 8 characters.
    • Of these 8 characters, must contain at least one letter, at least one digit, and at least one special character. .
  • A password cannot contain:
    • Username or the username reversed
    • Database name
    • Oracle or TimesTen

    Note:

    The comparisons are case insensitive.

TT_STRONG_VERIFY_FUNCTION

The TT_STRONG_VERIFY_FUNCTION performs the following password complexity checker verification:
  • Must have at least 9 characters.
  • Of these 9 characters, must contain at least two uppercase letters, at least two lowercase letters, at least two digits, and at least two special characters.

TT_STIG_VERIFY_FUNCTION

The TT_STIG_VERIFY_FUNCTION performs the following password complexity checker verification:
  • Must have at least 15 characters.
  • Of these 15 characters, must contain at least one uppercase letter, at least one lowercase letter, at least one digit, and at least one special character.

Description: Password complexity checker verification

  • EXECUTE privilege on TT_VERIFY_FUNCTION, TT_STRONG_VERIFY_FUNCTION, and TT_STIG_VERIFY_FUNCTION is required. TimesTen grants the EXECUTE privilege on these functions to PUBLIC by default.
  • The SYSTEM and the DEFAULT profiles are assigned a value of a NULL by default. A NULL value indicates that there is no password complexity checker function for these profiles, and as such, there is no password complexity checker verification done.
  • You cannot modify the SYSTEM profile to specify a password complexity checker function. Passwords for system users do not undergo password complexity checker verification.
  • You can use the ALTER PROFILE statement to modify the DEFAULT profile to specify a password complexity checker function. You specify such a function in the PASSWORD_COMPLEXITY_CHECKER (or PASSWORD_VERIFY_FUNCTION) clause.
  • The TT_VERIFY_FUNCTION, TT_STRONG_VERIFY_FUNCTION, and TT_STIG_VERIFY_FUNCTION functions in TimesTen are equivalent to the ORA12C_VERIFY_FUNCTION, ORA12C_STRONG_VERIFY_FUNCTION, and ORA12C_STIG_VERIFY_FUNCTION functions in Oracle Database.
  • If you use the ttMigrate utility to downgrade to an earlier major release (such as the 18.1 release), the PASSWORD_COMPLEXITY_CHECKER value is set to NULL for each profile in the database.

Restrictions on the password complexity checker functions

There are restrictions on the password complexity checker functions:
  • You cannot specify the SYS schema for the password complexity checker function. For example:
    Command> CREATE PROFILE my_profile LIMIT 
               PASSWORD_COMPLEXITY_CHECKER SYS.TT_VERIFY_FUNCTION;
    15187: Cannot specify schema name for password complexity checker function
    The command failed.
    
  • You cannot define your own password complexity checker function. Use only the TT_VERIFY_FUNCTION,the TT_STRONG_VERIFY_FUNCTION, or the TT_STIG_VERIFY_FUNCTION password complexity checker functions.
  • The password complexity checker verification is only done on a newly created password. You specify this new password by using the IDENTIFIED BY clause of the CREATE USER or the ALTER USER statement. TimesTen does not verify differences between an old and a new password.
  • Multi-byte characters are not supported when specifying passwords. TimesTen does not validate passwords with multi-byte characters.

Specify TT_VERIFY_FUNCTION for PASSWORD_COMPLEXITY_CHECKER

This example first creates a profile and specifies the TT_VERIFY_FUNCTION function. The example then queries the dba_profiles system view to verify the TT_VERIFY_FUNCTION has been assigned to this profile. A user who is assigned this profile must specify a password that meets the password verification requirements for this function.

Command> CREATE PROFILE myprofile_pw1 LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_VERIFY_FUNCTION;

Profile created.

Command> SELECT * FROM dba_profiles WHERE profile = 'MYPROFILE_PW1';
< MYPROFILE_PW1, FAILED_LOGIN_ATTEMPTS, PASSWORD, DEFAULT >
< MYPROFILE_PW1, PASSWORD_LIFE_TIME, PASSWORD, DEFAULT >
< MYPROFILE_PW1, PASSWORD_REUSE_TIME, PASSWORD, DEFAULT >
< MYPROFILE_PW1, PASSWORD_REUSE_MAX, PASSWORD, DEFAULT >
< MYPROFILE_PW1, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, TT_VERIFY_FUNCTION >
< MYPROFILE_PW1, PASSWORD_LOCK_TIME, PASSWORD, DEFAULT >
< MYPROFILE_PW1, PASSWORD_GRACE_TIME, PASSWORD, DEFAULT >
< MYPROFILE_PW1, TEMP_SPACE_PER_SESSION_MAX, MEMORY, DEFAULT >
8 rows found.

Create the sampleuser_pw1 user and assign the myprofile_pw1 profile to this user. Specify a password that meets the requirements of the TT_VERIFY_FUNCTION. See "TT_VERIFY_FUNCTION" for information on the TT_VERIFY_FUNCTION function.

Command> CREATE USER sampleuser_pw1 
           IDENTIFIED BY "A1!XXcg3" PROFILE myprofile_pw1;

User created.

Attempt to create the sampleuser_pw2. Assign the myprofile_pw1 profile to the user. Specify a password that contains the username reversed in uppercase. The CREATE USER statement fails. The password cannot contain the username reversed. Note that the comparison is case insensitive.

Command> CREATE USER sampleuser_pw2 
           IDENTIFIED BY "2WP_RESUELPMAS" PROFILE myprofile_pw1;
15186: Password complexity check for the specified password failed
15188: TT-20002: Password contains the username reversed
The command failed.

Specify TT_STRONG_VERIFY_FUNCTION for PASSWORD_COMPLEXITY_CHECKER

This example creates the myprofile_pw2 profile and specifies TT_STRONG_VERIFY_FUNCTION for the PASSWORD_COMPLEXITY_CHECKER password parameter. The example then creates the sampleuser_pw2 user, assigns the myprofile_pw2 profile to the user. The password meets the requirements of the TT_STRONG_VERIFY_FUNCTION function. See "TT_STRONG_VERIFY_FUNCTION" for more information on the TT_STRONG_VERIFY_FUNCTION function.

Command> CREATE PROFILE myprofile_pw2 LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_STRONG_VERIFY_FUNCTION;

Profile created.

Create the sampleuser_pw2, assign the myprofile_pw2 profile to the user. The password meets the requirements of the TT_STRONG_VERIFY_FUNCTION function. The user is successfully created.

Command> CREATE USER sampleuser_pw2 
           IDENTIFIED BY "!ddFF6C2?" PROFILE myprofile_pw2;

User created.

Specify TT_STIG_VERIFY_FUNCTION for PASSWORD_COMPLEXITY_CHECKER

This example creates the myprofile_pw3 profile and specifies TT_STIG_VERIFY_FUNCTION for the PASSWORD_COMPLEXITY_CHECKER password parameter. The example then creates the sampleuser_pw3 user, assigns the myprofile_pw3 profile to the user. The password meets the requirements of the TT_STIG_VERIFY_FUNCTION function. See "TT_STIG_VERIFY_FUNCTION" for more information on the TT_STIG_VERIFY_FUNCTION function.

Command> CREATE PROFILE myprofile_pw3 LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_STIG_VERIFY_FUNCTION;

Profile created.

Create the sampleuser_pw3, assign the myprofile_pw3 profile to the user. The password meets the requirements of the TT_STIG_VERIFY_FUNCTION function. The user is successfully created.

Command> CREATE USER sampleuser_pw3 
           IDENTIFIED BY "!ddBBKKUYT165>m" PROFILE myprofile_pw3;

User created.

Modify PASSWORD_COMPLEXITY_CHECKER value for SYSTEM and DEFAULT

This example queries the dba_profiles system view to check the value of the PASSWORD_COMPLEXITY_CHECKER password parameter for the SYSTEM and the DEFAULT profiles. The value is NULL by default.

Command> SELECT * FROM dba_profiles WHERE 
           resource_name='PASSWORD_COMPLEXITY_CHECKER' AND 
           profile IN ('DEFAULT','SYSTEM');
< DEFAULT, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, NULL >
< SYSTEM, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, NULL >
2 rows found.

Attempt to modify the PASSWORD_COMPLEXITY_CHECKER password parameter for the SYSTEM profile. An error results as this password parameter cannot be modified.

Command> ALTER PROFILE SYSTEM LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_STRONG_VERIFY_FUNCTION;
15176: Profile SYSTEM cannot be altered
The command failed.

Attempt to modify the PASSWORD_COMPLEXITY_CHECKER password parameter for the DEFAULT profile. The modification is successful. A user who is assigned the DEFAULT profile, or is not assigned a profile, must specify a password that meets the password verification requirements for the TT_STRONG_VERIFY_FUNCTION function.

Command> ALTER PROFILE "DEFAULT" LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_STRONG_VERIFY_FUNCTION;

Profile altered

Query the dba_profiles view to verify the TT_STRONG_VERIFY_FUNCTION has been assigned to the DEFAULT profile.

Command> SELECT * FROM dba_profiles WHERE 
           resource_name='PASSWORD_COMPLEXITY_CHECKER' AND 
           profile = 'DEFAULT';
< DEFAULT, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, TT_STRONG_VERIFY_FUNCTION >
1 row found.

Create a profile and attempt to specify an invalid password complexity checker function

This example specifies an invalid password complexity checker function for the PASSWORD_COMPLEXITY_CHECKER clause. Even though this function resides in the SYS schema, an error results, as you can only specify one of the three supported password complexity checker functions.

Command> CREATE PROFILE myprofile1 LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_COMPLEXITY_CHECK;
 8529: Invalid password complexity checker function TT_COMPLEXITY_CHECK
The command failed.

Create a profile and set limits on the password parameters

This example creates the profile1 profile and sets various limits on the password parameters. It then queries the dba_profiles system view to verify the limits.

Command> CREATE PROFILE profile1 LIMIT 
           FAILED_LOGIN_ATTEMPTS 5 
           PASSWORD_LIFE_TIME 60 
           PASSWORD_REUSE_TIME 60 
           PASSWORD_REUSE_MAX 5 
           PASSWORD_LOCK_TIME 1 
           PASSWORD_GRACE_TIME 10;
 
Profile created.

Query the dba_profiles system view to verify the limits. Note that since the PASSWORD_COMPLEXITY_CHECKER password parameter was not specified in the CREATE PROFILE statement, the value of PASSWORD_COMPLEXITY_CHECKER is DEFAULT (the value comes from the value that is in the DEFAULT profile).

Command> SELECT * FROM dba_profiles WHERE profile = 'PROFILE1' AND 
           resource_type='PASSWORD';
< PROFILE1, FAILED_LOGIN_ATTEMPTS, PASSWORD, 5 >
< PROFILE1, PASSWORD_LIFE_TIME, PASSWORD, 60 >
< PROFILE1, PASSWORD_REUSE_TIME, PASSWORD, 60 >
< PROFILE1, PASSWORD_REUSE_MAX, PASSWORD, 5 >
< PROFILE1, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, DEFAULT >
< PROFILE1, PASSWORD_LOCK_TIME, PASSWORD, 1 >
< PROFILE1, PASSWORD_GRACE_TIME, PASSWORD, 10 >
7 rows found.

Create a profile and specify FAILED_LOGIN_ATTEMPTS

This example creates the profile2 profile and specifies a value of 1 for FAILED_LOGIN_ATTEMPTS. The example then creates the user2 user and assigns user2 the profile2 profile. The user2 user attempts to connect to the database, but specifies an invalid password. The connection fails. After five minutes, the user2 user attempts to reconnect to the database. The connection succeeds due to the 0.0034722222222222 (equal to 5 minutes) value for PASSWORD_LOCK_TIME (specified in the DEFAULT profile).

Command> CREATE PROFILE profile2 LIMIT FAILED_LOGIN_ATTEMPTS 1;
 
Profile created.
 
Command> CREATE USER user2 IDENTIFIED BY user2 PROFILE profile2;
 
User created.

Grant admin privilege to user2.

Command> GRANT ADMIN TO user2;

Attempt to connect to the database. The connection fails due to an invalid password specified in the connection string.

Command> connect adding "UID=user2;PWD=user3" as user2;
 7001: User authentication failed
The command failed.

Attempt to connect again specifying the correct password in the connection string. The connection fails due to:

  • One previous failed connection attempt

  • An attempt to connect to the database before the five minute password lock time.

none: Command> use database1
database1: Command> connect adding "UID=user2;PWD=user2" as user2;
15179: the account is locked
The command failed.

After five minutes, attempt to connect to the database again. The connection succeeds.

none: Command> use database1
database1: Command> connect adding "UID=user2;PWD=user2" as user2;
Connection successful: DSN=database1;UID=user2;DataStore=/scratch/database1;
DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Determine the password parameter values in the DEFAULT profile

This example queries the dba_profiles system view to determine the password parameter values for the DEFAULT profile.

Command> SELECT * FROM dba_profiles WHERE profile = 'DEFAULT' AND 
           resource_type='PASSWORD';
< DEFAULT, FAILED_LOGIN_ATTEMPTS, PASSWORD, 10 >
< DEFAULT, PASSWORD_LIFE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_MAX, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, NULL >
< DEFAULT, PASSWORD_LOCK_TIME, PASSWORD, .0034 >
< DEFAULT, PASSWORD_GRACE_TIME, PASSWORD, UNLIMITED >
7 rows found.

Specify PASSWORD_LIFE_TIME and PASSWORD_GRACE_TIME

This example creates the profile4 profile and specifies a value of 0.0034722222222222 (equal to 5 minutes) for the PASSWORD_LIFE_TIME password parameter and a value of 0.01041667 (equal to 15 minutes) for the PASSWORD_GRACE_TIME password parameter. It then creates the user4 user and assigns the profile4 profile to user4. The example continues with attempts to connect to the database as user4.

Command> CREATE PROFILE profile4 LIMIT 
           PASSWORD_LIFE_TIME 0.0034722222222222 
           PASSWORD_GRACE_TIME 0.01041667;
 
Profile created.

Query the dba_profiles system view to verify the values for the password parameters.

Command> SELECT * FROM dba_profiles WHERE profile = 'PROFILE4' AND 
           resource_type='PASSWORD';
< PROFILE2, FAILED_LOGIN_ATTEMPTS, PASSWORD, DEFAULT >
< PROFILE2, PASSWORD_LIFE_TIME, PASSWORD, .0034 >
< PROFILE2, PASSWORD_REUSE_TIME, PASSWORD, DEFAULT >
< PROFILE2, PASSWORD_REUSE_MAX, PASSWORD, DEFAULT >
< PROFILE2, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, DEFAULT >
< PROFILE2, PASSWORD_LOCK_TIME, PASSWORD, DEFAULT >
< PROFILE2, PASSWORD_GRACE_TIME, PASSWORD, .0104 >
7 rows found.

Create the user4 user and assign user4 the profile4 profile. Grant the CONNECT privilege to user4.

Command> CREATE USER user4 IDENTIFIED BY user4 PROFILE profile4;
 
User created.
 
Command> GRANT CONNECT TO user4;

Connect to the database as user4. The connection succeeds.

Command> connect adding "UID=user4;PWD=user4" as user4;
Connection successful: DSN=access1;UID=user4;DataStore=/scratch/database1;
DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Disconnect from the database. After 5 minutes, reconnect to the database as user4. The connection succeeds but a warning is issued. The password lifetime is 5 minutes and the password grace time is 15 minutes.

user4: Command> disconnect user4;
Disconnecting from user4...
none: Command> use database1
database1: Command> connect adding "UID=user4;PWD=user4" as user4;
 
Warning 15182: Password will expire within 0.010417 days
 
Connection successful: DSN=access1;UID=user4;DataStore=/scratch/database1;
DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Disconnect from the database. After 15 minutes, reconnect to the database as user4. The connection fails as the password grace time of 15 minutes has ended.

user4: Command> disconnect user4;
Disconnecting from user4...
none: Command> use database1
database1: Command> connect adding "UID=user4;PWD=user4" as user4;
15180: the password has expired
The command failed.

Create a profile specifying only the LIMIT keyword

This example creates the profile5 profile and specifies just the LIMIT keyword. The example then queries the dba_profiles system view to illustrate the password parameter limits for the profile5 profile are all set to a value of DEFAULT.

Command> CREATE PROFILE profile5 LIMIT;
 
Profile created.
Command> SELECT * FROM dba_profiles WHERE profile = 'PROFILE5' AND 
           resource_type='PASSWORD
< PROFILE5, FAILED_LOGIN_ATTEMPTS, PASSWORD, DEFAULT >
< PROFILE5, PASSWORD_LIFE_TIME, PASSWORD, DEFAULT >
< PROFILE5, PASSWORD_REUSE_TIME, PASSWORD, DEFAULT >
< PROFILE5, PASSWORD_REUSE_MAX, PASSWORD, DEFAULT >
< PROFILE5, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, DEFAULT >
< PROFILE5, PASSWORD_LOCK_TIME, PASSWORD, DEFAULT >
< PROFILE5, PASSWORD_GRACE_TIME, PASSWORD, DEFAULT >
7 rows found.

Specify UNLIMITED for PASSWORD_REUSE_TIME

This example creates the profile6 profile and specifies a PASSWORD_REUSE_TIME of UNLIMITED. The password cannot be reused.

Command> CREATE PROFILE profile6 LIMIT 
           PASSWORD_REUSE_MAX 2 
           PASSWORD_REUSE_TIME UNLIMITED;
 
Profile created.

Create the user6 user and assign user6 the profile6 profile. Change the user6 password two times. Attempt to reuse the user6 password. The attempt fails due to the PASSWORD_REUSE_TIME value of UNLIMITED.

Command> CREATE USER user6 IDENTIFIED BY user6 PROFILE profile6;
 
User created.
 
Command> ALTER USER user6 IDENTIFIED BY user6_test1;
 
User altered.
 
Command> ALTER USER user6 IDENTIFIED BY user6_test2;
 
User altered.
 
Command> ALTER USER user6 IDENTIFIED BY user6;
15183: Password cannot be reused
The command failed.

Specify DEFAULT for PASSWORD_REUSE_TIME

This example creates the profile7 profile, specifying the value of DEFAULT for the PASSWORD_REUSE_TIME password parameter and the value of 3 for the PASSWORD_REUSE_MAX password parameter. TimesTen uses the value in the DEFAULT profile for the PASSWORD_REUSE_TIME password parameter.

Command> CREATE PROFILE profile7 LIMIT 
           PASSWORD_REUSE_TIME DEFAULT 
           PASSWORD_REUSE_MAX 3;
 
Profile created.

Query the dba_profiles system view to verify the password parameter values for the profile7 profile. Note the value of DEFAULT for PASSWORD_REUSE_TIME and a value of 3 for PASSWORD_REUSE_MAX (represented in bold).

Command> SELECT * FROM dba_profiles WHERE profile = 'PROFILE7' AND
           resource_type = 'PASSWORD';
< PROFILE7, FAILED_LOGIN_ATTEMPTS, PASSWORD, DEFAULT >
< PROFILE7, PASSWORD_LIFE_TIME, PASSWORD, DEFAULT >
< PROFILE7, PASSWORD_REUSE_TIME, PASSWORD, DEFAULT >
< PROFILE7, PASSWORD_REUSE_MAX, PASSWORD, 3 >
< PROFILE7, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, DEFAULT >
< PROFILE7, PASSWORD_LOCK_TIME, PASSWORD, DEFAULT >
< PROFILE7, PASSWORD_GRACE_TIME, PASSWORD, DEFAULT >
7 rows found.

Query the dba_profiles system view to verify the password parameter values for the DEFAULT profile. Note the value of UNLIMITED for PASSWORD_REUSE_TIME (represented in bold).

Command> SELECT * FROM dba_profiles WHERE profile = 'DEFAULT' AND
           resource_type = 'PASSWORD';
< DEFAULT, FAILED_LOGIN_ATTEMPTS, PASSWORD, 10 >
< DEFAULT, PASSWORD_LIFE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_MAX, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, NULL >
< DEFAULT, PASSWORD_LOCK_TIME, PASSWORD, .0034 >
< DEFAULT, PASSWORD_GRACE_TIME, PASSWORD, UNLIMITED >
7 rows found.

Create the user7 user and assign the profile7 profile to user7. Change the user7 password three times. The user7 password cannot be reused due to the value of UNLIMITED for the PASSWORD_REUSE_TIME parameter.

Command> CREATE USER user7 IDENTIFIED BY user7 PROFILE profile7;
 
User created.
 
Command> ALTER USER user7 IDENTIFIED BY user7_test1;
 
User altered.
 
Command> ALTER USER user7 IDENTIFIED BY user7_test2;
 
User altered.
 
Command> ALTER USER user7 IDENTIFIED BY user_test3;
 
User altered.
 
Command> ALTER USER user7 IDENTIFIED BY user7;
15183: Password cannot be reused
The command failed.

Specify PASSWORD_REUSE_TIME and PASSWORD_REUSE_MAX

This example creates the profile8 profile, specifying a value of 0.0020833 (equal to approximately 2 minutes) for the PASSWORD_REUSE_TIME password parameter and a value of 2 for the PASSWORD_REUSE_MAX password parameter. The example then creates the user8 user and assigns user8 the profile8 profile. The user8 password is changed two times within two minutes. Then, still within the two minutes, the original user8 password (user8_pwd) is reused. The ALTER USER operation fails. Even though the password is changed 2 times, the original password can only be reused after 0.00208333 days (equal to approximately two minutes). After two minutes, the original user8 password (user8_pwd) is reused again. The ALTER USER operation succeeds. The user's password was changed two times and more than two minutes had passed.

Command> CREATE PROFILE profile8 LIMIT 
           PASSWORD_REUSE_TIME 0.00208333 
           PASSWORD_REUSE_MAX 2;
 
Profile created.

Create the user8 user and assign user8 the profile8 profile.

Command> CREATE USER user8 IDENTIFIED BY user8_pwd PROFILE profile8;
 
User created.
 

Immediately alter the user, changing the password two times.

Command> ALTER USER user8 IDENTIFIED BY user8_test1;
 
User altered.
 
Command> ALTER USER user8 IDENTIFIED BY user8_test2;
 
User altered.

Within two minutes, attempt to reuse the original user8_pwd password (represented in bold). The ALTER USER operation fails as the original password can only be reused after two minutes.

Command> ALTER USER user8 IDENTIFIED BY user8_pwd;
15183: Password cannot be reused
The command failed.

After two minutes, attempt to reuse the original user8_pwd password (represented in bold). The ALTER USER operation succeeds. The original password can be reused as the password was changed two times and two minutes had expired.

Command> ALTER USER user8 IDENTIFIED BY user8_pwd;
 
User altered.

CREATE REPLICATION

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

The CREATE REPLICATION statement:

  • Defines a classic replication scheme on a participating database.

  • Installs the specified configuration in the executing database's replication system tables.

  • Typically consists of one or more replication element specifications and zero or more STORE specifications.

TimesTen SQL configuration for replication also provides a programmable way to configure a classic replication scheme. The configuration can be embedded in C, C++ or Java code. Replication can be configured locally or from remote systems using client/server.

In addition, you need to use the ttRepAdmin utility to maintain operations not covered by the supported SQL statements. Use ttRepAdmin to change replication state, duplicate databases, list the replication configuration, and view replication status.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

Definitions

A replication element is an entity that TimesTen synchronizes between databases. A replication element can be a whole table or a database. A database can include most types of tables and sequences. It can include only specified tables and sequences, or include all tables except specified tables and sequences. It cannot include temporary tables or views, whether materialized or nonmaterialized.

A replication scheme is a set of replication elements, as well as the databases that maintain copies of these elements.

For more detailed information on SQL configuration for classic replication, see "Defining a classic replication scheme" in the Oracle TimesTen In-Memory Database Replication Guide.

SQL syntax

CREATE REPLICATION [Owner.]ReplicationSchemeName
{ ELEMENT ElementName
  { DATASTORE | { TABLE [Owner.]TableName [CheckConflicts]} | 
       SEQUENCE [Owner.]SequenceName}
     { MASTER | PROPAGATOR } FullStoreName
     [TRANSMIT { NONDURABLE | DURABLE }]
     { SUBSCRIBER FullStoreName [,...]
        [ReturnServiceAttribute] } [,...] }
     [...]
     [{INCLUDE | EXCLUDE}
          {TABLE [[Owner.]TableName[,...]] |
           SEQUENCE [[Owner.]SequenceName[,...]} [,...]] 
[ STORE FullStoreName [StoreAttribute [... ]]] [...]
[ NetworkOperation[...]]

See "CHECK CONFLICTS" for CheckConflicts syntax.

Syntax for ReturnServiceAttribute:

{ RETURN RECEIPT [BY REQUEST] |
  RETURN TWOSAFE [BY REQUEST] |
  NO RETURN }

Syntax for StoreAttribute:

  DISABLE RETURN {SUBSCRIBER | ALL} NumFailures 
  RETURN SERVICES {ON | OFF} WHEN [REPLICATION] STOPPED 
  DURABLE COMMIT {ON | OFF}
  RESUME RETURN Milliseconds 
  LOCAL COMMIT ACTION {NO ACTION | COMMIT} 
  RETURN WAIT TIME Seconds 
  COMPRESS TRAFFIC {ON | OFF}
  PORT PortNumber 
  TIMEOUT Seconds 
  FAILTHRESHOLD Value 
  CONFLICT REPORTING SUSPEND AT Value 
  CONFLICT REPORTING RESUME AT Value 
  TABLE DEFINITION CHECKING {RELAXED|EXACT}

Syntax for NetworkOperation:

ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName
  { { MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost }
      PRIORITY Priority } [...]

Parameters

Parameter Description

[Owner.]ReplicationSchemeName

Name assigned to the new classic replication scheme. Classic replication schemes should have names that are unique from all other database objects.

CheckConflicts

Check for replication conflicts when simultaneously writing to bidirectionally replicated databases. See "CHECK CONFLICTS" for information on CheckConflicts.

COMPRESS TRAFFIC {ON | OFF}

Compress replicated traffic to reduce the amount of network bandwidth. ON specifies that all replicated traffic for the database defined by STORE be compressed. OFF (the default) specifies no compression. See "Compressing replicated traffic" in Oracle TimesTen In-Memory Database Replication Guide for details.

CONFLICT REPORTING SUSPEND AT Value

Suspends conflict resolution reporting.

Value is a non-negative integer. The default is 0 and means never suspend. Conflict reporting is suspended when the rate of conflict exceeds Value. If you set Value to 0, conflict reporting suspension is turned off.

This clause is valid for table level replication.

CONFLICT REPORTING RESUME AT Value

Resumes conflict resolution reporting.

Value is a non-negative integer. Conflict reporting is resumed when the rate of conflict falls below Value. The default is 1.

This clause is valid for table level replication.

DATASTORE

Define entire database as element. This type of element can only be defined for a master database that is not configured with an element of type TABLE in the same or a different replication scheme.

{INCLUDE|EXCLUDE}

{[TABLE[Owner.]TableName [,...]]|

SEQUENCE

[[Owner.]SequenceName[,...]]} [,...]

INCLUDE includes in the DATASTORE element only the tables or sequences listed. Use one INCLUDE clause for each object type (table or sequence).

EXCLUDE includes in the DATASTORE element all tables or sequences except for those listed. Use one EXCLUDE clause for each object type (table or sequence).

DISABLE RETURN {SUBSCRIBER|ALL} NumFailures

Set the return service failure policy so that return service blocking is disabled after the number of timeouts specified by NumFailures. Selecting SUBSCRIBER applies this policy only to the subscriber that fails to acknowledge replicated updates within the set timeout period. ALL applies this policy to all subscribers should any of the subscribers fail to respond. This failure policy can be specified for either the RETURN RECEIPT or RETURN TWOSAFE service.

If DISABLE RETURN is specified but RESUME RETURN is not specified, the return services remain off until the replication agent for the database has been restarted.

DURABLE COMMIT {ON|OFF}

Overrides the DurableCommits general connection attribute setting. DURABLE COMMIT ON enables durable commits regardless of whether the replication agent is running or stopped.

ELEMENT ElementName

The entity that TimesTen synchronizes between databases. TimesTen supports the entire database (DATASTORE) and whole tables (TABLE) as replication elements.

ElementName is the name given to the replication element. The ElementName for a TABLE element can be up to 30 characters in length. The ElementName for a DATASTORE element must be unique with respect to other DATASTORE element names within the first 20 characters. Each ElementName must be unique within a classic replication scheme. Also, you cannot define two element descriptions for the same element.

See "Defining replication elements" in Oracle TimesTen In-Memory Database Replication Guide for details.

FAILTHRESHOLD Value

The number of log files that can accumulate for a subscriber database. If this value is exceeded, the subscriber is set to the Failed state.The value 0 means "No Limit." This is the default.

See "Setting the transaction log failure threshold" in Oracle TimesTen In-Memory Database Replication Guide.

FullStoreName

The database, specified as one of the following:

  • SELF

  • The prefix of the database file name

For example, if the database path is directory/subdirectory/data.ds0, then data is the database name that should be used.

This is the database file name specified in the DataStore attribute of the DSN description with optional host ID in the form:

DataStoreName [ON Host]

Host can be either an IP address or a literal host name assigned to one or more IP addresses, as described in "Configuring the network" in Oracle TimesTen In-Memory Database Replication Guide. Host names containing special characters must be surrounded by double quotes. For example: "MyHost-500". Host names can be up to 30 characters long.

LOCAL COMMIT ACTION {NO ACTION | COMMIT}

Specifies the default action to be taken for a return twosafe transaction in the event of a timeout.

Note: This attribute is only valid when the RETURN TWOSAFE or RETURN TWOSAFE BY REQUEST attribute is set in the SUBSCRIBER clause.

NO ACTION: On timeout, the commit function returns to the application, leaving the transaction in the same state it was in when it entered the commit call, with the exception that the application is not able to update any replicated tables. The application can only reissue the commit. The transaction may not be rolled back. This is the default.

COMMIT: On timeout, the commit function attempts to perform a COMMIT to end the transaction locally. No more operations are possible on the same transaction.

This setting can be overridden for specific transactions by calling the localAction parameter in the ttRepSyncSet procedure.

MASTER FullStoreName

The database on which applications update the specified element. The MASTER database sends updates to its SUBSCRIBER databases. The FullStoreName must be the database specified in the DataStore attribute of the DSN description.

NO RETURN

Specifies that no return service is to be used. This is the default.

For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

PORT PortNumber

The TCP/IP port number on which the replication agent for the database listens for connections. If not specified, the replication agent automatically allocates a port number.

PROPAGATOR FullStoreName

The database that receives replicated updates and passes them on to other databases. The FullStoreName must be the database specified in the DataStore attribute of the DSN description.

RESUME RETURN Milliseconds

If return service blocking has been disabled by DISABLE RETURN, this attribute sets the policy on when to re-enable return service blocking. Return service blocking is re-enabled as soon as the failed subscriber acknowledges the replicated update in a period of time that is less than the specified Milliseconds.

If DISABLE RETURN is specified but RESUME RETURN is not specified, the return services remain off until the replication agent for the database has been restarted.

RETURN RECEIPT [BY REQUEST]

Enables the return receipt service, so that applications that commit a transaction to a master database are blocked until the transaction is received by all subscribers.

RETURN RECEIPT applies the service to all transactions. If you specify RETURN REQUEST BY REQUEST, you can use the ttRepSyncSet procedure to enable the return receipt service for selected transactions. For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

RETURN SERVICES {ON|OFF} WHEN [REPLICATION] STOPPED

Sets return services on or off when replication is disabled (stopped or paused state).

OFF disables return services when replication is disabled and is the default for RETURN RECEIPT service. ON allows return services to continue to be enabled when replication is disabled and is the default for RETURN TWOSAFE service.

RETURN TWOSAFE [BY REQUEST]

Enables the return twosafe service, so that applications that commit a transaction to a master database are blocked until the transaction is committed on all subscribers.

Note: This service can only be used in a bidirectional replication scheme where the elements are defined as DATASTORE.

Specifying RETURN TWOSAFE applies the service to all transactions. If you specify RETURN TWOSAFE BY REQUEST, you can use the ttRepSyncSet procedure to enable the return receipt service for selected transactions. For details on the use of the return services, see "Using a return service" in Oracle TimesTen In-Memory Database Replication Guide.

RETURN WAIT TIME Seconds

Specifies the number of seconds to wait for return service acknowledgment. The default value is 10 seconds. A value of 0 (zero) means that there is no timeout. Your application can override this timeout setting by calling the returnWait parameter in the ttRepSyncSet procedure.

SEQUENCE [Owner.]SequenceName

Define the sequence specified by [Owner.]SequenceName as element. See "Defining replication elements" in Oracle TimesTen In-Memory Database Replication Guide for details.

STORE FullStoreName

Defines the attributes for a given database. Attributes include PORT, TIMEOUT and FAILTHRESHOLD. The FullStoreName must be the database specified in the DataStore attribute of the DSN description.

SUBSCRIBER FullStoreName

A database that receives updates from the MASTER databases. The FullStoreName must be the database specified in the DataStore attribute of the DSN description.

TABLE [Owner.]TableName

Define the table specified by [Owner.]TableName as element. See "Defining replication elements" in Oracle TimesTen In-Memory Database Replication Guide for details.

TIMEOUT Seconds

The maximum number of seconds the replication agent waits for a response from remote replication agents. The default is 120 seconds.

Note: For large transactions that may cause a delayed response from the remote replication agent, the agent scales the timeout based on the size of the transaction. This scaling is disabled if you set TIMEOUT to less than or equal to 60 seconds. Also see "Setting wait timeout for response from remote replication agents" in Oracle TimesTen In-Memory Database Replication Guide.

TRANSMIT {DURABLE | NONDURABLE}

Specifies whether to flush the master log to the file system before sending a batch of committed transactions to the subscribers.

TRANSMIT NONDURABLE specifies that records in the master log are not to be flushed to the file system before they are sent to subscribers. This setting can only be used if the specified element is a DATASTORE. This is the default for RETURN TWOSAFE transactions.

TRANSMIT DURABLE specifies that records are to be flushed to the file system before they are sent to subscribers. This is the default for asynchronous and RETURN RECEIPT transactions.

Note: TRANSMIT DURABLE has no effect on RETURN TWOSAFE transactions.

Note: TRANSMIT DURABLE cannot be set for active standby pairs.

See "Setting transmit durability on DATASTORE element" in Oracle TimesTen In-Memory Database Replication Guide for more information.

TABLE DEFINITION CHECKING {EXACT|RELAXED}

Specifies type of table definition checking that occurs on the subscriber:

  • EXACT: The tables must be identical on master and subscriber.

  • RELAXED: The tables must have the same key definition, number of columns and column data types.

The default is RELAXED.

Note: If you use TABLE DEFINITION CHECKING EXACT, use ttMigrate -exactUpgrade if you migrate the database. If you use TABLE DEFINITION CHECKING RELAXED, use ttMigrate -relaxedUpgrade if you migrate the database.

ROUTE MASTER FullStoreName SUBSCRIBER FullStoreName

Denotes the NetworkOperation clause. If specified, enables you to control the network interface that a master store uses for every outbound connection to each of its subscriber stores.

Can be specified more than once.

For FullStoreName, ON "host" must be specified.

MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost

MasterHost and SubscriberHost are the IP addresses for the network interface on the master and subscriber stores. Specify in dot notation or canonical format or in colon notation for IPV6.

Clause can be specified more than once.

PRIORITY Priority

Variable expressed as an integer from 1 to 99. Denotes the priority of the IP address. Lower integral values have higher priority. An error is returned if multiple addresses with the same priority are specified. Controls the order in which multiple IP addresses are used to establish peer connections.

Required syntax of NetworkOperation clause. Follows MASTERIP MasterHost | SUBSCRIBERIP SubscriberHost clause.

CHECK CONFLICTS

Syntax

The syntax for CHECK CONFLICTS is:

{NO CHECK |
CHECK CONFLICTS BY ROW TIMESTAMP
      COLUMN ColumnName
      [ UPDATE BY { SYSTEM | USER } ]
      [ ON EXCEPTION { ROLLBACK [ WORK ] | NO ACTION } ]
      [ {REPORT TO 'FileName'
             [ FORMAT { XML | STANDARD } ] | NO REPORT
      } ]
}

Note:

A CHECK CONFLICT clause can only be used for elements of type TABLE.

Parameters

The CHECK CONFLICTS clause of the CREATE REPLICATION or ALTER REPLICATION statement has the following parameters:

Parameter Description

CHECK CONFLICTS BY ROW TIMESTAMP

Indicates that all update and uniqueness conflicts are to be detected. Conflicts are resolved in the manner specified by the ON EXCEPTION parameter.

It also detects delete conflicts with UPDATE operations.

COLUMN ColumnName

Indicates the column in the replicated table to be used for timestamp comparison. The table is specified in the ELEMENT description by TableName.

ColumnName is a nullable column of type BINARY(8) used to store a timestamp that indicates when the row was last updated. TimesTen rejects attempts to update a row with a lower timestamp value than the stored value. The specified ColumnName must exist in the replicated table on both the master and subscriber databases.

NO CHECK

Specify to suppress conflict resolution for a given element.

UPDATE BY {SYSTEM | USER}

Specifies whether the timestamp values are maintained by TimesTen (SYSTEM) or the application (USER). The replicated table in the master and subscriber databases must use the same UPDATE BY specification. See "Enabling system timestamp column maintenance" and "Enabling user timestamp column maintenance" in Oracle TimesTen In-Memory Database Replication Guide for more information. The default is UPDATE BY SYSTEM.

ON EXCEPTION {ROLLBACK[WORK |NO ACTION}

Specifies how to resolve a detected conflict. ROW TIMESTAMP conflict detection has the resolution options:

  • ROLLBACK [WORK]: Abort the transaction that contains the conflicting action.

  • NO ACTION: Complete the transaction without performing the conflicting action (UPDATE, INSERT or DELETE).

The default is ON EXCEPTION ROLLBACK [WORK].

REPORT TO 'FileName'

Specifies the file to log updates that fail the timestamp comparison. FileName is a SQL character string that cannot exceed 1,000 characters. (SQL character string literals are single-quoted strings that may contain any sequence of characters, including spaces.) The same file can be used to log failed updates for multiple tables.

[FORMAT {XML|STANDARD}]

Optionally specifies the conflict report format for an element. The default format is STANDARD.

NO REPORT

Specify to suppress logging of failed timestamp comparisons.

Description

  • The names of all databases on the same host must be unique for each classic replication scheme for each TimesTen instance.

  • Replication elements can only be updated (by normal application transactions) through the MASTER database. PROPAGATOR and SUBSCRIBER databases are read-only.

  • If you define a classic replication scheme that permits multiple databases to update the same table, see "Resolving Replication Conflicts" in Oracle TimesTen In-Memory Database Replication Guide for recommendations on how to avoid conflicts when updating rows.

  • SELF is intended for classic replication schemes where all participating databases are local. Do not use SELF for a distributed classic replication scheme in a production environment, where spelling out the host name for each database in a script enables it to be used at each participating database.

  • Each attribute for a given STORE may be specified only once, or not at all.

  • Specifying the PORT of a database for one classic replication scheme specifies it for all classic replication schemes. All other connection attributes are specific to the classic replication scheme specified in the command.

  • For replication schemes, DataStoreName is always the prefix of the TimesTen database checkpoint file names. These are the files with the.ds0 and.ds1 suffixes that are saved on the file system by checkpoint operations.

  • If a row with a default NOT INLINE VARCHAR value is replicated, the receiver creates a copy of this value for each row instead of pointing to the default value if and only if the default value of the receiving node is different from the sending node.

  • To use timestamp comparison on replicated tables, you must specify a nullable column of type BINARY(8) to hold the timestamp value. Define the timestamp column when you create the table. You cannot add the timestamp column with the ALTER TABLE statement. In addition, the timestamp column cannot be part of a primary key or index.

  • If you specify the XML report format, two XML documents are generated:

    • FileName.xml: This file contains the DTD for the report and the root node for the report. It includes the document definition and the include directive.

    • FileName.include: This file is included in FileName.xml and contains all the actual conflicts.

    • The FileName.include file can be truncated. Do not truncate the FileName.xml file.

    • For a complete description of the XML format, including examples of each conflict, see "Reporting conflicts to an XML file" in Oracle TimesTen In-Memory Database Replication Guide.

  • If you specify a report format for an element and then drop the element, the corresponding report files are not deleted.

  • Use the CONFLICT REPORTING SUSPEND AT clause to specify a high water mark threshold at which the reporting of conflict resolution is suspended.

  • Use the CONFLICT REPORTING RESUME AT clause to specify a low water mark threshold where the reporting of conflict resolution is resumed. When the rate of conflict falls below the low water mark threshold, conflict resolution reporting is resumed.

  • The state of whether conflict reporting is suspended or not by a replication agent does not persist across the local replication agent and the peer agent stop and restart.

  • Do not use the CREATE REPLICATION statement to replicate cache groups. Only active standby pairs can replicate cache groups. See the CREATE ACTIVE STANDBY PAIR statement.

Examples

Replicate the contents of repl.tab from masterds to two subscribers, subscriber1ds and subscriber2ds.

CREATE REPLICATION repl.twosubscribers
       ELEMENT e TABLE repl.tab
         MASTER masterds ON "server1"
         SUBSCRIBER subscriber1ds ON "server2",
                    subscriber2ds ON "server3";

Replicate the entire masterds database to the subscriber, subscriber1ds. The FAILTHRESHOLD specifies that a maximum of 10 log files can accumulate on masterds before it decides that subscriber1ds has failed.

CREATE REPLICATION repl.wholestore
  ELEMENT e DATASTORE
     MASTER masterds ON "server1"
     SUBSCRIBER subscriber1ds ON "server2"
  STORE masterds FAILTHRESHOLD 10;

Bidirectionally replicate the entire westds and eastds databases and enable the RETURN TWOSAFE service.

CREATE REPLICATION repl.biwholestore
  ELEMENT e1 DATASTORE
     MASTER westds ON "westcoast"
     SUBSCRIBER eastds ON "eastcoast"
        RETURN TWOSAFE
  ELEMENT e2 DATASTORE
     MASTER eastds ON "eastcoast"
     SUBSCRIBER westds ON "westcoast"
        RETURN TWOSAFE;

Enable the return receipt service for select transaction updates to the subscriber1ds subscriber.

CREATE REPLICATION repl.twosubscribers
  ELEMENT e TABLE repl.tab
     MASTER masterds ON "server1"
     SUBSCRIBER subscriber1ds ON "server2"
        RETURN RECEIPT BY REQUEST
     SUBSCRIBER subscriber2ds ON "server3";

Replicate the contents of the customerswest table from the west database to the ROUNDUP database and the customerseast table from the east database. Enable the return receipt service for all transactions.

CREATE REPLICATION r
       ELEMENT west TABLE customerswest
         MASTER west ON "serverwest"
         SUBSCRIBER roundup ON "serverroundup"
            RETURN RECEIPT
       ELEMENT east TABLE customerseast
         MASTER east ON "servereast"
         SUBSCRIBER roundup ON "serverroundup"
            RETURN RECEIPT;

Replicate the contents of the repl.tab table from the centralds database to the propds database, which propagates the changes to the backup1ds and backup2ds databases.

CREATE REPLICATION repl.propagator
       ELEMENT a TABLE repl.tab
         MASTER centralds ON "finance"
         SUBSCRIBER proprds ON "nethandler"
       ELEMENT b TABLE repl.tab
         PROPAGATOR proprds ON "nethandler"
         SUBSCRIBER backup1ds ON "backupsystem1"
                    bakcup2ds ON "backupsystem2";

Bidirectionally replicate the contents of the repl.accounts table between the eastds and westds databases. Each database is both a master and a subscriber for the repl.accounts table.

Because the repl.accounts table can be updated on either the eastds or westds database, it includes a timestamp column (tstamp). The CHECK CONFLICTS clause establishes automatic timestamp comparison to detect any update conflicts between the two databases. In the event of a comparison failure, the entire transaction that includes an update with the older timestamp is rolled back (discarded).

CREATE REPLICATION repl.r1
ELEMENT elem_accounts_1 TABLE repl.accounts
   CHECK CONFLICTS BY ROW TIMESTAMP
      COLUMN tstamp
      UPDATE BY SYSTEM
      ON EXCEPTION ROLLBACK
   MASTER westds ON "westcoast"
   SUBSCRIBER eastds ON "eastcoast"
ELEMENT elem_accounts_2 TABLE repl.accounts
   CHECK CONFLICTS BY ROW TIMESTAMP
      COLUMN tstamp
      UPDATE BY SYSTEM
      ON EXCEPTION ROLLBACK
   MASTER eastds ON "eastcoast"
   SUBSCRIBER westds ON "westcoast";

Replicate the contents of the repl.accounts table from the activeds database to the backupds database, using the return twosafe service, and using TCP/IP port 40000 on activeds and TCP/IP port 40001 on backupds. The transactions on activeds need to be committed whenever possible, so configure replication so that the transaction is committed even after a replication timeout using LOCAL COMMIT ACTION, and so that the return twosafe service is disabled when replication is stopped. To avoid significant delays in the application if the connection to the backupds database is interrupted, configure the return service to be disabled after five transactions have timed out, but also configure the return service to be re-enabled when the backupds database's replication agent responds in under 100 milliseconds. Finally, the bandwidth between databases is limited, so configure replication to compress the data when it is replicated from the activeds database.

CREATE REPLICATION repl.r
ELEMENT elem_accounts_1 TABLE repl.accounts
   MASTER activeds ON "active"
   SUBSCRIBER backupds ON "backup"
      RETURN TWOSAFE
ELEMENT elem_accounts_2 TABLE repl.accounts
   MASTER activeds ON "active"
   SUBSCRIBER backupds ON "backup"
      RETURN TWOSAFE
STORE activeds ON "active"
   PORT 40000
   LOCAL COMMIT ACTION COMMIT
   RETURN SERVICES OFF WHEN REPLICATION STOPPED
   DISABLE RETURN SUBSCRIBER 5
   RESUME RETURN 100
   COMPRESS TRAFFIC ON
STORE backupds ON "backup"
   PORT 40001;

Illustrate conflict reporting suspend and conflict reporting resume clauses for table level replication. Use these clauses for table level replication not database replication. Issue repschemes command to show that replication scheme is created.

Command> CREATE TABLE repl.accounts (tstamp BINARY (8) NOT NULL 
           PRIMARY KEY, tstamp1 BINARY (8));
Command> CREATE REPLICATION repl.r2
           ELEMENT elem_accounts_1 TABLE repl.accounts
           CHECK CONFLICTS BY ROW TIMESTAMP
           COLUMN tstamp1
           UPDATE BY SYSTEM
           ON EXCEPTION ROLLBACK WORK
           MASTER westds ON "west1"
           SUBSCRIBER eastds ON "east1"
  ELEMENT elem_accounts_2 TABLE repl.accounts
  CHECK CONFLICTS BY ROW TIMESTAMP 
  COLUMN tstamp1
  UPDATE BY SYSTEM 
  ON EXCEPTION ROLLBACK WORK 
  MASTER eastds ON "east1"
  SUBSCRIBER westds ON "west1"
  STORE westds
  CONFLICT REPORTING SUSPEND AT 20
  CONFLICT REPORTING RESUME AT 10;
Command> REPSCHEMES;

Replication Scheme REPL.R2:

  Element: ELEM_ACCOUNTS_1
  Type: Table REPL.ACCOUNTS
  Conflict Check Column: TSTAMP1
  Conflict Exception Action: Rollback Work
  Conflict Timestamp Update: System
  Conflict Report File: (none)
  Master Store: WESTDS on WEST1 Transmit Durable
  Subscriber Store: EASTDS on EAST1

  Element: ELEM_ACCOUNTS_2
  Type: Table REPL.ACCOUNTS
  Conflict Check Column: TSTAMP1
  Conflict Exception Action: Rollback Work
  Conflict Timestamp Update: System
  Conflict Report File: (none)
  Master Store: EASTDS on EAST1 Transmit Durable
  Subscriber Store: WESTDS on WEST1

  Store: EASTDS on EAST1
    Port: (auto)
    Log Fail Threshold: (none)
    Retry Timeout: 120 seconds
    Compress Traffic: Disabled

  Store: WESTDS on WEST1
    Port: (auto)
    Log Fail Threshold: (none)
    Retry Timeout: 120 seconds
    Compress Traffic: Disabled
    Conflict Reporting Suspend: 20
    Conflict Reporting Resume: 10

1 replication scheme found.

Example of NetworkOperation clause with 2 MASTERIP and SUBSCRIBERIP clauses:

CREATE REPLICATION r ELEMENT e DATASTORE
MASTER rep1 SUBSCRIBER rep2 RETURN RECEIPT
MASTERIP "1.1.1.1" PRIORITY 1 SUBSCRIBERIP "2.2.2.2"
    PRIORITY 1
MASTERIP "3.3.3.3" PRIORITY 2 SUBSCRIBERIP "4.4.4.4"
    PRIORITY 2;

Example of NetworkOperation clause. Use the default sending interface but a specific receiving network:

CREATE REPLICATION r
ELEMENT e DATASTORE
MASTER rep1 SUBSCRIBER rep2
ROUTE MASTER rep1 ON "machine1" SUBSCRIBER rep2 ON "machine2"
SUBSCRIBERIP "rep2nic2" PRIORITY 1;

Example of using the NetworkOperation clause with multiple subscribers:

CREATE REPLICATION r ELEMENT e DATASTORE
MASTER rep1 SUBSCRIBER rep2,rep3
ROUTE MASTER rep1 ON "machine1" SUBSCRIBER rep2 ON "machine2"
MASTERIP "1.1.1.1" PRIORITY 1 SUBSCRIBERIP "2.2.2.2"
    PRIORITY 1
ROUTE MASTER Rep1 ON "machine1" SUBSCRIBER Rep3 ON "machine2"
MASTERIP "3.3.3.3" PRIORITY 2 SUBSCRIBERIP "4.4.4.4";

CREATE SEQUENCE

The CREATE SEQUENCE statement creates a new sequence number generator that can subsequently be used by multiple users to generate unique integers. Use the CREATE SEQUENCE statement to define the initial value of the sequence, define the increment value, the maximum or minimum value and determine if the sequence continues to generate numbers after the minimum or maximum is reached.

Required privilege

CREATE SEQUENCE (if owner) or CREATE ANY SEQUENCE (if not owner).

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout. The BATCH clause is supported in TimesTen Scaleout only.

SQL syntax

CREATE SEQUENCE [Owner.]SequenceName
       [INCREMENT BY IncrementValue]
       [MINVALUE MinimumValue]
       [MAXVALUE MaximumValue]
       [CYCLE]
       [CACHE CacheValue]
       [START WITH StartValue]
       [BATCH BatchValue]

Parameters

Parameter Description

SEQUENCE [Owner.]SequenceName

Name of the sequence number generator.

INCREMENT BY IncrementValue

The incremental value between consecutive numbers. This value can be either a positive or negative integer. It cannot be 0. If the value is positive, it is an ascending sequence. If the value is negative, it is descending. The default value is 1. In a descending sequence, the range starts from MAXVALUE to MINVALUE, and vice versa for ascending sequence.

MINVALUE MinimumValue

Specifies the minimum value for the sequence. The default minimum value is 1.

MAXVALUE MaximumValue

The largest possible value for an ascending sequence, or the starting value for a descending sequence. The default maximum value is (263) -1, which is the maximum of BIGINT.

CYCLE

Indicates that the sequence number generator continues to generate numbers after it reaches the maximum or minimum value. By default, sequences do not cycle. Once the number reaches the maximum value in the ascending sequence, the sequence wraps around and generates numbers from its minimum value. For a descending sequence, when the minimum value is reached, the sequence number wraps around, beginning from the maximum value. If CYCLE is not specified, the sequence number generator stops generating numbers when the maximum/minimum is reached and TimesTen returns an error.

CACHE CacheValue

CACHE indicates the range of numbers that are cached each time. When a restart occurs, unused cached numbers are lost. If you specify a CacheValue of 1, then each use of the sequence results in an update to the database. Larger cache values result in fewer changes to the database and less overhead. The default is 20.

START WITH StartValue

Specifies the first sequence number to be generated. Use this clause to start an ascending sequence at a value that is greater than the minimum value or to start a descending sequence at a value less than the maximum. The StartValue must be greater or equal MinimumValue and StartValue must be less than or equal to MaximumValue.

BATCH BatchValue

Valid with TimesTen Scaleout only. Configures the range of unique sequence values that are stored at each element of the grid. The default value is 10 million.

Description

  • All parameters in the CREATE SEQUENCE statement must be integer values.

  • If you do not specify a value in the parameters, TimesTen defaults to an ascending sequence that starts with 1, increments by 1, has the default maximum value and does not cycle.

  • Do not create a sequence with the same name as a view or materialized view.

  • Sequences with the CYCLE attribute cannot be replicated (TimesTen Classic).

  • In TimesTen Classic, in which there is a replicated environment for an active standby pair, if DDL_REPLICATION_LEVEL is 3 or greater when you execute CREATE SEQUENCE on the active database, the sequence is replicated to all databases in the replication scheme. To include the sequence in the replication scheme, set DDL_REPLICATION_ACTION to INCLUDE. See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Usage with TimesTen Scaleout

  • The CREATE SEQUENCE statement creates a global object. Once you create the sequence, the sequence values are retrieved from any element of the database.

  • Sequence values are unique, but across elements the values might not be returned in monotonic order. Within a single element, sequence values are in monotonic order. But over time, across elements, sequence values are not returned monotonically. However, the monotonic property is guaranteed within an element.

  • The batch value is the range of unique sequence values stored in the element. Each element has its own batch. An element will get a new batch when its local batch is consumed. There is one element that owns the sequence and is responsible for allocating batch sequence blocks to other elements.

  • For the BATCH clause:

    • Use this clause to specify the range of sequence values that are stored on each element of the grid.

    • The default is 10 million.

    • BatchValue must be greater than or equal to CacheValue.

    • The maximum value for BatchValue is dependent on the maximum value of the signed integer for the platform.

  • Each element in a replica set has its own batch.

  • An element's batch sequence values are recoverable. Cache values are not recoverable.

See "Using sequences" in Oracle TimesTen In-Memory Database Scaleout User's Guide for detailed information and examples.

Using CURRVAL and NEXTVAL in TimesTen Scaleout

To refer to the SEQUENCE values in a SQL statement, use CURRVAL and NEXTVAL.

  • CURRVAL returns the value of the last call to NEXTVAL if there is one in the current session, otherwise it returns an error.

  • NEXTVAL increments the current sequence value by the specified increment and returns the value for each row accessed.

If you execute a single SQL statement with multiple NEXTVAL references, TimesTen only increments the sequence once, returning the same value for all occurrences of NEXTVAL. If a SQL statement contains both NEXTVAL and CURRVAL, NEXTVAL is executed first. CURRVAL and NEXTVAL have the same value in that SQL statement.

NEXTVAL and CURRVAL can be used in the following.

  • The SelectList of a SELECT statement, but not the SelectList of a subquery

  • The SelectList of an INSERT...SELECT statement

  • The SET clause of an UPDATE statement

See "Using sequences" in Oracle TimesTen In-Memory Database Scaleout User's Guide for information on the usage of CURRVAL and NEXTVAL in a grid and for examples.

Using CURRVAL and NEXTVAL in TimesTen Classic

To refer to the SEQUENCE values in a SQL statement, use CURRVAL and NEXTVAL.

  • CURRVAL returns the value of the last call to NEXTVAL if there is one in the current session, otherwise it returns an error.

  • NEXTVAL increments the current sequence value by the specified increment and returns the value for each row accessed.

The current value of a sequence is a connection-specific value. If there are two concurrent connections to the same database, each connection has its own CURRVAL of the same sequence set to its last NEXTVAL reference. When the maximum value is reached, SEQUENCE either wraps or issues an error statement, depending on the value of the CYCLE option of the CREATE SEQUENCE. In the case of recovery, sequences are not rolled back. It is possible that the range of values of a sequence can have gaps; however, each sequence value is still unique.

If you execute a single SQL statement with multiple NEXTVAL references, TimesTen only increments the sequence once, returning the same value for all occurrences of NEXTVAL. If a SQL statement contains both NEXTVAL and CURRVAL, NEXTVAL is executed first. CURRVAL and NEXTVAL have the same value in that SQL statement.

Note:

NEXTVAL cannot be used in a query on a standby node of an active standby pair.

NEXTVAL and CURRVAL can be used in the following.

  • The SelectList of a SELECT statement, but not the SelectList of a subquery

  • The SelectList of an INSERT...SELECT statement

  • The SET clause of an UPDATE statement

Examples: TimesTen Scaleout

For detailed examples, see "Using sequences" in the Oracle TimesTen In-Memory Database Scaleout User's Guide.

Syntax example:

Command> CREATE SEQUENCE mysequence BATCH 100;
Command> describe mysequence;
 
Sequence SAMPLEUSER.MYSEQUENCE:
  Minimum Value: 1
  Maximum Value: 9223372036854775807
  Current Value: 1
  Increment:     1
  Cache:         20
  Cycle:         Off
  Batch:         100
 
1 sequence found.

Examples: TimesTen Classic

Create a sequence.

CREATE SEQUENCE mysequence INCREMENT BY 1 MINVALUE 2 
       MAXVALUE 1000;

This example assumes that tab1 has 1 row in the table and that CYCLE is used:

CREATE SEQUENCE s1 MINVALUE 2 MAXVALUE 4 CYCLE;
SELECT s1.NEXTVAL FROM tab1;
/* Returns the value of 2; */
SELECT s1.NEXTVAL FROM tab1;
/* Returns the value of 3; */
SELECT s1.NEXTVAL FROM tab1;
/* Returns the value of 4; */

After the maximum value is reached, the cycle starts from the minimum value for an ascending sequence.

SELECT s1.NEXTVAL FROM tab1;
/* Returns the value of 2; */

To create a sequence and generate a sequence number:

CREATE SEQUENCE seq INCREMENT BY 1;
INSERT INTO student VALUES (seq.NEXTVAL, 'Sally');

To use a sequence in an UPDATE SET clause:

UPDATE student SET studentno = seq.NEXTVAL WHERE name = 'Sally';

To use a sequence in a query:

SELECT seq.CURRVAL FROM student;

CREATE SYNONYM

The CREATE SYNONYM statement creates a public or private synonym for a database object. A synonym is an alias for a database object. The object can be a table, view, synonym, sequence, PL/SQL stored procedure, PL/SQL function, PL/SQL package, materialized view or cache group.

A private synonym is owned by a specific user and exists in that user's schema. A private synonym is accessible to users other than the owner only if those users have appropriate privileges on the underlying object and specify the schema along with the synonym name.

A public synonym is accessible to all users as long as the user has appropriate privileges on the underlying object.

CREATE SYNONYM is a DDL statement.

Synonyms can be used in these SQL statements:

  • DML statements: SELECT, DELETE, INSERT, UPDATE, MERGE

  • Some DDL statements: GRANT, REVOKE, CREATE TABLE ... AS SELECT, CREATE VIEW ... AS SELECT, CREATE INDEX, DROP INDEX

  • Some cache group statements: LOAD CACHE GROUP, UNLOAD CACHE GROUP, REFRESH CACHE GROUP, FLUSH CACHE GROUP

Required privilege

CREATE SYNONYM (if owner) or CREATE ANY SYNONYM (if not owner) to create a private synonym.

CREATE PUBLIC SYNONYM to create a public synonym.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

CREATE [OR REPLACE] [PUBLIC] SYNONYM [Owner1.]synonym FOR [Owner2.]object

Parameters

Parameter Description

[OR REPLACE]

Specify OR REPLACE to recreate the synonym if it already exists. Use this clause to change the definition of an existing synonym without first dropping it.

[PUBLIC]

Specify PUBLIC to create a public synonym. Public synonyms are accessible to all users, but each user must have appropriate privileges on the underlying object in order to use the synonym.

When resolving references to an object, TimesTen uses a public synonym only if the object is not prefaced by a schema name.

[Owner1.]synonym

Specify the owner of the synonym. You cannot specify an owner if you have specified PUBLIC. If you omit both PUBLIC and Owner1, TimesTen creates the synonym in your own schema.

Specify the name for the synonym, which is limited to 30 bytes.

[Owner2.]object

Specify the owner in which the object resides. Specify the object name for which you are creating a synonym. If you do not qualify object with Owner2, the object is in your own schema. The Owner2 and object do not need to exist when the synonym is created.

Description

  • The schema object does not need to exist when its synonym is created.

  • Do not create a public synonym with the same name as a TimesTen built-in procedure.

  • In order to use the synonym, appropriate privileges must be granted to a user for the object aliased by the synonym before using the synonym.

  • A private synonym cannot have the same name as tables, views, sequences, PLSQL packages, functions, procedures, and cache groups that are in the same schema as the private synonym.

  • A public synonym may have the same name as a private synonym or an object name.

  • If the PassThrough attribute is set so that a query needs to executed in the Oracle database, the query is sent to the Oracle database without any changes. If the query uses a synonym for a table in a cache group, then a synonym with the same name must be defined for the corresponding Oracle database table for the query to be successful.

  • When an object name is used in the DML and DDL statements in which a synonym can be used, the object name is resolved as follows:

    1. Search for a match within the current schema. If no match is found, then:

    2. Search for a match with a public synonym name. If no match is found, then:

    3. Search for a match in the SYS schema. If no match is found, then:

    4. The object does not exist.

    TimesTen creates a public synonym for some objects in the SYS schema. The name of the public synonym is the same as the object name. Thus steps 2 and 3 in the object name resolution can be switched without changing the results of the search.

  • In a replicated environment for an active standby pair, if DDL_REPLICATION_LEVEL is 2 or greater when you execute CREATE SYNONYM on the active database, the synonym is replicated to all databases in the replication scheme. See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Examples

As user ttuser, create a synonym for the jobs table. Verify that you can retrieve the information using the synonym. Display the contents of the SYS.USER_SYNONYMS system view.

Command> CREATE SYNONYM synjobs FOR jobs;
Synonym created.

Command> SELECT FIRST 2 * FROM jobs;
< AC_ACCOUNT, Public Accountant, 4200, 9000 >
< AC_MGR, Accounting Manager, 8200, 16000 >
2 rows found.
Command> SELECT FIRST 2 * FROM synjobs;
< AC_ACCOUNT, Public Accountant, 4200, 9000 >
< AC_MGR, Accounting Manager, 8200, 16000 >
2 rows found.

Command> SELECT * FROM sys.user_synonyms;
< SYNJOBS, TTUSER, JOBS, <NULL> >
1 row found.

Create a public synonym for the employees table.

Command> CREATE PUBLIC SYNONYM pubemp FOR employees;
Synonym created.

Verify that pubemp is listed as a public synonym in the SYS.ALL_SYNONYMS system view.

Command> SELECT * FROM sys.all_synonyms;
< PUBLIC, TABLES, SYS, TABLES, <NULL> >
...
< TTUSER, SYNJOBS, TTUSER, JOBS, <NULL> >
< PUBLIC, PUBEMP, TTUSER, EMPLOYEES, <NULL> >
57 rows found.

Create a synonym for the tab table in the terry schema. Describe the synonym.

Command> CREATE SYNONYM syntab FOR terry.tab;
Synonym created.
Command> DESCRIBE syntab;
 
Synonym TTUSER.SYNTAB:
  For Table TERRY.TAB
  Columns:
    COL1                            VARCHAR2 (10) INLINE
    COL2                            VARCHAR2 (10) INLINE

1 Synonyms found.

Redefine the synjobs synonym to be an alias for the employees table by using the OR REPLACE clause. Describe synjobs.

Command> CREATE OR REPLACE synjobs FOR employees;
Synonym created.
 
Command> DESCRIBE synjobs;
 
Synonym TTUSER.SYNJOBS:
  For Table TTUSER.EMPLOYEES
  Columns:
   *EMPLOYEE_ID                     NUMBER (6) NOT NULL
    FIRST_NAME                      VARCHAR2 (20) INLINE
    LAST_NAME                       VARCHAR2 (25) INLINE NOT NULL
    EMAIL                           VARCHAR2 (25) INLINE UNIQUE NOT NULL
    PHONE_NUMBER                    VARCHAR2 (20) INLINE
    HIRE_DATE                       DATE NOT NULL
    JOB_ID                          VARCHAR2 (10) INLINE NOT NULL
    SALARY                          NUMBER (8,2)
    COMMISSION_PCT                  NUMBER (2,2)
    MANAGER_ID                      NUMBER (6)
    DEPARTMENT_ID                   NUMBER (4)
 
1 Synonyms found.

See also

DROP SYNONYM

CREATE TABLE

The CREATE TABLE statement defines a table.

The CREATE TABLE statement is supported in TimesTen Scaleout and in TimesTen Classic. However, there are differences in syntax and semantics. For simplicity, the supported syntax, parameters, description (semantics), and examples for TimesTen Scaleout and for TimesTen Classic are separated into the usage with TimesTen Scaleout and the usage with TimesTen Classic. While there is repetition in the usages, it is presented this way in order to allow you to progress from syntax to parameters to semantics to examples for each usage.

Review the required privilege section and then see:

Required privilege

CREATE TABLE (if owner) or CREATE ANY TABLE (if not owner).

The owner of the created table must have the REFERENCES privilege on tables referenced by the REFERENCE clause.

In TimesTen Classic:

  • ADMIN privilege is required if replicating a new table across an active standby pair when DDL_REPLICATION_LEVEL=2 or greater and DDL_REPLICATION_ACTION=INCLUDE.

  • These attributes cause the CREATE TABLE to implicitly execute an ALTER ACTIVE STANDBY PAIR... INCLUDE TABLE statement. See "ALTER SESSION" for more details.

After reviewing this section, see:

CREATE TABLE: Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout. Column-based compression and aging are not supported. The using index clause and the distribution clause is not supported for global temporary tables.

See:

SQL syntax for CREATE TABLE: TimesTen Scaleout

The syntax for a persistent table:

CREATE TABLE [Owner.]TableName
(
    ColumnDefinition [,...]
     [PRIMARY KEY (ColumnName [,...]) [UsingIndexClause1]]|
     [[CONSTRAINT ForeignKeyName]
        FOREIGN KEY ([ColumnName] [,...])
        REFERENCES RefTableName
            [(ColumnName [,...])] [ON DELETE CASCADE]] [...]
    
)
[UNIQUE HASH ON (HashColumnName [,...])
    PAGES = PrimaryPages]
[DistributionClause]
[AS SelectQuery]

The syntax for the UsingIndexClause1 is shown below. Note: The CreateIndexStmt is the TimesTen CREATE INDEX statement. See "CREATE INDEX" for details. You must create a unique index as this is a requirement for a primary key.

UsingIndexClause1::= USING INDEX {GLOBAL | LOCAL}| USING INDEX (CreateIndexStmt)
The syntax for the DistributionClause:
DistributionClause::= DISTRIBUTE BY HASH [(ColumnName [,...])] | 
DISTRIBUTE BY REFERENCE [(ForeignKeyConstraint)] | DUPLICATE

Note:

You cannot specify a PRIMARY KEY in both the ColumnDefinition clause and the PRIMARY KEY clause. If you are specifying the UsingIndexClause1 clause, you must specify PRIMARY KEY and PRIMARY KEY must be specified after the ColumnDefinition clause. The UsingIndexClause1 clause cannot be specified as part of the ColumnDefinition clause.

Syntax for global temporary table:

The UsingIndexClause1 and the DistributionClause are not supported for global temporary tables. The syntax is:

CREATE GLOBAL TEMPORARY TABLE [Owner.]TableName
(
    {{ColumnDefinition} [,...]
     [PRIMARY KEY (ColumnName [,...])] |
     [[CONSTRAINT ForeignKeyName]
        FOREIGN KEY ([ColumnName] [,...])
        REFERENCES RefTableName
             [(ColumnName [,...])] [ON DELETE CASCADE]] [...]
      }
)
[UNIQUE HASH ON (HashColumnName [,...])
    PAGES = PrimaryPages]
[ON COMMIT { DELETE | PRESERVE } ROWS ]

Parameters for CREATE TABLE: TimesTen Scaleout

Parameter Description

CREATE TABLE [Owner.]TableName

CREATE TABLE indicates you want to create a table. You must specify a name for the table and optionally the owner of the table.

(ColumnDefinition)

ColumnDefinition indicates the column name, data type, and so on. ColumnDefinition is described in Column definition: TimesTen Scaleout.

If you specify the AS SelectQuery clause, ColumnDefinition is optional.

PRIMARY KEY (ColumnName [,...])

The placement of the PRIMARY KEY keyword after the ColumnDefinition indicates the PRIMARY KEY is specified after the columns are defined. This enables you to specify more than one column for the primary key.

[UsingIndexClause1] UsingIndexClause1 is optional and is described in the next two rows of this table. You cannot specify two USING INDEX clauses in the CREATE TABLE definition. This clause enables you to define a global or local index for the PRIMARY KEY.
USING INDEX {GLOBAL|LOCAL} Part of [UsingIndexClause1]. If specified, indicates if a global or local index is to be created for the primary key.
USING INDEX (CreateIndexStmt) Part of the [UsingIndexClause1] clause. When this USING INDEX clause is specified, the (CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must create a unique index as that is the requirement for a primary key.

CONSTRAINT ForeignKeyName

Specifies an optional user-defined name for a foreign key. If not provided by the user, the system provides a default name.

FOREIGN KEY

This specifies a foreign key constraint between the new table and the referenced table identified by RefTableName. There are two lists of columns specified in the foreign key constraint.

Columns in the first list are columns of the new table and are called the referencing columns. Columns in the second list are columns of the referenced table and are called referenced columns. These two lists must match in data type, including length, precision and scale. The referenced table must already have a primary key or unique index on the referenced column.

The column name list of referenced columns is optional. If omitted, the primary index of RefTableName is used.

The declaration of a foreign key creates a range index on the referencing columns. The user cannot drop the referenced table or its referenced index until the referencing table is dropped.

The foreign key constraint asserts that each row in the new table must match a row in the referenced table such that the contents of the referencing columns are equal to the contents of the referenced columns. Any INSERT, DELETE or UPDATE statements that violate the constraint return TimesTen error 3001.

TimesTen supports SQL-92 "NO ACTION" update and delete rules and ON DELETE CASCADE. Foreign key constraints are not deferrable.

A foreign key can be defined on a global temporary table, but it can only reference a global temporary table. If a parent table is defined with COMMIT DELETE, the child table must also have the COMMIT DELETE attribute.

A foreign key cannot reference an active parent table. An active parent table is one that has some instance materialized for a connection.

If you specify the AS SelectQuery clause, you cannot define a foreign key on the table you are creating.

[ON DELETE CASCADE]

Enables the ON DELETE CASCADE referential action. If specified, when rows containing referenced key values are deleted from a parent table, rows in child tables with dependent foreign key values are also deleted.

UNIQUE HASH ON

Hash index for the table. UNIQUE HASH ON requires that a primary key be defined.

HashColumnName

Column defined in the table that is to participate in the hash key of this table. The columns specified in the hash index must be identical to the columns in the primary key.

If you specify the AS SelectQuery clause, you must define HashColumnName on the table you are creating.

PAGES = PrimaryPages

Sizes the hash index to reflect the expected number of pages in your table. To determine the value for PrimaryPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for PrimaryPages (256000/256=1000).

The value for PrimaryPages must be a positive constant and must be greater than 0.

If your estimate for PrimaryPages is too small, performance may be degraded.

[ON COMMIT {DELETE|PRESERVE} ROWS]

The optional statement specifies whether to delete or preserve rows when a transaction that touches a global temporary table is committed. If not specified, the rows of the temporary table are deleted.

AS SelectQuery

If specified, creates a new table from the contents of the result set of the SelectQuery. The rows returned by SelectQuery are inserted into the table.

Data types and data type lengths are derived from SelectQuery.

SelectQuery is a valid SELECT statement that may or may not contain a subquery.

You can specify a statement level optimizer hint after the SELECT verb. See "Statement level optimizer hints" for information on statement level optimizer hints.

DistributionClause

Supported in TimesTen Scaleout only. There are three options:

  • DISTRIBUTE BY HASH [(ColumnName [,...])]

  • DUPLICATE

  • DISTRIBUTE BY REFERENCE [(ForeignKeyConstraint)]

The DISTRIBUTE BY HASH clause specifies a hash distribution scheme which distributes data based on the hash of the primary key or the hash of the user-defined distribution column(s). Rows are distributed across the replica sets and each row exists in a replica set. The distribution key is optional. If specified, it consists of one or more columns and these columns are used to distribute the data.

The DUPLICATE clause specifies a duplicate distribution scheme which distributes identical copies of data in a table to all elements of the database. All rows of a table exist in each element.

The DISTRIBUTE BY REFERENCE clause specifies a reference distribution scheme which distributes the data of a child table based on the location of the parent row defined by the foreign key constraint. A child table row exists in the same replica set as its parent table. The foreign key constraint is optional in the DISTRIBUTE BY REFERENCE clause. However, if you define more than one foreign key constraint, you must specify one of the foreign key constraints in the DISTRIBUTE BY REFERENCE clause.

If you do not specify a clause, the default is DISTRIBUTE BY HASH.

You must specify the DistributionClause before the AS SelectQuery clause.

You cannot update the distribution key columns.

GLOBAL TEMPORARY

Specifies that the table being created is a global temporary table. A temporary table is similar to a persistent table but it is effectively materialized only when referenced in a connection.

A global temporary table definition is persistent and is visible to all connections, but the table instance is local to each connection. It is created when a command referencing the table is compiled for a connection and dropped when the connection is disconnected. All instances of the same temporary table have the same name but they are identified by an additional connection ID together with the table name. Global temporary tables are allocated in temp space.

The contents of a global temporary table cannot be shared between connections. Each connection sees only its own content of the table and compiled commands that reference temporary tables are not shared among connections.

Operations on temporary tables do generate log records. The amount of log they generate is less than for permanent tables.

The DistributionClause is not supported.

TRUNCATE TABLE is not supported with global temporary tables.

Local temporary tables are not supported.

No object privileges are needed to access global temporary tables.

Do not specify the AS SelectQuery clause with global temporary tables.

ColumnName

Name of the column in a table.

If the name is used in the primary key definition, it forms the primary key for the table to be created. Up to 16 columns can be specified for the primary key. For a foreign key, the ColumnName is optional. If not specified for a foreign key, the reference is to the parent table's primary key.

If you specify the AS SelectQuery clause, you do not have to specify the ColumnName. Do not specify the data type with the AS SelectQuery clause.

Column definition: TimesTen Scaleout

SQL syntax

You can only use the keyword, ENABLE, when defining columns in the CREATE TABLE statement.

The syntax is as follows:

ColumnName ColumnDataType
 [DEFAULT DefaultVal]
 [[NOT] INLINE]
 [PRIMARY KEY | UNIQUE | 
 NULL [UNIQUE] | 
 NOT NULL [ENABLE] [PRIMARY KEY | UNIQUE] 
]

Column definition parameters

The column definition has the following parameters:

Parameter Description

ColumnName

Name to be assigned to one of the columns in the new table. No two columns in the table can be given the same name. A table can have a maximum of 1000 columns.

If you specify the AS SelectQuery clause, ColumnName is optional. The number of column names must match the number of columns in SelectQuery.

ColumnDataType

Type of data the column can contain. Some data types require that you indicate a length. See Data Types for the data types that can be specified.

If you specify the AS SelectQuery clause, do not specify ColumnDataType.

DEFAULT DefaultVal

Indicates that if a value is not specified for the column in an INSERT statement, the default value DefaultVal is inserted into the column. The default value specified must have a type that is compatible with the data type of the column. A default value can be as long as the data type of the associated column allows. You cannot assign a default value for the ROWID data type or for columns in read-only cache groups. In addition, you cannot use a function within the DEFAULT clause.

The following are the supported data types for DefaultVal:

  • NULL

  • Constant expression (an expression that is evaluated to a constant value)

  • SYSDATE and GETDATE ("SYSDATE and GETDATE")

  • SYSTEM_USER ("SYSTEM_USER")

If the default value is one of the users, the data type of the column must be either CHAR or VARCHAR2 and the width of the column must be at least 30 characters.

If you specify the AS SelectQuery clause, optionally, you can specify the DEFAULT clause on the table you are creating.

INLINE|NOT INLINE

By default, variable-length columns whose declared column length is greater than 128 bytes are stored out of line. Variable-length columns whose declared column length is less than or equal to 128 bytes are stored inline. The default behavior can be overridden during table creation through the use of the INLINE and NOT INLINE keywords.

If you specify the AS SelectQuery clause, optionally, you can specify the INLINE | NOT INLINE clause on the table you are creating.

NULL

Indicates that the column can contain NULL values.

If you specify the AS SelectQuery clause, optionally, you can specify NULL on the table you are creating.

If you specify NULL, you cannot specify ENABLE.

NOT NULL [ENABLE]

Indicates that the column cannot contain NULL values. If NOT NULL is specified, any statement that attempts to place a NULL value in the column is rejected.

If you specify the AS SelectQuery clause, optionally, you can specify NOT NULL [ENABLE] on the table you are creating.

If you specify NOT NULL, you can optionally specify ENABLE. Because NOT NULL constraints are always enabled, you are not required to specify ENABLE.

You can only use the keyword, ENABLE, when defining columns in the CREATE TABLE statement.

UNIQUE

A unique constraint placed on the column. No two rows in the table may have the same value for this column. TimesTen creates a unique range index to enforce uniqueness. So a column with a unique constraint can use more memory and time during execution than a column without the constraint. Cannot be used with PRIMARY KEY.

If you specify the AS SelectQuery clause, optionally, you can specify UNIQUE on the table you are creating.

PRIMARY KEY

A unique NOT NULL constraint placed on the column. No two rows in the table may have the same value for this column. Cannot be used with UNIQUE.

If you specify the AS SelectQuery clause, optionally, you can specify PRIMARY KEY on the table you are creating.

Description for USING INDEX clauses in CREATE TABLE: TimesTen Scaleout

You have the option of specifying an additional clause after the PRIMARY KEY clause in your CREATE TABLE definition. This clause enables you to specify a global or local index for the primary key constraint.
  • The USING INDEX {GLOBAL | LOCAL} clause is one option that enables you to specify a global or local index for the primary key constraint. You must specify the GLOBAL or the LOCAL keyword. You can optionally specify the USE HASH INDEX clause after the USING INDEX {GLOBAL | LOCAL} clause if you want to define a hash index.

  • The USING INDEX (CreateIndexStmt) clause is your other option for specifying a global or local index. The(CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must create a unique index as that is the requirement for a primary key constraint. If you use the CREATE INDEX statement to create a hash index, see "CREATE INDEX" for information on the CREATE INDEX statement.

Note:

You cannot use both the USING INDEX {GLOBAL | LOCAL} and the USING INDEX (CreateIndexStmt) in the CREATE TABLE definition. Specify one clause or the other or specify neither.
When you create a hash index, the hash index must be sized. TimesTen provides the PAGES= clause for this purpose. Consider these options:
  • If you specify the USING INDEX...CreateIndexStmt clause to create a HASH index, you have the option of specifying the PAGES= clause. If you do not specify the PAGES= clause, TimesTen uses PAGES=CURRENT as the default to size the hash index. (See "CREATE INDEX" for details on the CREATE INDEX statement.)

  • If you specify the UNIQUE HASH ON clause (part of the CREATE TABLE definition), you must specify the PAGES= clause to size the hash index.

  • If you specify both the USING INDEX...CreateIndexStmt and the UNIQUE HASH ON clause (part of the CREATE TABLE definition), TimesTen uses the value specified in the UNIQUE HASH ON...PAGES= clause to size the hash index. TimesTen also issues a warning that there was a different number of pages specified for the hash index and it is using the value specified in the UNIQUE HASH ON...PAGES= clause.

In this example, the PAGES= clause is specified in both the CreateIndexStmt clause (PAGES=200) and the UNIQUE HASH ON clause (PAGES=400). TimesTen issues a warning and uses PAGES=400 to size the hash index:
Command> CREATE TABLE mytab (col1 TT_INTEGER, col2 TT_INTEGER, PRIMARY KEY (col1, col2)
           USING INDEX (CREATE GLOBAL UNIQUE HASH INDEX myindex on mytab (col1,col2) PAGES=200)) 
           UNIQUE HASH ON (col1,col2) PAGES=400 DISTRIBUTE BY HASH (col1);
Warning  2252: Different number of pages specified for hash index MYINDEX in table and index definition. 
Index created with pages = 400
Restrictions:
  • The USING INDEX clause cannot be used for foreign key constraints on a table.

  • The USING INDEX clause cannot be used with views.

See "CREATE INDEX" for information on global and local indexes and their use in TimesTen Scaleout.

Additional information for CREATE TABLE: TimesTen Scaleout

  • TimesTen Scaleout distributes data by one of three distribution schemes:

    • Hash: TimesTen Scaleout distributes data based on the hash of the primary key column(s) or one or more columns you specify in the DISTRIBUTED BY HASH clause. A given row is stored in a replica set. Rows are evenly distributed across the replica sets. Hash is the default distribution scheme as it is appropriate for most tables.

    • Reference: TimesTen Scaleout distributes data of a child table based on the location of the parent table that is identified by the foreign key. A given row of a child table is present in the same replica set as its parent table. This distribution scheme optimizes joins by distributing related data within a single replica set. You can distribute the parent table by hash or reference. The parent is called the root table if it is distributed by hash. You must define the child (foreign) key columns as NOT NULL.

    • Duplicate: TimesTen Scaleout distributes full identical copies of data to all elements of the database. All rows are present in all elements. This distribution scheme optimizes the performance of reads by storing identical data in every data instance. This distribution scheme is appropriate for tables that are relatively small, frequently read, and infrequently modified.

    See "Defining the distribution scheme for tables" and "Defining table distribution schemes" in the Oracle TimesTen In-Memory Database Scaleout User's Guide for more information.

  • For tables with a hash distribution scheme:

    • The distribution key is used if specified.

    • The primary key is used if the distribution key is not specified.

    • A hidden column is used if there is no primary key or distribution key. Data is distributed randomly and evenly.

    You should specify a distribution key if there is a primary key defined on the table, but the primary key is not the best way to distribute the data. If there is no primary key, but there is a unique column, then you may want to distribute the data on this unique column. If there is no primary key and no unique column, then do not specify a distribution key. TimesTen Scaleout distributes the data on the hidden column.

  • If the distribution scheme is by reference:

    • Only a single foreign key constraint can be referenced in the DISTRIBUTE BY REFERENCE clause. There may be multiple foreign key constraints in the child table, but only one can be used to determine the reference distribution.

    • A referenced foreign key constraint must be named in the constraint clause if there is more than one.

    • The foreign key constraint in the reference distribution clause must reference the primary key or a unique key of the parent table. If the parent table is the root, the referenced key must be the distribution key.

    • You can create a foreign key relationship to a non distribution key column of the parent table, but you cannot then distribute by reference based on this foreign key relationship.

    • You cannot update the foreign key column that is used in the DISTRIBUTE BY REFERENCE clause.

  • If you are planning to load your tables with data, consider creating your tables without indexes. After the data is loaded, you can then create your indexes. This reduces the time it take to load the data into the tables. The exception is if you are using foreign keys and reference tables.
  • You can use the CREATE TABLE...AS SELECT statement to create a new table based on the definition of the original table. Note that primary key constraints are not carried over to the new table so how the data is distributed changes if you do not define a primary key constraint on the new table.

    See "Use CREATE TABLE...AS SELECT" for more information.

  • You cannot update the distribution key column(s) unless you update the column(s) to the same value.

  • All columns participating in the primary key are NOT NULL.

  • A PRIMARY KEY that is specified in the ColumnDefinition can only be specified for one column.

  • You cannot specify a PRIMARY KEY in both the ColumnDefinition clause and the PRIMARY KEY clause.

  • For both primary key and foreign key constraints, duplicate column names are not allowed in the constraint column list.

  • You cannot update primary key column(s) unless you update the column(s) to the same value.

  • There are performance considerations when you define out of line columns instead of inline columns:

    • Accessing data is slower because TimesTen does not store data contiguously with out of line columns.

    • Populating data is slower because TimesTen generates more logging operations.

    • Deleting data is slower because TimesTen performs more reclaim and logging operations.

    • Storing a column requires less overhead.

  • If ON DELETE CASCADE is specified on a foreign key constraint for a child table, a user can delete rows from a parent table for which the user has the DELETE privilege without requiring explicit DELETE privilege on the child table.

  • To change the ON DELETE CASCADE triggered action, drop then redefine the foreign key constraint.

  • You cannot create a table that has a foreign key referencing a cached table.

  • UNIQUE column constraint and default column values are not supported with materialized views.

  • Use the ALTER TABLE statement to change the representation of the primary key index for a table.

  • If you specify the AS SelectQuery clause:

    • Data types and data type lengths are derived from the SelectQuery. Do not specify data types on the columns of the table you are creating.

    • TimesTen defines on columns in the new table NOT NULL constraints that were explicitly created on the corresponding columns of the selected table if SelectQuery selects the column rather than an expression containing the column.

    • NOT NULL constraints that were implicitly created by TimesTen on columns of the selected table (for example, primary keys) are carried over to the new table. You can override the NOT NULL constraint on the selected table by defining the new column as NULL. For example:

      CREATE TABLE newtable (newcol NULL) AS SELECT (col) FROM tab;
      
    • NOT INLINE/INLINE attributes are carried over to the new table.

    • Unique keys, foreign keys, indexes and column default values are not carried over to the new table.

    • If all expressions in SelectQuery are columns, rather than expressions, then you can omit the columns from the table you are creating. In this case, the name of the columns are the same as the columns in SelectQuery. If the SelectQuery contains an expression rather than a simple column reference, either specify a column alias or name the column in the CREATE TABLE statement.

    • Do not specify foreign keys on the table you are creating.

    • Do not specify the SELECT FOR UPDATE clause in SelectQuery.

    • The ORDER BY clause is not supported when you use the AS SelectQuery clause.

    • SelectQuery cannot contain set operators UNION, MINUS, INTERSECT.

  • By default, a range index is created to enforce the primary key. Use the UNIQUE HASH clause to specify a hash index for the primary key.

    • If your application performs range queries using a table's primary key, then choose a range index for that table by omitting the UNIQUE HASH clause.

    • If your application performs only exact match lookups on the primary key, then a hash index may offer better response time and throughput. In such a case, specify the UNIQUE HASH clause.

  • A hash index is created with a fixed size that remains constant for the life of the table or until the hash index is resized with the ALTER TABLE statement or when the index is dropped and recreated. A smaller hash index results in more hash collisions. A larger hash index reduces collisions but can waste memory. Hash key comparison is a fast operation, so a small number of hash collisions should not cause a performance problem for TimesTen.

    To ensure that your hash index is sized correctly, your application must indicate the expected size of your table with the value of the RowPages parameter of the SET PAGES clause. Compute this value by dividing the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for the value of RowPages (256000/256=1000).

  • At most 16 columns are allowed in a hash key.

  • ON DELETE CASCADE is supported on detail tables of a materialized view. If you have a materialized view defined over a child table, a deletion from the parent table causes cascaded deletes in the child table. This, in turn, triggers changes in the materialized view.

  • The total number of rows reported by the DELETE statement does not include rows deleted from child tables as a result of the ON DELETE CASCADE action.

  • For ON DELETE CASCADE: Since different paths may lead from a parent table to a child table, the following rule is enforced:

    • Either all paths from a parent table to a child table are "delete" paths or all paths from a parent table to a child table are "do not delete" paths. Specify ON DELETE CASCADE on all child tables on the "delete" path.

    • This rule does not apply to paths from one parent to different children or from different parents to the same child.

  • For ON DELETE CASCADE, the following rule is also enforced.

    • If a table is reached by a "delete" path, then all its children are also reached by a "delete" path.

  • The data in a global temporary table is private to the current connection and does not need to be secured between users. Thus, global temporary tables do not require object privileges.

Examples: Global and local indexes in TimesTen Scaleout

These examples show various uses of the syntax for using global indexes with CREATE TABLE...PRIMARY KEY.

Create a table specifying a primary key. Use the USING INDEX GLOBAL clause to create a global range index. The index must be unique as is a requirement for a primary key.
Command> CREATE TABLE mytab (c TT_INTEGER, b TT_INTEGER, a TT_INTEGER, 
           PRIMARY KEY (c,b) USING INDEX GLOBAL) DISTRIBUTE BY HASH (a,b);
Command> indexes mytab;

Indexes on table SAMPLEUSER.MYTAB:
  MYTAB: global unique range index on columns:
    C
    B
  1 index found.

1 index found on 1 table
Command> DROP TABLE mytab;

Create a table specifying a primary key. Use the USING INDEX LOCAL clause to create a local range index. The index must be unique as is a requirement for a primary key.

Command> CREATE TABLE mytab (c TT_INTEGER, b TT_INTEGER, a TT_INTEGER, 
  PRIMARY KEY (c,b) USING INDEX LOCAL DISTRIBUTE BY HASH (a,b);
Command> indexes mytab;

Indexes on table SAMPLEUSER.MYTAB:
  MYTAB: unique range index on columns:
    C
    B
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab;

Create a table specifying a primary key. Use the USING INDEX (CreateIndexStmt) clause to create a global range index. The (CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must specify the UNIQUE keyword when creating the index. This is a requirement for a primary key. See "CREATE INDEX" for information on this statement.

Command> CREATE TABLE mytab (c TT_INTEGER, b TT_INTEGER, a TT_INTEGER, 
           PRIMARY KEY (c,b) USING INDEX (CREATE GLOBAL UNIQUE INDEX GlobalUniqueIdx ON mytab (c,b))) DISTRIBUTE  BY HASH (a,b);
Command> indexes mytab;

Indexes on table SAMPLEUSER.MYTAB:
  GLOBALUNIQUEIDX: global unique range index on columns:
    C
    B
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab;

Create a table specifying a primary key. Use the USING INDEX (CreateIndexStmt) clause to create a global range index. The (CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. The CREATE INDEX definition specifies the INCLUDE clause to include additional column(s) in the index definition. You must specify the UNIQUE keyword when creating the index. This is a requirement for a primary key. See "CREATE INDEX" for information on this statement.

Command> CREATE TABLE mytab (c TT_INTEGER, b TT_INTEGER, a TT_INTEGER, 
           PRIMARY KEY (c,b) USING INDEX (CREATE GLOBAL UNIQUE INDEX GlobalUniqueIdx 
           ON mytab (c,b) INCLUDE (a))) DISTRIBUTE  BY HASH (a,b);
Command> indexes mytab;

Indexes on table SAMPLEUSER.MYTAB:
  GLOBALUNIQUEIDX: global unique range index on columns:
    C
    B
    Included columns:
      A
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab;

Create a table specifying a primary key. Use the USING INDEX (CreateIndexStmt) clause to create a global unique hash index. The (CreateIndexStmt) clause indicates that you want to define the index according to the TimesTen CREATE INDEX statement. The parentheses ( ) are required. You must specify the UNIQUE keyword when creating the index. This is a requirement for a primary key. See "CREATE INDEX" for information on this statement.

Command> CREATE TABLE mytab (c TT_INTEGER, b TT_INTEGER, a TT_INTEGER, 
           PRIMARY KEY (c,b) USING INDEX (CREATE GLOBAL UNIQUE HASH INDEX GlobalUniqueIdx 
           ON mytab (c,b))) DISTRIBUTE  BY HASH (a,b);
Command> indexes mytab;

Indexes on table SAMPLEUSER.MYTAB:
  GLOBALUNIQUEIDX: global unique hash index on columns:
    C
    B
  1 index found.

1 index found on 1 table.
Command> DROP TABLE mytab;

Additional examples: TimesTen Scaleout

These examples illustrate how to create tables with the duplicate, hash, and reference distribution schemes.

These examples illustrate how to create tables with the DISTRIBUTE BY REFERENCE distribution scheme:

"Use CREATE TABLE...AS SELECT" shows how to use the CREATE TABLE...AS SELECT clause in TimesTen Scaleout.

Create the account_type table

This example runs ttIsql to create the account_type table and use a duplicate distribution scheme to distribute the data. This table contains few rows and uses a duplicate distribution scheme to optimize reads. Copies of the data in the table are distributed to all elements of the database.

Command> CREATE TABLE account_type ( type CHAR(1) NOT NULL PRIMARY KEY,
           description VARCHAR2(100) NOT NULL) DUPLICATE;

Create the account_status table

This example runs ttIsql to create the account_status table and use a duplicate distribution scheme. The table size is small and uses a distribution scheme to optimize reads. Copies of the data in the table are distributed to all elements of the database.

Command> CREATE TABLE account_status(status NUMBER(2) NOT NULL PRIMARY KEY,
           description VARCHAR2(100) NOT NULL) DUPLICATE;

Create the customers table

This example runs ttIsql to create the customers table and distributes the table by hash. The data in the table is distributed to each element based on the hash of the cust_id column (the primary key).

Command> CREATE TABLE customers(cust_id NUMBER(10,0) NOT NULL PRIMARY KEY,
           first_name VARCHAR2(30) NOT NULL,last_name VARCHAR2(30) NOT NULL,
           addr1 VARCHAR2(64),addr2 VARCHAR2(64), zipcode VARCHAR2(5),
           member_since DATE NOT NULL)
         DISTRIBUTE BY HASH;

Create the accounts table

This example runs ttIsql to create the accounts table and defines three primary/foreign key relationships. The accounts table is distributed by reference and the data is distributed based on the fk_customer foreign key constraint. This scheme optimizes the performance of joins by distributing the data in the accounts table based on the location of the corresponding value of the customers.cust_id parent column (of the fk_customer foreign key constraint). The row of a child table exists in the same replica set as the parent table. If the join is performed on the primary or foreign key, the data is stored on one element, so TimesTen Scaleout does not have to access different elements.

Command> CREATE TABLE accounts(account_id NUMBER(10,0) NOT NULL PRIMARY KEY,
           phone VARCHAR2(15) NOT NULL,account_type CHAR(1) NOT NULL,
           status NUMBER(2) NOT NULL,current_balance NUMBER(10,2) NOT NULL,
           prev_balance NUMBER(10,2) NOT NULL,date_created DATE NOT NULL,
           cust_id NUMBER(10,0) NOT NULL,
         CONSTRAINT fk_customer FOREIGN KEY (cust_id)
         REFERENCES customers(cust_id),CONSTRAINT fk_acct_type
         FOREIGN KEY (account_type)
         REFERENCES account_type(type),
         CONSTRAINT fk_acct_status
         FOREIGN KEY (status)
         REFERENCES account_status(status)
         )
         DISTRIBUTE BY REFERENCE (fk_customer);

Create the transactions table

This example runs ttIsql to create the transactions table. The transactions table is distributed by reference and the data is distributed based on the fk_accounts foreign key constraint. This scheme optimizes the performance of joins by distributing the data in the transaction table based on the location of the corresponding value of the accounts.account_id parent column (of the fk_accounts foreign key constraint). The row of a child table exists in the same replica set as the parent table. If the join is performed on the primary or foreign key, the data is stored on one element, so TimesTen Scaleout does not have to access different elements.

The accounts parent table is also distributed by reference. This defines a two level distribute by reference distribution hierarchy.

Command> CREATE TABLE transactions(transaction_id NUMBER(10,0) NOT NULL,
           account_id NUMBER(10,0) NOT NULL ,
           transaction_ts TIMESTAMP NOT NULL,
           description VARCHAR2(60),
           optype CHAR(1) NOT NULL,
           amount NUMBER(6,2) NOT NULL,
         PRIMARY KEY (account_id, transaction_id, transaction_ts),
         CONSTRAINT fk_accounts FOREIGN KEY (account_id)
         REFERENCES accounts(account_id)
         )
         DISTRIBUTE BY REFERENCE (fk_accounts);

View the tables

This example runs the ttIsql tables command to view the tables in the database.

Command> tables;
  SAMPLEUSER.ACCOUNTS
  SAMPLEUSER.ACCOUNT_STATUS
  SAMPLEUSER.ACCOUNT_TYPE
  SAMPLEUSER.CUSTOMERS
  SAMPLEUSER.TRANSACTIONS
5 tables found.

View the definition of the accounts table

This example runs the ttIsql describe command to view the definition of the accounts table.

Command> describe accounts;
 
Table SAMPLEUSER.ACCOUNTS:
  Columns:
   *ACCOUNT_ID                      NUMBER (10) NOT NULL
    PHONE                           VARCHAR2 (15) INLINE NOT NULL
    ACCOUNT_TYPE                    CHAR (1) NOT NULL
    STATUS                          NUMBER (2) NOT NULL
    CURRENT_BALANCE                 NUMBER (10,2) NOT NULL
    PREV_BALANCE                    NUMBER (10,2) NOT NULL
    DATE_CREATED                    DATE NOT NULL
    CUST_ID                         NUMBER (10) NOT NULL
  DISTRIBUTE BY REFERENCE (FK_CUSTOMER)
 
1 table found.
(primary key columns are indicated with *)

DISTRIBUTE BY REFERENCE with one foreign key

This example illustrates that you do not have to specify the foreign key constraint in the DISTRIBUTE BY REFERENCE clause. There is only one foreign key.

First create the Orders table and distribute by hash.

Command> CREATE TABLE Orders
           (OrderId TT_INTEGER NOT NULL PRIMARY KEY,
            OrderDate DATE NOT NULL,
            discount BINARY_FLOAT)
         DISTRIBUTE BY HASH;

Create the OrderDetails table with one foreign key constraint. There is no need to name the constraint in the distribution clause.

Command> CREATE TABLE OrderDetails
           (OrderId TT_INTEGER NOT NULL,
            PartId TT_INTEGER NOT NULL,
            Quantity TT_INTEGER NOT NULL,
         FOREIGN KEY (OrderId)
         REFERENCES Orders (OrderId))
         DISTRIBUTE BY REFERENCE;

Run the ttIsql describe command to view the tables.

Command> describe Orders;
 
Table SAMPLEUSER.ORDERS:
  Columns:
   *ORDERID                         TT_INTEGER NOT NULL
    ORDERDATE                       DATE NOT NULL
    DISCOUNT                        BINARY_FLOAT
  DISTRIBUTE BY HASH (ORDERID)
 
1 table found.
(primary key columns are indicated with *)

Command> describe OrderDetails;
 
Table SAMPLEUSER.ORDERDETAILS:
  Columns:
    ORDERID                         TT_INTEGER NOT NULL
    PARTID                          TT_INTEGER NOT NULL
    QUANTITY                        TT_INTEGER NOT NULL
  DISTRIBUTE BY REFERENCE
 
1 table found.
(primary key columns are indicated with *)

Table with more than one foreign key

This example illustrates that if a table contains more than one foreign key constraint, the DISTRIBUTE BY REFERENCE clause must name the foreign key constraint that will be used as the reference. The customers2 table is the parent and is distributed by hash. The OrderDetails2 table contains two foreign key constraints and this table is distributed by reference on the c1_1 constraint. This constraint must be included in the DISTRIBUTED BY REFERENCE clause.

Command> CREATE TABLE customers2 (CustomerId TT_INTEGER NOT NULL PRIMARY KEY,
           LastOrderDate DATE NOT NULL,PromotionDiscount BINARY_FLOAT)
         DISTRIBUTE BY HASH;

Command> CREATE TABLE OrderDetails2 (OrderId TT_INTEGER NOT NULL,
           CustomerId TT_INTEGER NOT NULL, Quantity TT_INTEGER NOT NULL,
         CONSTRAINT c1_1 FOREIGN KEY (OrderId) 
         REFERENCES Orders (OrderId),
         CONSTRAINT c2_2 FOREIGN KEY (CustomerId) 
         REFERENCES Customers2 (CustomerId))
         DISTRIBUTE BY REFERENCE (c1_1);

Foreign key relationship not on distribution key of the parent table

This example creates the orders2 parent table with the OrderId primary key and the CouponId unique key. The table is distributed by hash. Since no distribution key is specified, the data is distributed by hash on the OrderId primary key. The coupons child table establishes a foreign key relationship on the CouponId unique key. Since this key is not the distribution key of the orders2 parent table, TimesTen Scaleout throws an error.

Command> CREATE TABLE Orders2 (OrderId TT_INTEGER NOT NULL PRIMARY KEY,
           CouponId TT_INTEGER NOT NULL UNIQUE, OrderDate DATE NOT NULL,
           discount BINARY_FLOAT)
         DISTRIBUTE BY HASH;

Command> CREATE TABLE Coupons (CouponId TT_INTEGER NOT NULL, 
           discount BINARY_FLOAT,
         CONSTRAINT CouponC1 FOREIGN KEY (CouponId) 
         REFERENCES Orders2 (CouponId) )
         DISTRIBUTE BY REFERENCE (CouponC1);
 1067: The Parent keys for a distribute by reference table with hash distributed
parent must include the distribution keys of the parent.
The command failed.

Using first and second level child foreign key relationship

This example creates the Coupons2 parent table and distributes the data by hash. The Orders3 child table is created as a first level foreign key relationship and the parent table (Coupons2) is the root table. The OrderDetails3 child table is created as a second level foreign key relationship and the parent table (Orders3) is a reference table.

Command> CREATE TABLE Coupons2 (CouponId TT_INTEGER NOT NULL PRIMARY KEY, 
           discount BINARY_FLOAT)
         DISTRIBUTE BY HASH;

Command> CREATE TABLE Orders3 (OrderId TT_INTEGER NOT NULL PRIMARY KEY,
           CouponId TT_INTEGER NOT NULL, OrderDate DATE NOT NULL,
           discount BINARY_FLOAT, CONSTRAINT c1_coupons FOREIGN KEY (CouponId)
         REFERENCES Coupons2 (CouponId))
         DISTRIBUTE BY REFERENCE (c1_coupons);

Command> CREATE TABLE OrderDetails3 (OrderId TT_INTEGER NOT NULL,
           PartId TT_INTEGER NOT NULL, quantity TT_INTEGER NOT NULL,
         CONSTRAINT c1_orders FOREIGN KEY (OrderId)
         REFERENCES Orders3 (OrderId))
         DISTRIBUTE BY REFERENCE (C1_orders);

Use CREATE TABLE...AS SELECT

This example creates the NewCustomers table based on the customers table. It defines a primary key constraint to maintain the same distribution scheme and ensure the data is distributed on the primary key.

Command> CREATE TABLE NewCustomers(cust_id PRIMARY KEY, first_name, last_name, 
           addr1, addr2, zipcode, member_since) AS SELECT * FROM customers;
0 rows inserted.
Command> describe NewCustomers;
 
Table SAMPLEUSER.NEWCUSTOMERS:
  Columns:
   *CUST_ID                         NUMBER (10) NOT NULL
    FIRST_NAME                      VARCHAR2 (30) INLINE NOT NULL
    LAST_NAME                       VARCHAR2 (30) INLINE NOT NULL
    ADDR1                           VARCHAR2 (64) INLINE
    ADDR2                           VARCHAR2 (64) INLINE
    ZIPCODE                         VARCHAR2 (5) INLINE
    MEMBER_SINCE                    DATE NOT NULL
  DISTRIBUTE BY HASH (CUST_ID)
 
1 table found.
(primary key columns are indicated with *)

Run ttIsql describe to view the original customers table:

Command> describe Customers;

Table SAMPLEUSER.CUSTOMERS:
  Columns:
   *CUST_ID                         NUMBER (10) NOT NULL
    FIRST_NAME                      VARCHAR2 (30) INLINE NOT NULL
    LAST_NAME                       VARCHAR2 (30) INLINE NOT NULL
    ADDR1                           VARCHAR2 (64) INLINE
    ADDR2                           VARCHAR2 (64) INLINE
    ZIPCODE                         VARCHAR2 (5) INLINE
    MEMBER_SINCE                    DATE NOT NULL
  DISTRIBUTE BY HASH (CUST_ID)
 
1 table found.
(primary key columns are indicated with *)

SQL syntax for CREATE TABLE: TimesTen Classic

You cannot specify a PRIMARY KEY in both the ColumnDefinition clause and the PRIMARY KEY clause.

The syntax for a persistent table:

CREATE TABLE [Owner.]TableName
(
    {{ColumnDefinition} [,...]
     [PRIMARY KEY (ColumnName [,...]) |
     [[CONSTRAINT ForeignKeyName]
        FOREIGN KEY ([ColumnName] [,...])
        REFERENCES RefTableName
            [(ColumnName [,...])] [ON DELETE CASCADE]] [...]
    }
)
[ColumnBasedCompression]
[UNIQUE HASH ON (HashColumnName [,...])
    PAGES = PrimaryPages]
[AGING {LRU|
         USE ColumnName
             LIFETIME Num1 {SECOND[S] | MINUTE[S] | HOUR[S] |DAY[S]}
               [CYCLE Num2 {SECOND[S] | MINUTE[S] |HOUR[S] |DAY[S]}]
       }[ON|OFF]
] 
[AS SelectQuery]

The syntax for a global temporary table is:

CREATE GLOBAL TEMPORARY TABLE [Owner.]TableName
(
    {{ColumnDefinition} [,...]
     [PRIMARY KEY (ColumnName [,...]) |
     [[CONSTRAINT ForeignKeyName]
        FOREIGN KEY ([ColumnName] [,...])
        REFERENCES RefTableName
             [(ColumnName [,...])] [ON DELETE CASCADE]] [...]
      }
)
[UNIQUE HASH ON (HashColumnName [,...])
    PAGES = PrimaryPages]
[ON COMMIT { DELETE | PRESERVE } ROWS]

Parameters for CREATE TABLE: TimesTen Classic

Parameter Description

[Owner.]TableName

Name to be assigned to the new table. Two tables cannot have the same owner name and table name.

If you do not specify the owner name, your login name becomes the owner name for the new table. Owners of tables in TimesTen are determined by the user ID settings or login names. Oracle Database table owner names must always match TimesTen table owner names.

See "Basic names" for rules on defining names.

GLOBAL TEMPORARY

Specifies that the table being created is a global temporary table. A temporary table is similar to a persistent table but it is effectively materialized only when referenced in a connection.

A global temporary table definition is persistent and is visible to all connections, but the table instance is local to each connection. It is created when a command referencing the table is compiled for a connection and dropped when the connection is disconnected. All instances of the same temporary table have the same name but they are identified by an additional connection ID together with the table name. Global temporary tables are allocated in temp space.

The contents of a global temporary table cannot be shared between connections. Each connection sees only its own content of the table and compiled commands that reference temporary tables are not shared among connections.

When DDL_REPLICATION_LEVEL is 2 or greater, the creation of a global temporary table is replicated in an active standby pair, but the global temporary table is not included in the replication scheme.

Temporary tables are automatically excluded from active standby pairs or when the DATASTORE element has been specified.

A cache group table cannot be defined as a temporary table.

Changes to temporary tables cannot be tracked with XLA.

Operations on temporary tables do generate log records. The amount of log they generate is less than for permanent tables.

Truncate table is not supported with global temporary tables.

Local temporary tables are not supported.

No object privileges are needed to access global temporary tables.

Do not specify the AS SelectQuery clause with global temporary tables.

ColumnDefinition

An individual column in a table. Each table must have at least one column.

If you specify the AS SelectQuery clause, ColumnDefinition is optional.

ColumnName

Name of the column in a table. Is used in various clauses of the CREATE TABLE statement.

If the name is used in the primary key definition, it forms the primary key for the table to be created. Up to 16 columns can be specified for the primary key. For a foreign key, the ColumnName is optional. If not specified for a foreign key, the reference is to the parent table's primary key.

If you specify the AS SelectQuery clause, you do not have to specify the ColumnName. Do not specify the data type with the AS SelectQuery clause.

PRIMARY KEY

PRIMARY KEY may only be specified once in a table definition. It provides a way of identifying one or more columns that, together, form the primary key of the table. The contents of the primary key have to be unique and NOT NULL. You cannot specify a column as both UNIQUE and a single column PRIMARY KEY.

CONSTRAINT ForeignKeyName

Specifies an optional user-defined name for a foreign key. If not provided by the user, the system provides a default name.

FOREIGN KEY

This specifies a foreign key constraint between the new table and the referenced table identified by RefTableName. There are two lists of columns specified in the foreign key constraint.

Columns in the first list are columns of the new table and are called the referencing columns. Columns in the second list are columns of the referenced table and are called referenced columns. These two lists must match in data type, including length, precision and scale. The referenced table must already have a primary key or unique index on the referenced column.

The column name list of referenced columns is optional. If omitted, the primary index of RefTableName is used.

The declaration of a foreign key creates a range index on the referencing columns. The user cannot drop the referenced table or its referenced index until the referencing table is dropped.

The foreign key constraint asserts that each row in the new table must match a row in the referenced table such that the contents of the referencing columns are equal to the contents of the referenced columns. Any INSERT, DELETE or UPDATE statements that violate the constraint return TimesTen error 3001.

TimesTen supports SQL-92 "NO ACTION" update and delete rules and ON DELETE CASCADE. Foreign key constraints are not deferrable.

A foreign key can be defined on a global temporary table, but it can only reference a global temporary table. If a parent table is defined with COMMIT DELETE, the child table must also have the COMMIT DELETE attribute.

A foreign key cannot reference an active parent table. An active parent table is one that has some instance materialized for a connection.

If you specify the AS SelectQuery clause, you cannot define a foreign key on the table you are creating.

[ON DELETE CASCADE]

Enables the ON DELETE CASCADE referential action. If specified, when rows containing referenced key values are deleted from a parent table, rows in child tables with dependent foreign key values are also deleted.

ColumnBasedCompression

Defines compression at the column level, which stores data more efficiently. Eliminates redundant storage of duplicate values within columns and improves the performance of SQL queries that perform full table scans. See "Column-based compression of tables (TimesTen Classic)" for details.

UNIQUE

UNIQUE provides a way of identifying a column where each row must contain a unique value.

UNIQUE HASH ON

Hash index for the table. This parameter is used for equality predicates. UNIQUE HASH ON requires that a primary key be defined.

HashColumnName

Column defined in the table that is to participate in the hash key of this table. The columns specified in the hash index must be identical to the columns in the primary key.

If you specify the AS SelectQuery clause, you must define HashColumnName on the table you are creating.

PAGES = PrimaryPages

Sizes the hash index to reflect the expected number of pages in your table. To determine the value for PrimaryPages, divide the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for PrimaryPages (256000/256=1000).

The value for PrimaryPages must be a positive constant and must be greater than 0.

If your estimate for PrimaryPages is too small, performance may be degraded.

[ON COMMIT {DELETE|PRESERVE} ROWS]

The optional statement specifies whether to delete or preserve rows when a transaction that touches a global temporary table is committed. If not specified, the rows of the temporary table are deleted.

[AGING LRU [ON|OFF]]

If specified, defines the LRU aging policy for the table. The LRU aging policy defines the type of aging (least recently used (LRU)), the aging state (ON or OFF) and the LRU aging attributes.

Set the aging state to either ON or OFF. ON indicates that the aging state is enabled and aging is done automatically. OFF indicates that the aging state is disabled and aging is not done automatically. In both cases, the aging policy is defined. The default is ON.

LRU attributes are defined by calling the ttAgingLRUConfig and/or the ttAgingTableLRUConfig built-in procedures. LRU attributes are not defined at the SQL level. See "ttAgingLRUConfig" and "ttAgingTableLRUConfig" in the Oracle TimesTen In-Memory Database Reference and "Implementing an aging policy in your tables" in the Oracle TimesTen In-Memory Database Operations Guide for more information.

[AGING USE ColumnName...[ON|OFF]]

If specified, defines the time-based aging policy for the table. The time-based aging policy defines the type of aging (time-based), the aging state (ON or OFF) and the time-based aging attributes.

Set the aging state to either ON or OFF. ON indicates that the aging state is enabled and aging is done automatically. OFF indicates that the aging state is disabled and aging is not done automatically. In both cases, the aging policy is defined. The default is ON.

Time-based aging attributes are defined at the SQL level and are specified by the LIFETIME and CYCLE clauses.

Specify ColumnName as the name of the column used for time-based aging. Define the column as NOT NULL and of data type TIMESTAMP or DATE. The value of this column is subtracted from SYSDATE, truncated using the specified unit (second, minute, hour, day) and then compared to the LIFETIME value. If the result is greater than the LIFETIME value, then the row is a candidate for aging.

The values of the column that you use for aging are updated by your applications. If the value of this column is unknown for some rows, and you do not want the rows to be aged, define the column with a large default value (the column cannot be NULL).

You can define your aging column with a data type of TT_TIMESTAMP or TT_DATE. If you choose data type TT_DATE, then you must specify the LIFETIME unit as days.

If you specify the AS SelectQuery clause, you must define the ColumnName on the table you are creating.

For more information about time-based aging, see "Implementing an aging policy in your tables" in Oracle TimesTen In-Memory Database Operations Guide.

LIFETIME Num1 {SECOND[S]| MINUTE[S]|HOUR[S]| DAY[S]}

LIFETIME is a time-based aging attribute and is a required clause.

Specify the LIFETIME clause after the AGING USE ColumnName clause.

The LIFETIME clause specifies the minimum amount of time data is kept in cache.

Specify Num1 as a positive integer constant to indicate the unit of time expressed in seconds, minutes, hours or days that rows should be kept in cache. Rows that exceed the LIFETIME value are aged out (deleted from the table). If you define your aging column with data type TT_DATE, then you must specify DAYS as the LIFETIME unit.

The concept of time resolution is supported. If DAYS is specified as the time resolution, then all rows whose timestamp belongs to the same day are aged out at the same time. If HOURS is specified as the time resolution, then all rows with timestamp values within that hour are aged at the same time. A LIFETIME of 3 days is different than a LIFETIME of 72 hours (3*24) or a LIFETIME of 432 minutes (3*24*60).

[CYCLE Num2 {SECOND[S] |MINUTE[S]|HOUR[S]| DAY[S]}]

CYCLE is a time-based aging attribute and is optional. Specify the CYCLE clause after the LIFETIME clause.

The CYCLE clause indicates how often the system should examine rows to see if data exceeds the specified LIFETIME value and should be aged out (deleted).

Specify Num2 as a positive integer constant.

If you do not specify the CYCLE clause, then the default value is 5 minutes. If you specify 0 for Num2, then the aging thread wakes up every second.

If the aging state is OFF, then aging is not done automatically and the CYCLE clause is ignored.

AS SelectQuery

If specified, creates a new table from the contents of the result set of the SelectQuery. The rows returned by SelectQuery are inserted into the table.

Data types and data type lengths are derived from SelectQuery.

SelectQuery is a valid SELECT statement that may or may not contain a subquery. See "SELECT" for information on the SELECT statement.

You can specify a statement level optimizer hint after the SELECT verb. For more information on statement level optimizer hints, see "Statement level optimizer hints".

Column definition: TimesTen Classic

SQL syntax

You can only use the keyword, ENABLE, when defining columns in the CREATE TABLE statement.

For all data types other than LOBs, the syntax is as follows:

ColumnName ColumnDataType
 [DEFAULT DefaultVal]
 [[NOT] INLINE]
 [PRIMARY KEY | UNIQUE | 
 NULL [UNIQUE] | 
 NOT NULL [ENABLE] [PRIMARY KEY | UNIQUE] 
]

For LOB data types, you cannot create a primary key or unique constraint on LOB columns. In addition, LOB data types are stored out of line, so the INLINE attribute cannot be specified.

LOB data types are not supported with TimesTen Scaleout.

For all LOB data types, the syntax is:

ColumnName ColumnDataType
 [DEFAULT DefaultVal] [[NOT] NULL [ENABLE]] |
 [[NOT] NULL [ENABLE]] [DEFAULT DefaultVal]

Parameters

The column definition has the following parameters:

Parameter Description

ColumnName

Name to be assigned to one of the columns in the new table. No two columns in the table can be given the same name. A table can have a maximum of 1000 columns.

If you specify the AS SelectQuery clause, ColumnName is optional. The number of column names must match the number of columns in SelectQuery.

ColumnDataType

Type of data the column can contain. Some data types require that you indicate a length. See Data Types for the data types that can be specified.

If you specify the AS SelectQuery clause, do not specify ColumnDataType.

DEFAULT DefaultVal

Indicates that if a value is not specified for the column in an INSERT statement, the default value DefaultVal is inserted into the column. The default value specified must have a type that is compatible with the data type of the column. A default value can be as long as the data type of the associated column allows. You cannot assign a default value for the ROWID data type or for columns in read-only cache groups. In addition, you cannot use a function within the DEFAULT clause.

The following are the supported data types for DefaultVal:

  • NULL

  • Constant expression (an expression that is evaluated to a constant value)

  • SYSDATE and GETDATE ("SYSDATE and GETDATE")

  • SYSTEM_USER ("SYSTEM_USER")

If the default value is one of the users, the data type of the column must be either CHAR or VARCHAR2 and the width of the column must be at least 30 characters.

If you specify the AS SelectQuery clause, optionally, you can specify the DEFAULT clause on the table you are creating.

INLINE|NOT INLINE

By default, variable-length columns whose declared column length is greater than 128 bytes are stored out of line. Variable-length columns whose declared column length is less than or equal to 128 bytes are stored inline. The default behavior can be overridden during table creation through the use of the INLINE and NOT INLINE keywords.

If you specify the AS SelectQuery clause, optionally, you can specify the INLINE | NOT INLINE clause on the table you are creating.

NULL

Indicates that the column can contain NULL values.

If you specify the AS SelectQuery clause, optionally, you can specify NULL on the table you are creating.

If you specify NULL, you cannot specify ENABLE.

NOT NULL [ENABLE]

Indicates that the column cannot contain NULL values. If NOT NULL is specified, any statement that attempts to place a NULL value in the column is rejected.

If you specify the AS SelectQuery clause, optionally, you can specify NOT NULL [ENABLE] on the table you are creating.

If you specify NOT NULL, you can optionally specify ENABLE. Because NOT NULL constraints are always enabled, you are not required to specify ENABLE.

You can only use the keyword, ENABLE, when defining columns in the CREATE TABLE statement.

UNIQUE

A unique constraint placed on the column. No two rows in the table may have the same value for this column. TimesTen creates a unique range index to enforce uniqueness. So a column with a unique constraint can use more memory and time during execution than a column without the constraint. Cannot be used with PRIMARY KEY.

If you specify the AS SelectQuery clause, optionally, you can specify UNIQUE on the table you are creating.

PRIMARY KEY

A unique NOT NULL constraint placed on the column. No two rows in the table may have the same value for this column. Cannot be used with UNIQUE.

If you specify the AS SelectQuery clause, optionally, you can specify PRIMARY KEY on the table you are creating.

Description for CREATE TABLE: TimesTen Classic

  • If you are planning to load your tables with data, consider creating your tables without indexes. After the data is loaded, you can then create your indexes. This reduces the time it take to load the data into the tables.
  • All columns participating in the primary key are NOT NULL.

  • A PRIMARY KEY that is specified in the ColumnDefinition can only be specified for one column.

  • You cannot specify a PRIMARY KEY in both the ColumnDefinition clause and the PRIMARY KEY clause.

  • For both primary key and foreign key constraints, duplicate column names are not allowed in the constraint column list.

  • You cannot update primary key column(s) unless you update the column(s) to the same value.

  • There are performance considerations when you define out of line columns instead of inline columns:

    • Accessing data is slower because TimesTen does not store data contiguously with out of line columns.

    • Populating data is slower because TimesTen generates more logging operations.

    • Deleting data is slower because TimesTen performs more reclaim and logging operations.

    • Storing a column requires less overhead.

  • If ON DELETE CASCADE is specified on a foreign key constraint for a child table, a user can delete rows from a parent table for which the user has the DELETE privilege without requiring explicit DELETE privilege on the child table.

  • To change the ON DELETE CASCADE triggered action, drop then redefine the foreign key constraint.

  • You cannot create a table that has a foreign key referencing a cached table.

  • UNIQUE column constraint and default column values are not supported with materialized views.

  • Use the ALTER TABLE statement to change the representation of the primary key index for a table.

  • If you specify the AS SelectQuery clause:

    • Data types and data type lengths are derived from the SelectQuery. Do not specify data types on the columns of the table you are creating.

    • TimesTen defines on columns in the new table NOT NULL constraints that were explicitly created on the corresponding columns of the selected table if SelectQuery selects the column rather than an expression containing the column.

    • NOT NULL constraints that were implicitly created by TimesTen on columns of the selected table (for example, primary keys) are carried over to the new table. You can override the NOT NULL constraint on the selected table by defining the new column as NULL. For example:

      CREATE TABLE newtable (newcol NULL) AS SELECT (col) FROM tab;
      
    • NOT INLINE/INLINE attributes are carried over to the new table.

    • Unique keys, foreign keys, indexes and column default values are not carried over to the new table.

    • If all expressions in SelectQuery are columns, rather than expressions, then you can omit the columns from the table you are creating. In this case, the name of the columns are the same as the columns in SelectQuery. If the SelectQuery contains an expression rather than a simple column reference, either specify a column alias or name the column in the CREATE TABLE statement.

    • Do not specify foreign keys on the table you are creating.

    • Do not specify the SELECT FOR UPDATE clause in SelectQuery.

    • The ORDER BY clause is not supported when you use the AS SelectQuery clause.

    • SelectQuery cannot contain set operators UNION, MINUS, INTERSECT.

    • In a replicated environment, be aware of the following.

      To include a new table, including global temporary tables, into an active standby pair when the table is created, set DDL_REPLICATION_LEVEL to 2 or greater and DDL_REPLICATION_ACTION to INCLUDE before executing the CREATE TABLE statement on the active database. In this configuration, the table is included in the active standby pair and is replicated to all databases in the replication scheme.

      If DDL_REPLICATION_ACTION is set to EXCLUDE, then the new table is not included in the active standby pair but is replicated to all databases in the replication scheme. Any DML issued on that table will not be replicated, as the table will not be part of the replication scheme. To enable DML replication for the table, you must execute the ALTER ACTIVE STANDBY PAIR ... INCLUDE TABLE statement to include the table. In this case, the table must be empty and present on all databases before executing ALTER ACTIVE STANDBY PAIR ... INCLUDE TABLE, as the table contents will be truncated when this statement is executed.

      See "ALTER SESSION" for more information.

  • By default, a range index is created to enforce the primary key. Use the UNIQUE HASH clause to specify a hash index for the primary key.

    • If your application performs range queries using a table's primary key, then choose a range index for that table by omitting the UNIQUE HASH clause.

    • If your application performs only exact match lookups on the primary key, then a hash index may offer better response time and throughput. In such a case, specify the UNIQUE HASH clause.

  • A hash index is created with a fixed size that remains constant for the life of the table or until the hash index is resized with the ALTER TABLE statement or when the index is dropped and recreated. A smaller hash index results in more hash collisions. A larger hash index reduces collisions but can waste memory. Hash key comparison is a fast operation, so a small number of hash collisions should not cause a performance problem for TimesTen.

    To ensure that your hash index is sized correctly, your application must indicate the expected size of your table with the value of the RowPages parameter of the SET PAGES clause. Compute this value by dividing the number of expected rows in your table by 256. For example, if your table has 256,000 rows, specify 1000 for the value of RowPages (256000/256=1000).

  • At most 16 columns are allowed in a hash key.

  • ON DELETE CASCADE is supported on detail tables of a materialized view. If you have a materialized view defined over a child table, a deletion from the parent table causes cascaded deletes in the child table. This, in turn, triggers changes in the materialized view.

  • The total number of rows reported by the DELETE statement does not include rows deleted from child tables as a result of the ON DELETE CASCADE action.

  • For ON DELETE CASCADE: Since different paths may lead from a parent table to a child table, the following rule is enforced:

    • Either all paths from a parent table to a child table are "delete" paths or all paths from a parent table to a child table are "do not delete" paths. Specify ON DELETE CASCADE on all child tables on the "delete" path.

    • This rule does not apply to paths from one parent to different children or from different parents to the same child.

  • For ON DELETE CASCADE, the following rule is also enforced.

    • If a table is reached by a "delete" path, then all its children are also reached by a "delete" path.

  • For ON DELETE CASCADE with replication, the following restrictions apply:

    • The foreign keys specified with ON DELETE CASCADE must match between the Master and subscriber for replicated tables. Checking is done at runtime. If there is an error, the receiver thread stops working.

    • All tables in the delete cascade tree have to be replicated if any table in the tree is replicated. This restriction is checked when the replication scheme is created or when a foreign key with ON DELETE CASCADE is added to one of the replication tables. If an error is found, the operation is aborted. You may be required to drop the replication scheme first before trying to change the foreign key constraint.

    • You must stop the replication agent before adding or dropping a foreign key on a replicated table.

  • The data in a global temporary table is private to the current connection and does not need to be secured between users. Thus, global temporary tables do not require object privileges.

  • After you have defined an aging policy for the table, you cannot change the policy from LRU to time-based or from time-based to LRU. You must first drop aging and then alter the table to add a new aging policy.

  • The aging policy must be defined to change the aging state.

  • For the time-based aging policy, you cannot add or modify the aging column. This is because you cannot add or modify a NOT NULL column.

  • LRU and time-based aging can be combined in one system. If you use only LRU aging, the aging thread wakes up based on the cycle specified for the whole database. If you use only time-based aging, the aging thread wakes up based on an optimal frequency. This frequency is determined by the values specified in the CYCLE clause for all tables. If you use both LRU and time-based aging, then the thread wakes up based on a combined consideration of both types.

  • The following rules determine if a row is accessed or referenced for LRU aging:

    • Any rows used to build the result set of a SELECT statement.

    • Any rows used to build the result set of an INSERT ... SELECT statement.

    • Any rows that are about to be updated or deleted.

  • Compiled commands are marked invalid and need recompilation when you either drop LRU aging from or add LRU aging to tables that are referenced in the commands.

  • Call the ttAgingScheduleNow procedure to schedule the aging process immediately regardless of the aging state.

  • Aging restrictions:

    • LRU aging and time-based aging are not supported on detail tables of materialized views.

    • LRU aging and time-based aging are not supported on global temporary tables.

    • You cannot drop the column that is used for time-based aging.

    • The aging policy and aging state must be the same in all sites of replication.

    • Tables that are related by foreign keys must have the same aging policy.

    • For LRU aging, if a child row is not a candidate for aging, neither this child row nor its parent row are deleted. ON DELETE CASCADE settings are ignored.

    • For time-based aging, if a parent row is a candidate for aging, then all child rows are deleted. ON DELETE CASCADE (whether specified or not) is ignored.

Column-based compression of tables (TimesTen Classic)

You can compress tables at the column level, which stores data more efficiently. This eliminates redundant storage of duplicate values within columns and improves the performance of SQL queries that perform full table scans.

You can define one or more columns in a table to be compressed together, which is called a compressed column group. You can define one or more compressed column groups in each table.

A dictionary table is created for each compressed column group that contains a column with all the distinct values of the compressed column group. The compressed column group now contains a pointer to the row in the dictionary table for the appropriate value. The width of this pointer can be 1, 2, or 4 bytes long depending on the maximum number of entries you defined for the dictionary table. So if the sum of the widths of the columns in a compressed column group is wider than the 1, 2, or 4 byte pointer width, and if there are a lot of duplicate values of those column values, you have reduced the amount of space used by the table.

Figure 6-1 shows the compressed column group in the table pointing to the appropriate row in the dictionary table.

Figure 6-1 Column-Based Compression

Description of Figure 6-1 follows
Description of "Figure 6-1 Column-Based Compression"

The dictionary table has a column of pointers to each of the distinct values. When the user configures the maximum number of distinct entries for the compressed column group, the size of the compressed column group is set as follows:

  • 1 byte for a maximum number of entries of 255 (28-1). When the maximum number is between 1 and 255, the dictionary size is set to 255 (28-1) values and the compressed column group pointer column is 1 byte.

  • 2 bytes for a maximum number of entries of 65,535 (216-1). When the maximum number is between 256 and 65,535, the dictionary size is set to 65,535 (216-1) values and the compressed column group pointer column is 2 bytes.

  • 4 bytes for a maximum number of entries of 4,294,967,295 (232-1). When the maximum number is between 65,536 and 4,294,967,295, the dictionary size is set to 4,294,967,295 (232-1) values and the compressed column group pointer column is 4 bytes. This is the default.

Syntax: column-based compression (TimesTen Classic)

The syntax for ColumnBasedCompression is:

[COMPRESS (CompressColumns [,...])]

The CompressColumns syntax is as follows:

{ColumnDefinition | (ColumnDefinition [,...])} BY DICTIONARY 
   [MAXVALUES = CompressMax]

Parameters

ColumnBasedCompression syntax has the following parameters:

Parameter Description

COMPRESS (CompressColumns [,...])

Defines a compressed column group for a table that is enabled for compression. This can include one or more columns in the table. However, a column can be included in only one compressed column group.

Only INLINE columns are supported when you specify multiple columns in a compressed column group. An out-of-line column cannot be in a multi-column compression group.

Each compressed column group is limited to a maximum of 16 columns.

BY DICTIONARY

Defines a compression dictionary for each compressed column group.

MAXVALUES = CompressMax

CompressMax is the total number of distinct values in the table and sets the size for the compressed column group pointer column to 1, 2, or 4 bytes and sets the size for the maximum number of entries in the dictionary table.

For the dictionary table, NULL is counted as one unique value.

CompressMax takes an integer between 1 and 429497295 (232-1).

The maximum size defaults to size of 232-1 if the MAXVALUES clause is omitted, which uses 4 bytes for the pointer column. An error is thrown if the value is greater than 232-1.

Description: column-based compression (TimesTen Classic)

  • Compressed column groups can be added at the time of table creation or added later using ALTER TABLE. You can drop a compressed column group with the ALTER TABLE statement, but you must drop the entire group.

  • You can create indexes on any columns in the table and on columns that exist in separate compression column groups. However, you cannot create single column compression groups on unique columns or on single column primary keys. You also cannot create unique indexes or primary keys where all the indexes or primary keys are in the same compression group.

  • LOB columns cannot be compressed.

  • Compression is not supported on columns in replicated tables, cache group tables, or on global temporary tables. You cannot create a table with the CREATE TABLE AS SELECT statement when defining column-based compression for that table in that statement.

  • You cannot create materialized views on tables enabled for compression.

  • Column-based compression is not supported with TimesTen Scaleout.

Examples: TimesTen Classic

A range index is created on partnumber because it is the primary key.

Command> CREATE TABLE price
          (partnumber INTEGER NOT NULL PRIMARY KEY,
           vendornumber INTEGER NOT NULL,
           vendpartnum CHAR(20) NOT NULL,
           unitprice DECIMAL(10,2),
           deliverydays SMALLINT,
           discountqty SMALLINT);
Command> INDEXES price;
Indexes on table SAMPLEUSER.PRICE:
 PRICE: unique range index on columns:
    PARTNUMBER
  1 index found.
1 index found on 1 table.

A hash index is created on column clubname, the primary key.

CREATE TABLE recreation.clubs
(clubname CHAR(15) NOT NULL PRIMARY KEY,
 clubphone SMALLINT,
 activity CHAR(18))
UNIQUE HASH ON (clubname) PAGES = 30;

A range index is created on the two columns membername and club because together they form the primary key.

Command> CREATE TABLE recreation.members
          (membername CHAR(20) NOT NULL,
           club CHAR(15) NOT NULL,
           memberphone SMALLINT,
           PRIMARY KEY (membername, club));
Command> INDEXES recreation.members;
Indexes on table RECREATION.MEMBERS:
  MEMBERS: unique range index on columns:
    MEMBERNAME
    CLUB
  1 index found on 1 table.

No hash index is created on the table recreation.events.

CREATE TABLE recreation.events
(sponsorclub CHAR(15),
 event CHAR(30),
 coordinator CHAR(20),
 results VARBINARY(10000));

A hash index is created on the column vendornumber.

CREATE TABLE purchasing.vendors
(vendornumber INTEGER NOT NULL PRIMARY KEY,
 vendorname CHAR(30) NOT NULL,
 contactname CHAR(30),
 phonenumber CHAR(15),
 vendorstreet CHAR(30) NOT NULL,
 vendorcity CHAR(20) NOT NULL,
 vendorstate CHAR(2) NOT NULL,
 vendorzipcode CHAR(10) NOT NULL,
 vendorremarks VARCHAR(60))
UNIQUE HASH ON (vendornumber) PAGES = 101;

A hash index is created on the columns membername and club because together they form the primary key.

CREATE TABLE recreation.members
    (membername CHAR(20) NOT NULL,
        club CHAR(15) NOT NULL,
        memberphone SMALLINT,
        PRIMARY KEY (membername, club))
    UNIQUE HASH ON (membername, club) PAGES = 100;

A hash index is created on the columns firstname and lastname because together they form the primary key in the table authors. A foreign key is created on the columns authorfirstname and authorlastname in the table books that references the primary key in the table authors.

CREATE TABLE authors
    (firstname VARCHAR(255) NOT NULL,
        lastname VARCHAR(255) NOT NULL,
        description VARCHAR(2000),
        PRIMARY KEY (firstname, lastname))
    UNIQUE HASH ON (firstname, lastname) PAGES=20;
CREATE TABLE books
    (title VARCHAR(100),
    authorfirstname VARCHAR(255),
    authorlastname VARCHAR(255),
    price DECIMAL(5,2),
    FOREIGN KEY (authorfirstname, authorlastname) 
    REFERENCES authors(firstname, lastname));

The following statement overrides the default character of VARCHAR columns and creates a table where one VARCHAR (10) column is NOT INLINE and one VARCHAR (144) is INLINE.

CREATE TABLE t1
    (c1 VARCHAR(10) NOT INLINE NOT NULL,
    c2 VARCHAR(144) INLINE NOT NULL);

The following statement creates a table with a UNIQUE column for book titles.

CREATE TABLE books
    (title VARCHAR(100) UNIQUE,
        authorfirstname VARCHAR(255),
        authorlastname VARCHAR(255),
        price DECIMAL(5,2),
        FOREIGN KEY (authorfirstname, authorlastname)
        REFERENCES authors(firstname, lastname));

The following statement creates a table with a default value of 1 on column x1 and a default value of SYSDATE on column d.

CREATE TABLE t1
    (x1 INT DEFAULT 1, d TIMESTAMP DEFAULT SYSDATE);

This example creates the rangex table and defines col1 as the primary key. A range index is created by default.

Command> CREATE TABLE rangex (col1 TT_INTEGER PRIMARY KEY);
Command> INDEXES rangex;
Indexes on table SAMPLEUSER.RANGEX:
  RANGEX: unique range index on columns:
    COL1
  1 index found
1 index found on 1 table.

The following statement illustrates the use of the ON DELETE CASCADE clause for parent/child tables of the HR schema. Tables with foreign keys have been altered to enable ON DELETE CASCADE.

ALTER TABLE countries
ADD CONSTRAINT countr_reg_fk
         FOREIGN KEY (region_id)
           REFERENCES regions(region_id) ON DELETE CASCADE;
ALTER TABLE locations
     ADD CONSTRAINT loc_c_id_fk
          FOREIGN KEY (country_id)
          	  REFERENCES countries(country_id) ON DELETE CASCADE;
ALTER TABLE departments
     ADD CONSTRAINT dept_loc_fk
         FOREIGN KEY (location_id)
           REFERENCES locations (location_id) ON DELETE CASCADE;
ALTER TABLE employees
     ADD CONSTRAINT     emp_dept_fk
         FOREIGN KEY (department_id)
           REFERENCES departments ON DELETE CASCADE;
ALTER TABLE employees
     ADD CONSTRAINT     emp_job_fk
         FOREIGN KEY (job_id)
           REFERENCES jobs (job_id);
ALTER TABLE job_history
     ADD CONSTRAINT     jhist_job_fk
         FOREIGN KEY (job_id)
           REFERENCES jobs;
ALTER TABLE job_history
     ADD CONSTRAINT     jhist_emp_fk
         FOREIGN KEY (employee_id)
           REFERENCES employees ON DELETE CASCADE;
ALTER TABLE job_history
     ADD CONSTRAINT     jhist_dept_fk
         FOREIGN KEY (department_id)
           REFERENCES departments ON DELETE CASCADE;
     ;

This example shows how time resolution works with aging.

If lifetime is three days (resolution is in days):

  • If (SYSDATE - ColumnValue) <= 3, do not age.

  • If (SYSDATE - ColumnValue) > 3, then the row is a candidate for aging.

  • If (SYSDATE - ColumnValue) = 3 days, 22 hours, then the row is not aged out if you specified a lifetime of three days. The row would be aged out if you had specified a lifetime of 72 hours.

This example creates a table with LRU aging. Aging state is ON by default.

CREATE TABLE agingdemo
       (agingid NUMBER NOT NULL PRIMARY KEY,
        name  VARCHAR2 (20)
       )
       AGING LRU;
Command> DESCRIBE agingdemo;
Table USER.AGINGDEMO:
  Columns:
     *AGINGID NUMBER NOT NULL
     NAME VARCHAR2 (20) INLINE
     AGING LRU ON
1 table found.
(primary key columns are indicated with *)

This example creates a table with time-based aging. Lifetime is three days. Cycle is not specified, so the default is five minutes. Aging state is OFF.

CREATE TABLE agingdemo2
       (agingid NUMBER NOT NULL PRIMARY KEY,
        name  VARCHAR2 (20),
        agingcolumn TIMESTAMP NOT NULL
        )
        AGING USE agingcolumn LIFETIME 3 DAYS OFF;
Command> DESCRIBE agingdemo2;
Table USER.AGINGDEMO2:
  Columns:
     *AGINGID NUMBER NOT NULL
     NAME VARCHAR2 (20) INLINE
     AGINGCOLUMN TIMESTAMP (6) NOT NULL
  Aging use AGINGCOLUMN lifetime 3 days cycle 5 minutes off
1 table found.
(primary key columns are indicated with *)

This example generates an error message. It illustrates that after you create an aging policy, you cannot change it. You must drop aging and redefine aging.

CREATE TABLE agingdemo2
        (agingid NUMBER NOT NULL PRIMARY KEY,
        name  VARCHAR2 (20),
        agingcolumn TIMESTAMP NOT NULL
        )
        AGING USE agingcolumn LIFETIME 3 DAYS OFF;
ALTER TABLE agingdemo2
        ADD AGING LRU;
 2980: Cannot add aging policy to a table with an existing aging policy. Have to
drop the old aging first
The command failed.
DROP aging on the table and redefine with LRU aging.
ALTER TABLE agingdemo2
        DROP AGING;
ALTER TABLE agingdemo2
        ADD AGING LRU;
Command> DESCRIBE agingdemo2;
Table USER.AGINGDEMO2:
  Columns:
   *AGINGID                         NUMBER NOT NULL
    NAME                            VARCHAR2 (20) INLINE
    AGINGCOLUMN                     TIMESTAMP (6) NOT NULL
  Aging lru on
1 table found.
(primary key columns are indicated with *)

Attempt to create a table with time-based aging. Define aging column with data type TT_DATE and LIFETIME 3 hours. An error is generated because the LIFETIME unit must be expressed as DAYS.

Command> CREATE TABLE aging1 (col1 TT_INTEGER PRIMARY KEY, 
         col2 TT_DATE NOT NULL) AGING USE col2 LIFETIME 3 HOURS;
 2977: Only DAY lifetime unit is allowed with a TT_DATE column
The command failed.

Use AS SelectQuery clause to create the table emp. Select last_name from the employees table where employee_id between 100 and 105. You see six rows inserted into emp. First issue the SELECT statement to see rows that should be returned.

Command> SELECT last_name FROM employees 
         WHERE employee_id BETWEEN 100 AND 105;
< King >
< Kochhar >
< De Haan >
< Hunold >
< Ernst >
< Austin >
6 rows found.
Command> CREATE TABLE emp AS SELECT last_name FROM employees
         WHERE employee_id BETWEEN 100 AND 105;
6 rows inserted.
Command> SELECT * FROM emp;
< King >
< Kochhar >
< De Haan >
< Hunold >
< Ernst >
< Austin >
6 rows found.

Use AS SelectQuery to create table totalsal. Sum salary and insert result into totalsalary. Define alias s for SelectQuery expression.

Command> CREATE TABLE totalsal AS SELECT SUM (salary) s FROM employees;
1 row inserted.
Command> SELECT * FROM totalsal;
< 691400 >
1 row found.

Use AS SelectQuery to create table defined with column commission_pct. Set default to .3. First describe table employees to show that column commission_pct is of type NUMBER (2,2). For table c_pct, column commission_pct inherits type NUMBER (2,2) from column commission_pct of employees table.

Command> DESCRIBE employees;
Table SAMPLEUSER.EMPLOYEES:
  Columns:
   *EMPLOYEE_ID                     NUMBER (6) NOT NULL
    FIRST_NAME                      VARCHAR2 (20) INLINE
    LAST_NAME                       VARCHAR2 (25) INLINE NOT NULL
    EMAIL                           VARCHAR2 (25) INLINE UNIQUE NOT NULL
    PHONE_NUMBER                    VARCHAR2 (20) INLINE
    HIRE_DATE                       DATE NOT NULL
    JOB_ID                          VARCHAR2 (10) INLINE NOT NULL
    SALARY                          NUMBER (8,2)
    COMMISSION_PCT                  NUMBER (2,2)
    MANAGER_ID                      NUMBER (6)
    DEPARTMENT_ID                   NUMBER (4)

1 table found.
(primary key columns are indicated with *)
Command> CREATE TABLE c_pct (commission_pct DEFAULT .3) AS SELECT
         commission_pct FROM employees;
107 rows inserted.
Command> DESCRIBE c_pct;

Table SAMPLEUSER.C_PCT:
  Columns:
    COMMISSION_PCT                 NUMBER (2,2) DEFAULT .3

1 table found.
(primary key columns are indicated with *)

The following example creates the employees table where the job_id is compressed.

Command> CREATE TABLE EMPLOYEES
          (EMPLOYEE_ID NUMBER (6) PRIMARY KEY, 
           FIRST_NAME VARCHAR2(20), 
           LAST_NAME VARCHAR2(25) NOT NULL, 
           EMAIL VARCHAR2(25) NOT NULL, 
           PHONE_NUMBER VARCHAR2(20),
           HIRE_DATE DATE NOT NULL, 
           JOB_ID VARCHAR2(10) NOT NULL, 
           SALARY NUMBER (8,2), 
           COMMISSION_PCT NUMBER (2,2), 
           MANAGER_ID NUMBER(6), 
           DEPARTMENT_ID NUMBER(4)) 
          COMPRESS (JOB_ID BY DICTIONARY);

Command> DESCRIBE EMPLOYEES;
 
Table MYSCHEMA.EMPLOYEES:
  Columns:
   *EMPLOYEE_ID                     NUMBER (6) NOT NULL
    FIRST_NAME                      VARCHAR2 (20) INLINE
    LAST_NAME                       VARCHAR2 (25) INLINE NOT NULL
    EMAIL                           VARCHAR2 (25) INLINE NOT NULL
    PHONE_NUMBER                    VARCHAR2 (20) INLINE
    HIRE_DATE                       DATE NOT NULL
    JOB_ID                          VARCHAR2 (10) INLINE NOT NULL
    SALARY                          NUMBER (8,2)
    COMMISSION_PCT                  NUMBER (2,2)
    MANAGER_ID                      NUMBER (6)
    DEPARTMENT_ID                   NUMBER (4)
  COMPRESS ( JOB_ID BY DICTIONARY )
 
1 table found.
(primary key columns are indicated with *)

The following example shows that there are three dictionary table sizes. The value you specify for the maximum number of entries is rounded up to the next size. For example, specifying 400 as the maximum number of job IDs creates a dictionary table that can have at most 65535 entries. The default size of 232-1 is not shown in the DESCRIBE output.

Command> CREATE TABLE employees 
          (employee_id NUMBER(6) PRIMARY KEY, 
           first_name VARCHAR2(20), 
           last_name VARCHAR2(25), 
           email VARCHAR2(25) NOT NULL, 
           job_id VARCHAR2(10) NOT NULL, 
           manager_id NUMBER(6), 
           department_id NUMBER(4)) 
          COMPRESS (last_name BY DICTIONARY MAXVALUES=70000, 
             job_id BY DICTIONARY MAXVALUES=400, 
             department_id BY DICTIONARY MAXVALUES=100);

Command> DESCRIBE employees;
Table MYSCHEMA.EMPLOYEES:
  Columns:
   *EMPLOYEE_ID                     NUMBER (6) NOT NULL
    FIRST_NAME                      VARCHAR2 (20) INLINE
    LAST_NAME                       VARCHAR2 (25) INLINE
    EMAILS                          VARCHAR2 (25) INLINE NOT NULL
    JOB_ID                          VARCHAR2 (10) INLINE NOT NULL
    MANAGER_ID                      NUMBER (6)
    DEPARTMENT_ID                   NUMBER (4)
  COMPRESS ( LAST_NAME BY DICTIONARY,
             JOB_ID BY DICTIONARY MAXVALUES=65535,
             DEPARTMENT_ID BY DICTIONARY MAXVALUES=255 )
 
1 table found.
(primary key columns are indicated with *)

CREATE USER

The CREATE USER statement creates a user in the TimesTen database.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

CREATE USER user IDENTIFIED BY {password | "password"}
  [PROFILE profile] [ACCOUNT {LOCK|UNLOCK}] [PASSWORD EXPIRE]

or

CREATE USER user IDENTIFIED EXTERNALLY
  [PROFILE profile] [ACCOUNT {LOCK|UNLOCK}]

Parameters

Parameter Description

user

Name of the user.

IDENTIFIED BY {password | "password"}

Identification clause for an internal user. You must supply a password for an internal user.

The password you can specify is dependent on the profile assigned to the user. Specifically, the value of the PASSWORD_COMPLEXITY_CHECKER password parameter determines the complexity of the password. If the value is TT_VERIFY_FUNCTION, TT_STRONG_VERIFY_FUNCTION, or TT_STIG_VERIFY_FUNCTION, the password must meet specific password verification requirements. For example, if the value is TT_VERIFY_FUNCTION, the password cannot contain the name of the database. See "About password complexity checker verification" for details.

IDENTIFIED EXTERNALLY

Identifies an external user (the operating system user). To perform database operations as an external user, the external user name must match the user name authenticated by the operating system or network. A password is not required by TimesTen as the user has been authenticated by the operating system at login time.

PROFILE profile

Use the PROFILE clause to specify the name of the profile (designated by profile) that you want to assign to the user. The profile sets the limits for the password parameters for the user. See "CREATE PROFILE" for information on these password parameters. If you omit the PROFILE clause, TimesTen assigns the DEFAULT profile to the user. If you create an external user (denoted by specifying the EXTERNALLY keyword), you can specify a PROFILE clause, but the password parameters have no effect on external users. Additionally, if you do not specify the PROFILE clause for an external user, TimesTen assigns the DEFAULT profile to the user (but the password parameters have no effect).

ACCOUNT [LOCK|UNLOCK]

Specify ACCOUNT LOCK to lock the user's account and disable connections to the database. Specify ACCOUNT UNLOCK to unlock the user's account and enable connections to the database. The default is ACCOUNT UNLOCK.

PASSWORD EXPIRE

Specify PASSWORD EXPIRE if you want the user's password to expire. This setting forces a user with ADMIN privileges to change the password before the user can connect to the database. In order to change the expired password, a user with ADMIN privileges must use the ALTER USER statement with the IDENTIFIED BY clause to change the password. Once the password is changed, the user can log in to the database with the new password. Note that even if the newly created user is granted ADMIN privileges, that newly created user cannot login to the database and therefore cannot initially change the password. See "ALTER USER" for information. This clause is not valid for an externally identified user (as denoted by the IDENTIFIED EXTERNALLY clause).

Description

  • Database users can be internal or external.

    • Internal users are defined for a TimesTen database.

    • External users are defined by the operating system. External users cannot be assigned a TimesTen password.

  • Password requirements:
    • Cannot exceed 30 characters.
    • Is case-sensitive.
    • Must start with a letter. A password cannot start with a digit or a special character unless the password is enclosed in double quotation marks.
    • If a special character is used, the password must be contained in double quotation marks. The exceptions are the # and the @ special characters. A password that contains the # or the @ special character does not need to be enclosed in double quotation marks.
    • Cannot contain a semi-colon (;) or a double quotation mark (").
  • When a user is created, the user has the privileges granted to PUBLIC and no additional privileges.

  • Use the PROFILE clause to assign a profile to a user. If you assign the profile to an internal user, the user cannot exceed the limits specified for the profile. If you do not assign a profile to an internal user, a DEFAULT profile is assigned to that user. See "CREATE PROFILE" for details.

  • Use the ACCOUNT LOCK or ACCOUNT UNLOCK to lock or unlock the user account.

  • Use the PASSWORD EXPIRE clause to expire the user's password and force a password change before the user can connect to the database.

  • You can create a user over a client/sever connection if the connection is encrypted with TLS. See "Transport Layer Security for TimesTen Client/Server" in the Oracle TimesTen In-Memory Database Security Guide for details.

  • In TimesTen, user brad is the same as user "brad". In both cases, the name of the user is created as BRAD.

  • User names are TT_CHAR data type.

  • This statement is replicated.

Examples

Create users and observe password verification

This example creates the user_pw1 user and does not assign a profile to the user1_pw user. The user is subject to the limits of the DEFAULT profile. The PASSWORD_COMPLEXITY_CHECKER password parameter is set to NULL for the DEFAULT profile. Therefore, there is no password verification performed on this user's password. The example then alters the DEFAULT profile, changing the value of the PASSWORD_COMPLEXITY_CHECKER to TT_VERIFY_FUNCTION. The user1_p1 user can still connect to the database with the original password. Password verification is performed only on newly created users.

Command> CREATE USER user_pw1 IDENTIFIED BY user1_pw1;

User created.
Query the dba_profiles system view to check the limits of the password parameters for the DEFAULT profile. The PASSWORD_COMPLEXITY_CHECKER password parameter has a value of NULL.
Command> SELECT * FROM dba_profiles WHERE profile = 'DEFAULT';
< DEFAULT, FAILED_LOGIN_ATTEMPTS, PASSWORD, 10 >
< DEFAULT, PASSWORD_LIFE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_REUSE_MAX, PASSWORD, UNLIMITED >
< DEFAULT, PASSWORD_COMPLEXITY_CHECKER, PASSWORD, NULL >
< DEFAULT, PASSWORD_LOCK_TIME, PASSWORD, .0034 >
< DEFAULT, PASSWORD_GRACE_TIME, PASSWORD, UNLIMITED >
< DEFAULT, TEMP_SPACE_PER_SESSION_MAX, MEMORY, UNLIMITED >
8 rows found.
Alter the DEFAULT profile, changing the value of the PASSWORD_COMPLEXITY_CHECKER parameter to TT_VERIFY_FUNCTION. Attempt to connect to the database as the user_pw1 user. The connection is successful, as password verification is only performed on newly created passwords.
Command> ALTER PROFILE "DEFAULT" LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_VERIFY_FUNCTION;

Profile altered.

Command> connect adding "UID=user_pw1;PWD=user_pw1" as user1;
Connection successful: DSN=access1;UID=user_pw1;
DataStore=/scratch/user1/mydatabase1;DatabaseCharacterSet=AL32UTF8;
ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)
Create the user_pw2 user and specify user_pw2 for the password. The CREATE USER statement fails. Password verification is performed on the password for user_pw2, as the password is a newly created password. Create the user_pw2 user again, specifying a password that meets the requirements of the TT_VERIFY_FUNCTION function. The CREATE USER statement is successful, and the user is created. See "TT_VERIFY_FUNCTION" for more information on the TT_VERIFY_FUNCTION function.
Command> CREATE USER user_pw2 IDENTIFIED BY user_pw2;
15186: Password complexity check for the specified password failed
15188: TT-20002: Password contains the username
The command failed.

Command> CREATE USER user_pw2 IDENTIFIED BY abc75#n4;

User created.

Create user with TT_STRONG_VERIFY_FUNCTION password requirements

This example illustrates the password verification requirements for the TT_STRONG_VERIFY_FUNCTION function. Create the profile_pw3 profile and specify a value of TT_STRONG_VERIFY_FUNCTION for the PASSWORD_COMPLEXITY_CHECKER password parameter. Create the user_pw3 user and assign this user the profile_pw3 profile. Experiment with different passwords to confirm that the password meets the requirements of the TT_STRONG_VERIFY_FUNCTION function. If the password meets the requirements, the CREATE USER statement is successful and the user is created. See "TT_STRONG_VERIFY_FUNCTION" for more information on the TT_STRONG_VERIFY_FUNCTION function.
Command> CREATE PROFILE profile_pw3 LIMIT 
           PASSWORD_COMPLEXITY_CHECKER TT_STRONG_VERIFY_FUNCTION;

Profile created.
Create the user_pw3 user and experiment with various passwords. Recall that special characters must be enclosed in double quotation marks (with the exception of # and @).
Command> CREATE USER user_pw3 IDENTIFIED BY abcABC1# 
           PROFILE profile_pw3;
15186: Password complexity check for the specified password failed
15188: TT-20001: Password length less than 9
The command failed.

Command> CREATE USER user_pw3 IDENTIFIED BY abcABCD1# 
           PROFILE profile_pw3;
15186: Password complexity check for the specified password failed
15188: TT-20001: Password must contain at least 2 digit(s)
The command failed.

Command> CREATE USER user_pw3 IDENTIFIED BY abcABCD11# 
           PROFILE profile_pw3;
15186: Password complexity check for the specified password failed
15188: TT-20001: Password must contain at least 2 special character(s)
The command failed.

Command> CREATE USER user_pw3 IDENTIFIED BY "!abcABCD11#" 
           PROFILE profile_pw3;

User created.

Create a user and assign a profile

This example creates the user1 user and assigns the profile1 profile to the user.

Command> CREATE USER user1 IDENTIFIED BY user1 PROFILE profile1;
 
User created.

Create a user and do not assign a profile

This example creates the user2 user and does not assign a profile. The user2 user is assigned the values of the password parameters in the DEFAULT profile.

Command> CREATE USER user2 identified by user2;
 
User created.

Query the dba_users system view to verify the user2 user is assigned the DEFAULT profile.

Command> SELECT profile FROM dba_users WHERE username='USER2';
< DEFAULT >
1 row found.

Create a user and lock the user account

This example creates the user3 user and locks the user3 account. The user3 account must be unlocked by a user with the ADMIN privilege before the user3 user can connect to the database.

Command> CREATE USER user3 IDENTIFIED BY user3 ACCOUNT LOCK;
 
User created.

Grant the CONNECT privilege to user3;

Command> GRANT CONNECT TO user3;

Attempt to connect to the database as user3. The user3 account is locked so the connection fails.

Command> connect adding "UID=user3;PWD=user3" as user3;
15179: the account is locked
The command failed.

As the instance administrator, reconnect to the database and use the ALTER USER statement to unlock the user3 account.

none: Command> use database1
database1: Command> ALTER USER user3 ACCOUNT UNLOCK;
 
User altered.

Attempt to connect to the database a the user3 user. The connection succeeds.

database1: Command> connect adding "UID=user3;PWD=user3" as user3;
Connection successful: DSN=database1;UID=user3;DataStore=/scratch/database1;
DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Lock the user account and enforce a password change

This example creates the user4 user. The user4 user is assigned the profile1 profile. The user4 account is locked and the password for user4 must be changed before the user4 user can connect to the database.

Command> CREATE USER user4 identified by user4 PROFILE profile1 
         ACCOUNT LOCK PASSWORD EXPIRE;
 
User created.

Attempt to connect to the database as user4. The user4 account is locked and the password must be changed before the user4 user can connect to the database.

Command>  connect adding "UID=user4;PWD=user4" as user4;
15179: the account is locked
The command failed.

As the instance administrator, reconnect to the database and use the ALTER USER statement to unlock the user4 account.

none: Command> use database1
database1: Command> ALTER USER user4 ACCOUNT UNLOCK;
 
User altered.

Grant the CONNECT privilege to user4. Then change the user4's expired password. (This example changes the password to user4_changed, represented in bold.)

database1: Command> GRANT CONNECT TO user4;
database1: Command> ALTER USER user4 IDENTIFIED BY user4_changed;
 
User altered.

Attempt to connect to the database as the user4 user. The connection succeeds. The account is unlock and the password is changed.

database1: Command> connect adding "UID=user4;PWD=user4_changed" as user4;
Connection successful: DSN=database1;UID=user4;DataStore=/scratch/database1;
DatabaseCharacterSet=AL32UTF8;ConnectionCharacterSet=AL32UTF8;PermSize=128;
(Default setting AutoCommit=1)

Create an external user

This example creates the user5 user as an external user.

Command> CREATE USER user5 IDENTIFIED EXTERNALLY;
 
User created.

CREATE VIEW

The CREATE VIEW statement creates a view of the tables specified in the SelectQuery clause. A view is a logical table that is based on one or more detail tables. The view itself contains no data. It is sometimes called a nonmaterialized view to distinguish it from a materialized view, which does contain data that has already been calculated from detail tables.

In a replicated environment for an active standby pair, if DDL_REPLICATION_LEVEL is 3 or greater when you execute CREATE VIEW on the active database, the view is replicated to all databases in the replication scheme. See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Required privilege

The user executing the statement must have the CREATE VIEW privilege (if owner) or CREATE ANY VIEW (if not the owner) for another user's view.

The owner of the view must have the SELECT privilege on the detail tables.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

CREATE VIEW [Owner.]ViewName AS SelectQuery

Parameters

Parameter Description

[Owner.]ViewName

Name of view

SelectQuery

Selects column from the detail tables to be used in the view.

You can also create indexes on the view.

Restrictions on the SELECT query

There are several restrictions on the query that is used to define the view.

  • A SELECT * query in a view definition is expanded when the view is created. Any columns added after a view is created do not affect the view.

  • Do not use the following in a SELECT statement that is used to create a view:

    • FIRST

    • ORDER BY

      If used, this is ignored by CREATE VIEW. The result will not be sorted.

    • Arguments

  • Each expression in the select list must have a unique name. A name of a simple column expression would be that column's name unless a column alias is defined. ROWID is considered an expression and needs an alias.

  • Do not use SELECT FOR UPDATE to create a view.

  • Certain TimesTen query restrictions are not checked when a non-materialized view is created. Views that violate those restrictions may be allowed to be created, but an error is returned when the view is referenced later in an executed statement.

  • When a view is referenced in the FROM clause of a SELECT statement, its name is replaced by its definition as a derived table at parsing time. If it is not possible to merge all clauses of a view to the same clause in the original select query to form a supported query without the derived table, the content of this derived table is materialized. For example, if both the view and the referencing select specify aggregates, the view is materialized before its result can be joined with other tables of the select.

  • Use the DROP VIEW statement to drop a view.

  • A view cannot be altered with an ALTER TABLE statement.

  • Referencing a view can fail because of dropped or altered detail tables.

Examples

Create a nonmaterialized view from the employees table.

Command> CREATE VIEW v1 AS SELECT employee_id, email FROM employees;
Command> SELECT FIRST 5 * FROM v1;
< 100, SKING >
< 101, NKOCHHAR >
< 102, LDEHAAN >
< 103, AHUNOLD >
< 104, BERNST >
5 rows found.

Create a nonmaterialized view tview with column max1 from an aggregate query on the table t1.

CREATE VIEW tview (max1) AS SELECT MAX(x1) FROM t1;

DELETE

The DELETE statement deletes rows from a table.

Required privilege

No privilege is required for the table owner.

DELETE on the table for another user's table.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DELETE [hint] [FIRST NumRows] FROM [Owner.]TableName [CorrelationName] 
[WHERE SearchCondition]
[RETURNING|RETURN Expression[,...]INTO DataItem[,...]]

Parameters

Parameter Description

hint

Specifies a statement level optimizer hint for the DELETE statement. See "Statement level optimizer hints" for information on optimizer hints.

FIRST NumRows

Specifies the number of rows to delete. FIRST NumRows is not supported in subquery statements. NumRows must be either a positive INTEGER or a dynamic parameter placeholder. The syntax for a dynamic parameter placeholder is either ? or :DynamicParameter. The value of the dynamic parameter is supplied when the statement is executed.

[Owner.]TableName [CorrelationName]

Designates a table from which any rows satisfying the search condition are to be deleted.

[Owner.]TableName identifies a table to be deleted.

CorrelationName specifies an alias for the immediately preceding table. Use the correlation name to reference the table elsewhere in the DELETE statement. The scope of the CorrelationName is the SQL statement in which it is used. It must conform to the syntax rules for a basic name. See "Basic names" for details.

SearchCondition

Specifies which rows are to be deleted. If no rows satisfy the search condition, the table is not changed. If the WHERE clause is omitted, all rows are deleted. The search condition can contain a subquery.

Expression

Valid expression syntax. See Expressions for details.

DataItem

Host variable or PL/SQL variable that stores the retrieved Expression value.

Description

  • If all the rows of a table are deleted, the table is empty but continues to exist until you issue a DROP TABLE statement.

  • If your table has out of line columns and there are millions of rows to delete, consider calling the ttCompact built-in procedure to free memory.

  • The DELETE operation fails if it violates any foreign key constraint. See "CREATE TABLE" for a description of the foreign key constraint.

  • The total number of rows reported by the DELETE statement does not include rows deleted from child tables as a result of the ON DELETE CASCADE action.

  • If ON DELETE CASCADE is specified on a foreign key constraint for a child table, a user can delete rows from a parent table for which the user has the DELETE privilege without requiring explicit DELETE privilege on the child table.

  • Restrictions on the RETURNING clause:

    • Each Expression must be a simple expression. Aggregate functions are not supported.

    • You cannot return a sequence number into an OUT parameter.

    • ROWNUM and subqueries cannot be used in the RETURNING clause.

    • Parameters in the RETURNING clause cannot be duplicated anywhere in the DELETE statement.

    • Using the RETURNING clause to return multiple rows requires PL/SQL BULK COLLECT functionality. See "FORALL and BULK COLLECT operations" in Oracle TimesTen In-Memory Database PL/SQL Developer's Guide for information about BULK COLLECT.

    • In PL/SQL, you cannot use a RETURNING clause with a WHERE CURRENT operation.

Examples

Rows for orders whose quantity is less than 50 are deleted.

DELETE FROM purchasing.orderitems
WHERE quantity < 50;

The following query deletes all the duplicate orders assuming that id is not a primary key:

DELETE FROM orders a
WHERE EXISTS (SELECT 1 FROM orders b
WHERE a.id = b.id and a.rowid < b.rowid);

The following sequence of statements causes a foreign key violation.

CREATE TABLE master (name CHAR(30), id CHAR(4) NOT NULL PRIMARY KEY);
CREATE TABLE details
  (masterid CHAR(4),description VARCHAR(200), 
   FOREIGN KEY (masterid) REFERENCES master(id));
INSERT INTO master('Elephant', '0001');
INSERT INTO details('0001', 'A VERY BIG ANIMAL');
DELETE FROM master WHERE id = '0001';

If you attempt to delete a "busy" table, an error results. In this example, t1 is a "busy" table that is a parent table with foreign key constraints based on it.

CREATE TABLE t1 (a INT NOT NULL, b INT NOT NULL,
       PRIMARY KEY (a));
CREATE TABLE t2 (c INT NOT NULL, 
       FOREIGN KEY (c) REFERENCES t1(a));
INSERT INTO t1 VALUES (1,1);
INSERT INTO t2 VALUES (1);
DELETE FROM t1;

An error is returned:

SQL ERROR (3001):  Foreign key violation [TTFOREIGN_0] a row in child table T2 
has a parent in the delete range.

Delete an employee from employees. Declare empid and name as variables with the same data types as employee_id and last_name. Delete the row, returning employee_id and last_name into the variables. Verify that the correct row was deleted.

Command> VARIABLE empid NUMBER(6) NOT NULL;
Command> VARIABLE name VARCHAR2(25) INLINE NOT NULL;
Command> DELETE FROM employees WHERE last_name='Ernst'
           RETURNING employee_id, last_name INTO :empid,:name;
1 row deleted.
Command> PRINT empid name;
EMPID                : 104
NAME                 : Ernst

DROP ACTIVE STANDBY PAIR

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

This statement drops an active standby pair replication scheme.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

DROP ACTIVE STANDBY PAIR

Parameters

DROP ACTIVE STANDBY PAIR has no parameters.

Description

The active standby pair is dropped, but all objects such as tables, cache groups, and materialized views still exist on the database on which the statement was issued.

You cannot execute the DROP ACTIVE STANDBY PAIR statement when Oracle Clusterware is used with TimesTen.

DROP CACHE GROUP

This statement is supported in TimesTen Scaleout.

In TimesTen Classic:

The DROP CACHE GROUP statement drops the table associated with the cache group, and removes the cache group definition from the CACHE_GROUP system table.

Required privilege

No privilege is required for the cache group owner.

If not the cache group owner, DROP ANY CACHE GROUP and

DROP ANY TABLE if at least one table in the cache group is not owned by the current user.

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

DROP CACHE GROUP [Owner.]GroupName

Parameters

Parameter Description

[Owner.]GroupName

Name of the cache group to be deleted.

Description

  • If you attempt to delete a cache group table that is in use, TimesTen returns an error.

  • Asynchronous writethrough cache groups cannot be dropped while the replication agent is running.

  • Automatically installed Oracle Database objects for read-only cache groups and cache groups with the AUTOREFRESH attribute are uninstalled by the cache agent. If the cache agent is not running during the DROP CACHE GROUP operation, the Oracle Database objects are uninstalled on the next startup of the cache agent.

  • You cannot execute the DROP CACHE GROUP statement when performed under the serializable isolation level. An error message is returned when attempted.

  • If you issue a DROP CACHE GROUP statement, and there is an autorefresh operation currently running, then:

    • If LockWait interval is 0, the DROP CACHE GROUP statement fails with a lock timeout error.

    • If LockWait interval is nonzero, then the current autorefresh transaction is preempted (rolled back), and the DROP statement continues. This affects all cache groups with the same autorefresh interval.

Examples

DROP CACHE GROUP westerncustomers;

DROP FUNCTION

The DROP FUNCTION statement removes a standalone stored function from the database. Do not use this statement to remove a function that is part of a package.

Required privilege

No privilege is required for the function owner.

DROP ANY PROCEDURE for another user's function.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP FUNCTION [Owner.]FunctionName

Parameters

Parameter Description

[Owner.]FunctionName

Name of the function to be dropped.

Description

  • When you drop a function, TimesTen invalidates objects that depend on the dropped function. If you subsequently reference one of these objects, TimesTen attempts to recompile the object and returns an error message if you have not recreated the dropped function.

  • Do not use this statement to remove a function that is part of a package. Either drop the package or redefine the package without the function using the CREATE PACKAGE statement with the OR REPLACE clause.

  • To use the DROP FUNCTION statement, you must have PL/SQL enabled in your database. If you do not have PL/SQL enabled in your database, an error is thrown.

Examples

The following statement drops the function myfunc and invalidates all objects that depend on myfunc:

Command> DROP FUNCTION myfunc;
 
Function dropped.

If PL/SQL is not enabled in your database, TimesTen returns an error:

Command> DROP FUNCTION myfunc;
 8501: PL/SQL feature not installed in this TimesTen database
The command failed.

See also

CREATE FUNCTION

DROP INDEX

The DROP INDEX statement deletes the specified index. The index can be global (TimesTen Scaleout or local (TimesTen Scaleoutor TimesTen Classic.

Required privilege

No privilege is required for the index owner. DROP ANY INDEX is required for an index owned by another user.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout. Use the DROP INDEX statement to drop both global and local indexes.

SQL syntax

DROP INDEX [Owner.]IndexName [FROM [Owner.]TableName]

Parameters

Parameter Description

[Owner.]IndexName

Name of the index to be dropped. You can include the name of the owner of the table for the index. For TimesTen Scaleout, the index can be global or local.

[Owner.]TableName

Name of the table upon which the index was created.

Description

  • If you attempt to drop a "busy" index—an index that is in use or that enforces a foreign key—an error results. To drop a foreign key and the index associated with it, use the ALTER TABLE statement.

  • If an index is created on a UNIQUE column constraint, it can only be dropped by dropping the constraint. Use the ALTER TABLE DROP UNIQUE statement for this purpose. Also, see "CREATE TABLE" for more information about the UNIQUE column constraint.

  • If a DROP INDEX operation is or was active in an uncommitted transaction, other transactions that are performing DML operations, that do not access that index, are blocked.

  • If an index is dropped, any prepared statement that uses the index is automatically prepared again the next time the statement is executed.

  • If no table name is specified, the index name must be unique for the specified or implicit owner.

  • If no index owner is specified and a table is specified, the default owner is the table owner.

  • If a table is specified and no owner is specified for it, the default table owner is the current user.

  • The table and index owners must be the same.

  • An index on a temporary table cannot be dropped by a connection if some other connection has an instance of the table that is not empty.

  • If the index is replicated across an active standby pair and if DDL_REPLICATION_LEVEL is 2 or greater, use the DROP INDEX statement to drop the index from the active standby pair in the replication scheme. See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Examples

Drop index partsorderedindex which is defined on table orderitems using one of the following:

DROP INDEX partsorderedindex
   FROM purchasing.orderitems;

Or:

DROP INDEX purchasing.partsorderedindex;

See also

CREATE INDEX

DROP MATERIALIZED VIEW

The DROP MATERIALIZED VIEW statement removes the specified materialized view, including any hash indexes and any range indexes associated with it.

Required privilege

View owner or DROP ANY MATERIALIZED VIEW (if not owner) and

Table owner or DROP ANY TABLE (if not owner) and

Index owner or DROP ANY INDEX (if not owner) if there is an index on the view.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP MATERIALIZED VIEW [Owner.]ViewName

Parameters

Parameter Description

[Owner.]ViewName

Identifies the materialized view to be dropped.

Description

When you execute a DROP MATERIALIZED VIEW operation, the detail tables are updated and locked. An error may result if the detail table was already locked by another transaction.

Examples

The following statement drops the custorder materialized view.

DROP MATERIALIZED VIEW custorder;

DROP PACKAGE [BODY]

The DROP PACKAGE statement removes a stored package from the database. Both the specification and the body are dropped. DROP PACKAGE BODY removes only the body of the package.

Required privilege

No privilege is required for the package owner.

DROP ANY PROCEDURE for another user's package.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP PACKAGE [BODY] [Owner.]PackageName

Parameters

Parameter Description

PACKAGE [BODY]

Specify BODY to drop only the body of the package. Omit BODY to drop both the specification and body of the package.

[Owner.]PackageName

Name of the package to be dropped.

Description

  • When you drop only the body of the package, TimesTen does not invalidate dependent objects. However, you cannot execute one of the procedures or stored functions declared in the package specification until you recreate the package body.

  • TimesTen invalidates any objects that depend on the package specification. If you subsequently reference one of these objects, then TimesTen tries to recompile the object and returns an error if you have not recreated the dropped package.

  • Do not use this statement to remove a single object from the package. Instead, recreate the package without the object using the CREATE PACKAGE and CREATE PACKAGE BODY statements with the OR REPLACE clause.

  • To use the DROP PACKAGE [BODY] statement, you must have PL/SQL enabled in your database. If you do not have PL/SQL enabled in your database, TimesTen returns an error.

Example

The following statement drops the body of package samplePackage:

Command> DROP PACKAGE BODY SamplePackage;
Package body dropped.

To drop both the specification and body of package samplepackage:

Command> DROP PACKAGE samplepackage;
Package dropped.

See also

CREATE PACKAGE

DROP PROCEDURE

The DROP PROCEDURE statement removes a standalone stored procedure from the database. Do not use this statement to remove a procedure that is part of a package.

Required privilege

No privilege is required for the procedure owner.

DROP ANY PROCEDURE for another user's procedure.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP PROCEDURE [Owner.]ProcedureName

Parameters

Parameter Description

[Owner.]ProcedureName

Name of the procedure to be dropped.

Description

  • When you drop a procedure, TimesTen invalidates objects that depend on the dropped procedure. If you subsequently reference one of these objects, TimesTen attempts to recompile the object and returns an error message if you have not recreated the dropped procedure.

  • Do not use this statement to remove a procedure that is part of a package. Either drop the package or redefine the package without the procedure using the CREATE PACKAGE statement with the OR REPLACE clause.

  • To use the DROP PROCEDURE statement, you must have PL/SQL enabled in your database. If you do not have PL/SQL enabled in your database, an error is thrown.

Examples

The following statement drops the procedure myproc and invalidates all objects that depend on myproc:

Command> DROP PROCEDURE myproc;
Procedure dropped.

If PL/SQL is not enabled in your database, TimesTen returns an error:

Command> DROP PROCEDURE myproc;
 
 8501: PL/SQL feature not installed in this TimesTen databaseThe command failed.

See also

CREATE PROCEDURE

DROP PROFILE

The DROP PROFILE statement removes a profile from the database.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP PROFILE profile [CASCADE]

Parameters

Parameter Description

profile

Name of the profile to be dropped.

CASCADE

Specify CASCADE to de-assign the profile from any users to whom the profile is assigned. TimesTen reassigns the DEFAULT profile to such users. You must specify CASCADE to drop a profile that is currently assigned to users.

Description

  • Use this statement to drop an existing profile. You cannot drop the DEFAULT profile. See "CREATE PROFILE" for information on the DEFAULT profile.

  • If you create a profile that is not currently assigned to a user, you do not need to specify CASCADE to drop the profile. If, however, the profile is currently assigned to a user, you must specify CASCADE to drop the profile.

Example

This example creates the test_profile profile and the test_profile_assign_to_user profile. It then creates the test_user user and assigns the test_profile_assign_to_user profile to that user. The example attempts to drop the test_profile profile. The operation succeeds as there are no users assigned to this profile. The example then attempts to drop the test_profile_assign_to_user profile. The operation succeeds if CASCADE is specified. After the test_profile_assign_to_user profile is dropped, the test_user user is assigned the DEFAULT profile.

  1. Create the test_profile profile. Set FAILED_LOGIN_ATTEMPTS to a value of 5.

    Command> CREATE PROFILE test_profile LIMIT FAILED_LOGIN_ATTEMPTS 5;
     
    Profile created.
    
  2. Create the test_profile_assign_to_user profile. Set FAILED_LOGIN_ATTEMPTS to a value of 3.

    Command> CREATE PROFILE test_profile_assign_to_user 
             LIMIT FAILED_LOGIN_ATTEMPTS 3;
     
    Profile created.
    
  3. Create the test_user user and assign the test_profile_assign_to_user profile to this user.

    Command> CREATE USER test_user identified by test_user_pwd 
             PROFILE test_profile_assign_to_user;
     
    User created.
    
  4. Drop the test_profile profile. The DROP PROFILE operation succeeds. There are no users assigned to this test_profile profile.

    Command> DROP PROFILE test_profile;
     
    Profile dropped.
    
  5. Attempt to drop the test_profile_assign_to_user profile. The DROP PROFILE operation fails. There is a user assigned to this profile. Repeat the DROP PROFILE operation again, but this time specify CASCADE. The DROP PROFILE operation succeeds.

    Command> DROP PROFILE test_profile_assign_to_user;
    15178: Profile TEST_PROFILE_ASSIGN_TO_USER has users assigned, cannot drop without CASCADE
    The command failed.
    
    Command> DROP PROFILE test_profile_assign_to_user CASCADE;
     
    Profile dropped.
    
  6. Query the DBA_USERS system view to verify that the test_user user has been assigned the DEFAULT profile.

    Command> SELECT profile FROM dba_users WHERE username = 'TEST_USER';
     
    PROFILE
    < DEFAULT >
    1 row found.

DROP REPLICATION

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

The DROP REPLICATION statement destroys a classic replication scheme and removes it from the executing database.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

DROP REPLICATION [Owner.]ReplicationSchemeName

Parameters

Parameter Description

[Owner.]ReplicationSchemeName

Name assigned to the classic replication scheme.

Description

Dropping the last replication scheme on a database does not delete the replicated tables. These tables exist and persist at a database whether any replication schemes are defined.

Examples

The following statement erases the executing database's knowledge of a classic replication scheme, r:

DROP REPLICATION r;

DROP SEQUENCE

The DROP SEQUENCE statement removes an existing sequence number generator.

If the sequence is replicated across an active standby pair and if DDL_REPLICATION_LEVEL is 3 or greater, the DROP SEQUENCE statement drops the sequence from the active standby pair for all databases in the replication scheme. See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Required privilege

No privilege is required for the sequence owner.

DROP ANY SEQUENCE for another user's sequence.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP SEQUENCE [Owner.]SequenceName

Parameters

Parameter Description

[Owner.]SequenceName

Name of the sequence number generator.

Description

  • Sequences can be dropped while they are in use.

  • If you are using TimesTen Scaleout, you can modify the batch value with the ALTER SEQUENCE statement. Otherwise, to alter a sequence, use the DROP SEQUENCE statement and then create a new sequence with the same name. For example, to change the MINVALUE, drop the sequence and recreate it with the same name and with the desired MINVALUE.

  • If the sequence is part of a replication scheme, use the ALTER REPLICATION statement to drop the sequence from the replication scheme. Then use the DROP SEQUENCE statement to drop the sequence.

Examples

The following statement drops mysequence:

DROP SEQUENCE mysequence;

See also

CREATE SEQUENCE

DROP SYNONYM

The DROP SYNONYM statement removes a synonym from the database.

If the synonym is replicated across an active standby pair and if DDL_REPLICATION_LEVEL is 2 or greater, the DROP SYNONYM statement drops the synonym from the active standby pair for all databases in the replication scheme. See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Required privilege

No privilege is required to drop the private synonym by its owner. The DROP ANY SYNONYM privilege is required to drop another user's private synonym.

The DROP PUBLIC SYNONYM privilege is required to drop a PUBLIC synonym.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

To drop a private synonym, use the following syntax:

DROP SYNONYM [Owner.]SynonymName

To drop a public synonym:

DROP PUBLIC SYNONYM SynonymName

Parameters

Parameter Description

PUBLIC

Specify PUBLIC to drop a public synonym.

Owner

Optionally, specify the owner for a private synonym. If you omit the owner, the private synonym must exist in the current user's schema.

SynonymName

Specify the name of the synonym to be dropped.

Examples

Drop the public synonym pubemp:

DROP PUBLIC SYNONYM pubemp;
Synonym dropped.

Drop the private synjobs synonym:

DROP SYNONYM synjobs;
Synonym dropped.

As user terry with DROP ANY SYNONYM privilege, drop the private syntab synonym owned by ttuser.

DROP SYNONYM ttuser.syntab;
Synonym dropped.

See also

CREATE SYNONYM

DROP TABLE

The DROP TABLE statement removes the specified table, including any hash indexes and any range indexes associated with it.

Required privilege

No privilege is required for the table owner.

DROP ANY TABLE for another user's table.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP TABLE [Owner.]TableName

Parameters

Parameter Description

[Owner.]TableName

Identifies the table to be dropped.

Description

  • If you attempt to drop a table that is in use, an error results.

  • If DROP TABLE is or was active in an uncommitted transaction, other transactions doing DML operations that do not access that table are allowed to proceed.

  • If the table is a replicated table, you can do one of the following:

    • Use the DROP REPLICATION statement to drop the replication scheme before issuing the DROP TABLE statement.

    • If DDL_REPLICATION_LEVEL is 2 or greater, the DROP TABLE statement drops the table from the active standby pair for all databases in the replication scheme.

      If DDL_REPLICATION_LEVEL is 1, stop the replication agent and use the ALTER ACTIVE STANDBY PAIR ... EXCLUDE TABLE statement to exclude the table from the replication scheme. Then use the DROP TABLE statement to drop the table.

      See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

  • A temporary table cannot be dropped by a connection if some other connection has some non-empty instance of the table.

Examples

CREATE TABLE vendorperf
   (ordernumber INTEGER,
    delivday TT_SMALLINT,
    delivmonth TT_SMALLINT,
    delivyear TT_SMALLINT,
    delivqty TT_SMALLINT,
    remarks VARCHAR2(60));
CREATE UNIQUE INDEX vendorperfindex ON vendorperf (ordernumber);

The following statement drops the table and index.

DROP TABLE vendorperf;

DROP USER

The DROP USER statement removes a user from the database.

Required privilege

ADMIN

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP USER user

Parameters

Parameter Description

user

Name of the user that is being dropped from the database.

Description

Before you can drop a user:

  • The user must exist either internally or externally in the database.

  • You must drop objects that the user owns.

  • When replication is configured, this statement is replicated.

Examples

Drop user terry from the database:

DROP USER terry;
User dropped.

DROP VIEW

The DROP VIEW statement removes the specified view.

If the view is replicated across an active standby pair and if DDL_REPLICATION_LEVEL is 3 or greater, the DROP VIEW statement drops the view from the active standby pair for all databases in the replication scheme. See "Making DDL changes in an active standby pair" in the Oracle TimesTen In-Memory Database Replication Guide for more information.

Required privilege

View owner or DROP ANY VIEW (if not owner)

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

DROP VIEW [Owner.]ViewName

Parameters

Parameter Description

[Owner.]ViewName

Identifies the view to be dropped.

Examples

The following statement drops the custorder view.

DROP VIEW custorder;

See also

FLUSH CACHE GROUP

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

The FLUSH CACHE GROUP statement flushes data from TimesTen cache tables to Oracle Database tables. This statement is available only for user managed cache groups.

There are two variants to this operation: one that accepts a WHERE clause, and one that accepts a WITH ID clause.

FLUSH CACHE GROUP is meant to be used when commit propagation (from TimesTen to Oracle Database) is turned off. Instead of propagating every transaction upon commit, many transactions can be committed before changes are propagated to Oracle Database. For each cache instance ID, if the cache instance exists in the Oracle database, the operation in the Oracle database consists of an update. If the cache instance does not exist in the Oracle database, TimesTen inserts it.

This is useful, for example, in a shopping cart application in which many changes may be made to the cart, which uses TimesTen as a high-speed cache, before the order is committed to the master Oracle database table.

Note:

Using a WITH ID clause usually results in better system performance than using a WHERE clause.

Only inserts and updates are flushed. Inserts are propagated as inserts if the record does not exist in the Oracle database table or as updates (if the record already exists). It is not possible to flush a delete. That is, if a record is deleted on TimesTen, there is no way to "flush" that delete to the Oracle database table. Deletes must be propagated either manually or by turning commit propagation on. Attempts to flush deleted records are silently ignored. No error or warning is issued. Records from tables that are specified as READ ONLY or PROPAGATE cannot be flushed to the Oracle database tables.

Required privilege

No privilege is required for the cache group owner.

FLUSH or FLUSH ANY CACHE GROUP for another user's cache group.

INSERT, DELETE, UPDATE privileges on underlying tables.

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

FLUSH CACHE GROUP [Owner.]GroupName
[WHERE ConditionalExpression]

or

FLUSH CACHE GROUP [Owner.]GroupName
WITH ID (ColumnValueList)

Parameters

Parameter Description

[Owner.]GroupName

Name of the cache group to be flushed.

WHERE ConditionalExpression

Use the WHERE clause to specify a search condition to qualify the target rows of the cache operation. If you use more than one table in the WHERE clause and the tables have columns with the same names, fully qualify the table names.

WITH ID ColumnValueList

The WITH ID clauses enables you to use primary key values to flush the cache instance. Specify ColumnValueList as either a list of literals or binding parameters to represent the primary key values.

Description

  • WHERE clauses are generally used to apply the operation to a set of cache instances, rather than to a single cache instance or to all cache instances. The flush operation uses the WHERE clause to determine which cache instances to send to the Oracle database.

  • Generally, you do not have to fully qualify the column names in the WHERE clause of the FLUSH CACHE GROUP statement. However, since TimesTen automatically generates queries that join multiple tables in the same cache group, a column must be fully qualified if there is more than one table in the cache group that contains columns with the same name. Without an owner name, all tables referenced by cache group WHERE clauses are owned by the current login name executing the cache group operation.

  • When the WHERE clause is omitted, the entire contents of the cache group is flushed to the Oracle database tables. When the WHERE clause is included, it is allowed to include only the root table.

  • Following the execution of a FLUSH CACHE GROUP statement, the ODBC function SQLRowCount(), the JDBC method getUpdateCount(), and the OCI function OCIAttrGet() with the OCI_ATTR_ROW_COUNT argument return the number of cache instances that were flushed.

  • Use the WITH ID clause to specify binding parameters.

Restrictions

Do not use the WITH ID clause when flushing:

  • Static user managed cache group with the AUTOREFRESH attribute

  • AWT or SWT cache groups

Examples

FLUSH CACHE GROUP marketbasket;

FLUSH CACHE GROUP marketbasket
WITH ID(10);

GRANT

The GRANT statement assigns one or more privileges to a user.

Required privilege

ADMIN to grant system privileges.

ADMIN or the object owner to grant object privileges.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

GRANT {SystemPrivilege [,...] | ALL [PRIVILEGES]} [...] TO {user |PUBLIC} [,...]

or

GRANT {{ObjectPrivilege [,...] | ALL [PRIVILEGES]} ON {[Owner.]object}[,...]} TO
{user | PUBLIC} [,...]

Parameters

The following parameters are for granting system privileges:

Parameter Description

SystemPrivilege

This is the system privilege to grant. See "System privileges" for a list of acceptable values.

ALL [PRIVILEGES]

Assigns all system privileges to the user.

user

Name of the user to whom privileges are being granted. The user name must first have been introduced to the TimesTen database by a CREATE USER statement.

PUBLIC

Specifies that the privilege is granted to all users.

The following parameters are for granting object privileges:

Parameter Description

ObjectPrivilege

This is the object privilege to grant. See "Object privileges" for a list of acceptable values.

ALL [PRIVILEGES]

Assigns all object privileges to the user.

[Owner.]object

object is the name of the object on which privileges are being granted. Owner is the owner of the object. If Owner is not specified, the user who is granting the privilege is the owner.

user

Name of the user to whom privileges are being granted. The user must exist in the database.

PUBLIC

Specifies that the privilege is granted to all users.

Description

  • One or more system privileges can be granted to a user by a user with ADMIN privilege.

  • One or more object privileges can be granted to a user by the owner of the object.

  • One or more object privileges can be granted to a user on any object by a user with ADMIN privilege.

  • To remove a privilege from a user, use the REVOKE statement.

  • You cannot grant system privileges and object privileges in the same statement.

  • Only one object can be specified in an object privilege statement.

  • When replication is configured, this statement is replicated.

Examples

Grant the ADMIN privilege to the user terry:

GRANT admin TO terry;

Assuming the grantor has ADMIN privilege, grant the SELECT privilege to user terry on the customers table owned by user pat:

GRANT SELECT ON pat.customers TO terry;

Grant an object privilege to user terry:

GRANT SELECT ON emp_details_view TO terry;

INSERT

The INSERT statement adds rows to a table.

The following expressions can be used in the VALUES clause of an INSERT statement:

Required privilege

No privilege is required for the table owner.

INSERT for another user's table.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

INSERT [hint] INTO [Owner.]TableName [(Column [,...])]
VALUES (SingleRowValues)
[RETURNING|RETURN Expression[,...] INTO DataItem[,...]]

The SingleRowValues parameter has the syntax:

{NULL|{?|:DynamicParameter}|{Constant}| DEFAULT}[,...]

Parameters

Parameter Description

hint

Specifies a statement level optimizer hint for the INSERT statement. See "Statement level optimizer hints" for information on statement level optimizer hints.

Owner

The owner of the table into which data is inserted.

TableName

Name of the table into which data is inserted.

Column

Each column in this list is assigned a value from SingleRowValues.

If you omit one or more of the table's columns from this list, then the value of the omitted column in the inserted row is the column default value as specified when the table was created or last altered. If any omitted column has a NOT NULL constraint and has no default value, then the database returns an error.

If you omit a list of columns completely, then you must specify values for all columns in the table.

?

:DynamicParameter

Placeholder for a dynamic parameter in a prepared SQL statement. The value of the dynamic parameter is supplied when the statement is executed.

Constant

A specific value. See "Constants" for information on constants.

DEFAULT

Specifies that the column should be updated with the default value.

Expression

Valid expression syntax. See Expressions for information on expressions.

DataItem

Host variable or PL/SQL variable that stores the retrieved Expression value.

Description

  • If you omit any of the table's columns from the column name list, the INSERT statement places the default value in the omitted columns. If the table definition specifies NOT NULL for any of the omitted columns and there is no default value, the INSERT statement fails.

  • BINARY and VARBINARY data can be inserted in character or hexadecimal format:

    • Character format requires single quotes.

    • Hexadecimal format requires the prefix 0x before the value.

  • The INSERT operation fails if it violates a foreign key constraint. See "CREATE TABLE" for a description of the foreign key constraint.

  • Restrictions on the RETURNING clause:

    • Each Expression must be a simple expression. Aggregate functions are not supported.

    • You cannot return a sequence number into an OUT parameter.

    • ROWNUM and subqueries cannot be used in the RETURNING clause.

    • Parameters in the RETURNING clause cannot be duplicated anywhere in the INSERT statement.

    • In PL/SQL, you cannot use a RETURNING clause with a WHERE CURRENT operation.

Examples

A new single row is added to the purchasing.vendors table.

INSERT INTO purchasing.vendors
VALUES (9016,
       'Secure Systems, Inc.',
       'Jane Secret',
       '454-255-2087',
       '1111 Encryption Way',
       'Hush',
       'MD',
       '00007',
       'discount rates are secret');

For dynamic parameters :pno and :pname, values are supplied at runtime.

INSERT INTO purchasing.parts (partnumber, partname)
  VALUES (:pno, :pname);

Return the annual salary and job_id of a new employee. Declare the variables sal and jobid with the same data types as salary and job_id. Insert the row into employees. Print the variables for verification.

Command> VARIABLE sal12 NUMBER(8,2);
Command> VARIABLE jobid VARCHAR2(10) INLINE NOT NULL;

Command> INSERT INTO employees(employee_id, last_name, email, hire_date, 
           job_id, salary)
         VALUES (211,'Doe','JDOE',sysdate,'ST_CLERK',2400)
         RETURNING salary*12, job_id INTO :sal12,:jobid;
1 row inserted.

PRINT sal12 jobid;
SAL12                 : 28800
JOBID                 : ST_CLERK

INSERT...SELECT

The INSERT...SELECT statement inserts the results of a query into a table.

Required privilege

No privilege is required for the object owner.

INSERT and SELECT for another user's object.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

INSERT INTO [Owner.]TableName [(ColumnName [,...])] InsertQuery

Parameters

Parameter Description

[Owner.]TableName

Table to which data is to be added.

ColumnName

Column for which values are supplied. If you omit any of the table's columns from the column name list, the INSERT...SELECT statement places the default value in the omitted columns. If the table definition specifies NOT NULL, without a default value, for any of the omitted columns, the INSERT...SELECT statement fails. You can omit the column name list if you provide values for all columns of the table in the same order the columns were specified in the CREATE TABLE statement. If too few values are provided, the remaining columns are assigned default values.

InsertQuery

Any supported SELECT query. See "SELECT". You can specify a statement level optimizer hint after the SELECT verb. See "Statement level optimizer hints" for more information on statement level optimizer hints.

Description

  • The column types of the result set must be compatible with the column types of the target table.

  • You can specify a sequence CURRVAL or NEXTVAL when inserting values. See "Using CURRVAL and NEXTVAL in TimesTen Classic" for more details.

  • In the InsertQuery, the ORDER BY clause is allowed. The sort order may be modified using the ORDER BY clause when the result set is inserted into the target table, but the order is not guaranteed.

  • The INSERT operation fails if there is an error in the InsertQuery.

  • A RETURNING clause cannot be used in an INSERT...SELECT statement.

  • The SELECT subquery in a UNION, UNION ALL, MINUS, or INTERSECT must have the same number of projected expressions.

Examples

New rows are added to the purchasing.parts table that describe which parts are delivered in 20 days or less.

INSERT INTO purchasing.parts
SELECT partnumber, deliverydays
FROM purchasing.supplyprice
WHERE deliverydays < 20;

LOAD CACHE GROUP

The LOAD CACHE GROUP statement loads data from Oracle database tables into a TimesTen cache group.

Required privilege

No privilege is required for the cache group owner.

LOAD or LOAD ANY CACHE GROUP for another user's cache group.

INSERT, DELETE, UPDATE privileges on underlying tables.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

LOAD CACHE GROUP [Owner.]GroupName 
[WHERE ConditionalExpression] 
COMMIT EVERY n ROWS
[PARALLEL NumThreads [READERS NumReaders]]

or

LOAD CACHE GROUP [Owner.]GroupName
WITH ID (ColumnValueList)

Note:

The WITH ID clause is not supported in TimesTen Scaleout.

Parameters

Parameter Description

[Owner.]GroupName

Name assigned to the cache group.

WHERE ConditionalExpression

Use the WHERE clause to specify a search condition to qualify the target rows of the cache operation. If you use more than one table in the WHERE clause and the tables have columns with the same names, fully qualify the table names.

COMMIT EVERY n ROWS

Use the COMMIT EVERY n ROWS clause to indicate the frequency (based on the number of rows that are loaded into the cache group) at which a commit is issued during the load operation. This clause is required if you do not specify the WITH ID clause.

n represents the number of rows that are loaded into the cache group before the transaction is committed. Express n as an integer (where n >= 0). If you specify 0 for n, the LOAD CACHE GROUP statement is executed as one transaction.

[PARALLEL NumThreads]

Provides parallel loading for cache group tables. Specifies the number of loading threads to run concurrently. One thread performs the bulk fetch from the Oracle database and the other threads (NumThreads - 1 threads) perform the inserts into TimesTen. Each thread uses its own connection or transaction.

The minimum value for NumThreads is 2. The maximum value is 10. If you specify a value greater than 10, TimesTen assigns the value 10.

[READERS NumReaders]

This option specifies the total number of threads from the NumThreads parameter to use for bulk fetching from the Oracle database.

For example, if you specify a NumThreads parameter of 8 and a readers option of 3, then 3 threads are used to bulk fetch data from the Oracle database and 5 threads are used to load data into the TimesTen database.

Express NumReaders as an integer where 0 < NumReaders < NumThreads.

WITH ID ColumnValueList

The WITH ID clauses enables you to use primary key values to load the cache instance. Specify ColumnValueList as either a list of literals or binding parameters to represent the primary key values.

The WITH ID clause is not supported in TimesTen Scaleout.

Description

  • LOAD CACHE GROUP loads all new cache instances from the Oracle database that satisfy the cache group definition and are not yet present in the cache group.

  • Before issuing the LOAD CACHE GROUP statement, ensure that the replication agent is running if the cache group is replicated or is an AWT cache group. Make sure the cache agent is running.

  • LOAD CACHE GROUP is executed in its own transaction, and must be the first operation in a transaction.

  • LOAD CACHE GROUP only loads new (inserted) rows on the Oracle database tables into the corresponding TimesTen cache tables.

  • Errors cause a rollback. When cache instances are committed periodically, errors abort the remainder of the load. The load is rolled back to the last commit.

  • If the LOAD CACHE GROUP statement fails when you specify COMMIT EVERY n ROWS (where n >= 0), the content of the target cache group could be in an inconsistent state since some loaded rows are already committed. Some cache instances may be partially loaded. Use the UNLOAD CACHE GROUP statement to unload the cache group, then reload the cache group.

  • Generally, you do not have to fully qualify the column names in the WHERE clause of the LOAD CACHE GROUP statement. However, since TimesTen automatically generates queries that join multiple tables in the same cache group, a column must be fully qualified if there is more than one table in the cache group that contains columns with the same name.

  • When loading a read-only cache group:

    • The AUTOREFRESH state must be paused.

    • The LOAD CACHE GROUP statement cannot have a WHERE clause (except on a dynamic cache group).

    • The cache group must be empty.

  • The automatic refresh state of a cache group may change after a LOAD CACHE GROUP operation completes. See "Loading and refreshing a dynamic cache group with autorefresh" in the Oracle TimesTen In-Memory Database Cache Guide for information.

  • Following the execution of a LOAD CACHE GROUP statement, the ODBC function SQLRowCount(), the JDBC method getUpdateCount(), and the OCI function OCIAttrGet() with the OCI_ATTR_ROW_COUNT argument return the number of cache instances that were loaded.

  • Use the WITH ID clause as follows:

    • In place of the WHERE clause for faster loading of the cache instance

    • To specify binding parameters

    • To roll back the load transaction upon failure

Restrictions

  • The LOAD CACHE GROUP...WITH ID clause is not supported in TimesTen Scaleout.

  • Do not reference child tables in the WHERE clause.

  • Do not specify the PARALLEL clause in the following circumstances:

    • With the WITH ID clause

    • With the COMMIT EVERY 0 ROWS clause

    • When database level locking is enabled (connection attribute LockLevel is set to 1)

  • Do not use the WITH ID clause when loading these types of cache groups:

    • Static read-only cache group

    • Static user managed cache group with the autorefresh attribute

    • User managed cache group with the AUTOREFRESH and PROPAGATE attributes

  • Do not use the WITH ID clause with the COMMIT EVERY n ROWS clause.

Examples

CREATE CACHE GROUP recreation.cache
    FROM recreation.clubs (
        clubname CHAR(15) NOT NULL,
        clubphone SMALLINT,
        activity CHAR(18),
        PRIMARY KEY(clubname))
    WHERE (recreation.clubs.activity IS NOT NULL);

LOAD CACHE GROUP recreation.cache
    COMMIT EVERY 30 ROWS;

Use the HR schema to illustrate the use of the PARALLEL clause with the LOAD CACHE GROUP statement. The COMMIT EVERY n ROWS clause is required. Issue the CACHEGROUPS command. You see cache group cg2 is defined and the autorefresh state is paused. Unload cache group cg2, then specify the LOAD CACHE GROUP statement with the PARALLEL clause to provide parallel loading. You see 25 cache instances loaded.

Command> CACHEGROUPS;

Cache Group SAMPLEUSER.CG2:

  Cache Group Type: Read Only
  Autorefresh: Yes
  Autorefresh Mode: Incremental
  Autorefresh State: Paused
  Autorefresh Interval: 1.5 Minutes

  Root Table: SAMPLEUSER.COUNTRIES
  Table Type: Read Only

  Child Table: SAMPLEUSER.LOCATIONS
  Table Type: Read Only

  Child Table: SAMPLEUSER.DEPARTMENTS
  Table Type: Read Only

1 cache group found.

Command> UNLOAD CACHE GROUP cg2;
25 cache instances affected.
Command> COMMIT;
Command> LOAD CACHE GROUP cg2 COMMIT EVERY 10 ROWS PARALLEL 2;
25 cache instances affected.
Command> COMMIT;

The following example loads only the cache instances for customers whose customer number is greater than or equal to 5000 into the TimesTen cache tables in the new_customers cache group from the corresponding Oracle database tables:

LOAD CACHE GROUP new_customers WHERE (oratt.customer.cust_num >= 5000) COMMIT
EVERY 256 ROWS;

MERGE

This statement is not supported in TimesTen Scaleout.

In TimesTen Classic:

The MERGE statement enables you to select rows from one or more sources for update or insertion into a target table. You can specify conditions that are used to evaluate which rows are updated or inserted into the target table.

Use this statement to combine multiple INSERT and UPDATE statements.

MERGE is a deterministic statement: You cannot update the same row of the target table multiple times in the same MERGE statement.

Required privilege

No privilege is required for the owner of the target table and the source table.

INSERT or UPDATE on a target table owned by another user and SELECT on a source table owned by another user.

Usage with TimesTen Scaleout

This statement is not supported with TimesTen Scaleout.

SQL syntax

MERGE [hint] INTO [Owner.]TargetTableName [Alias] USING
    {[Owner.]SourceTableName|(Subquery)}[Alias] ON (Condtion)
    {MergeUpdateClause MergeInsertClause |
     MergeInsertClause MergeUpdateClause |
     MergeUpdateClause | MergeInsertClause
    }

The syntax for MergeUpdateClause is as follows:

WHEN MATCHED THEN UPDATE SET SetClause [WHERE Condition1]

The syntax for MergeInsertClause is as follows:

WHEN NOT MATCHED THEN INSERT [Columns [,...]] VALUES
    ( {{Expression | DEFAULT|NULL} [,...] }) [WHERE Condition2]

Parameters

Parameter Description

hint

Specifies a statement level optimizer hint for the MERGE statement. See "Statement level optimizer hints" for information on statement level optimizer hints.

[Owner.]TargetTableName

Name of the target table. This is the table in which rows are either updated or inserted.

[Alias]

You can optionally specify an alias name for the target or source table.

USING {[Owner.]SourceTableName | (Subquery)} [Alias]

The USING clause indicates the table name or the subquery that is used for the source of the data. Use a subquery to use joins or aggregates. Optionally, you can specify an alias for the table name or the subquery.

ON (Condition)

Specify the condition used to evaluate each row of the target table to determine if the row should be considered for either a merge insert or a merge update. If the condition is true when evaluated, then the MergeUpdateClause is considered for the target row using the matching row from the SourceTableName. An error is generated if more than one row in the source table matches the same row in the target table. If the condition is not true when evaluated, then the MergeInsertClause is considered for that row.

SET SetClause

Clause used with the UPDATE statement. See "UPDATE" for information on the UPDATE statement.

[WHERE Condition1]

For each row that matches the ON (Condition), Condition1 is evaluated. If the condition is true when evaluated, the row is updated. You can refer to either the target table or the source table in this clause. You cannot use a subquery. The clause is optional.

INSERT [Columns[,...]]VALUES ({{Expression |DEFAULT|NULL} [,...]})

Columns to insert into the target table. See "INSERT" for information on the INSERT statement.

[WHERE Condition2]

If specified, Condition2 is evaluated. If the condition is true when evaluated, the row is inserted into the target table. The condition can refer to the source table only. You cannot use a subquery.

Description

  • You can specify the MergeUpdateClause and MergeInsertClause together or separately. If you specify both, they can be in either order.

  • If DUAL is the only table specified in the USING clause and it is not referenced elsewhere in the MERGE statement, specify DUAL as a simple table rather than using it in a subquery. In this simple case, to help performance, specify a key condition on a unique index of the target table in the ON clause.

  • Restrictions on the MergeUpdateClause:

    • You cannot update a column that is referenced in the ON condition clause.

    • You cannot update source table columns.

  • Restrictions on the MergeInsertClause:

    • You cannot insert values of target table columns.

  • Other restrictions:

    • Do not use the set operators in the subquery of the source table.

    • Do not use a subquery in the WHERE condition of either the MergeUpdateClause or the MergeInsertClause.

    • The target table cannot be a detail table of a materialized view.

    • The RETURNING clause cannot be used in a MERGE statement.

Examples

In this example, dual is specified as a simple table. There is a key condition on the UNIQUE index of the target table specified in the ON clause. The DuplicateBindMode attribute is set to 1 in this example. (The default is 0.)

Command> CREATE TABLE mergedualex (col1 TT_INTEGER NOT NULL, 
           col2 TT_INTEGER, PRIMARY KEY (col1));
Command> MERGE INTO mergedualex USING dual ON (col1 = :v1)
         WHEN MATCHED THEN UPDATE SET col2 = col2 + 1
         WHEN NOT MATCHED THEN INSERT VALUES (:v1, 1);
Type '?' for help on entering parameter values.
Type '*' to end prompting and abort the command.
Type '-' to leave the parameter unbound.
Type '/;' to leave the remaining parameters unbound and execute the command.

Enter Parameter 1  'V1' (TT_INTEGER) > 10
1 row merged.
Command> SELECT * FROM mergedualex;
< 10, 1 >
1 row found.

In this example, a table called contacts is created with columns employee_id and manager_id. One row is inserted into contacts with values 101 and NULL for employee_id and manager_id, respectively. The MERGE statement is used to insert rows into contacts using the data in the employees table. A SELECT FIRST 3 rows is used to illustrate that in the case where employee_id is equal to 101, manager_id is updated to 100. The remaining 106 rows from the employees table are inserted into contacts:

Command> CREATE TABLE contacts (employee_id NUMBER (6) NOT NULL PRIMARY KEY, 
                                manager_id NUMBER (6));
Command> SELECT employee_id, manager_id FROM employees WHERE employee_id =101;
< 101, 100 >
1 row found.
Command> INSERT INTO contacts VALUES (101,null);
1 row inserted.
Command> SELECT COUNT (*) FROM employees;
< 107 >
1 row found.
Command> MERGE INTO contacts c
         USING employees e
         ON (c.employee_id = e.employee_id)
         WHEN MATCHED THEN
         UPDATE SET c.manager_id = e.manager_id
         WHEN NOT MATCHED THEN
         INSERT (employee_id, manager_id)
         VALUES (e.employee_id, e.manager_id);
107 rows merged.
Command> SELECT COUNT (*) FROM contacts;
< 107 >
1 row found.
Command> SELECT FIRST 3 employee_id,manager_id FROM employees;
< 100, <NULL> >
< 101, 100 >
< 102, 100 >
3 rows found.
Command> SELECT FIRST 3 employee_id, manager_id FROM contacts;
< 100, <NULL> >
< 101, 100 >
< 102, 100 >
3 rows found.

REFRESH CACHE GROUP

The REFRESH CACHE GROUP statement replaces data in the TimesTen cache tables with the most current committed data from the Oracle database cached tables.

Required privilege

CREATE SESSION on the Oracle Database schema and SELECT on the Oracle Database tables.

No privilege for the cache group is required for the cache group owner.

REFRESH or REFRESH ANY CACHE GROUP for another user's cache group.

INSERT, DELETE, UPDATE privileges on underlying tables.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

REFRESH CACHE GROUP [Owner.]GroupName 
[WHERE ConditionalExpression]
COMMIT EVERY n ROWS
[PARALLEL NumThreads]

or

REFRESH CACHE GROUP [Owner.]GroupName 
WITH ID (ColumnValueList)

Note:

The WITH ID clause is not supported in TimesTen Scaleout.

Parameters

Parameter Description

[Owner.]GroupName

Name assigned to the cache group.

WHERE ConditionalExpression

Use the WHERE clause to specify a search condition to qualify the target rows of the cache operation. If you use more than one table in the WHERE clause and the tables have columns with the same names, fully qualify the table names.

COMMIT EVERY n ROWS

Use the COMMIT EVERY n ROWS clause to indicate the frequency (based on the number of rows that are refreshed in the cache group) at which a commit is issued during the refresh operation. This clause is required if you do not specify the WITH ID clause.

n represents the number of rows that are refreshed before the transaction is committed. Express n as an integer (where n >= 0). If you specify 0 for n, the REFRESH CACHE GROUP statement is executed as one transaction.

[PARALLEL NumThreads]

Provides parallel loading for cache group tables. Specifies the number of loading threads to run concurrently. One thread performs the bulk fetch from the Oracle database and the other threads (NumThreads - 1 threads) perform the inserts into TimesTen. Each thread uses its own connection or transaction.

The minimum value for NumThreads is 2. The maximum value is 10. If you specify a value greater than 10, TimesTen assigns the value 10.

WITH ID ColumnValueList

The WITH ID clauses enables you to use primary key values to refresh the cache instance. Specify ColumnValueList as either a list of literals or binding parameters to represent the primary key values.

The WITH ID clause is not supported in TimesTen Scaleout.

Description

  • A REFRESH CACHE GROUP statement must be executed in its own transaction.

  • Before issuing the REFRESH CACHE GROUP statement, ensure that the replication agent is running if the cache group is replicated or is an AWT cache group. Make sure the cache agent is running.

  • The REFRESH CACHE GROUP statement replaces data in the TimesTen cached tables with the most current committed data from the cached Oracle database tables, including data that already exists in the TimesTen cached tables. For an explicitly loaded cache group, a refresh operation is equivalent to issuing an UNLOAD CACHE GROUP statement followed by a LOAD CACHE GROUP statement. Operations on all rows in the Oracle database tables including inserts, updates, and deletes are applied to the cache tables. For dynamic cache groups, a refresh operation refreshes only rows that are updated or deleted on the Oracle database tables into the cache tables. For more information on explicitly loaded and dynamic cache groups, see "Transmitting changes between the TimesTen and Oracle databases" in Oracle TimesTen In-Memory Database Cache Guide.

  • When refreshing a read-only cache group:

    • The AUTOREFRESH state must be paused.

    • If the cache group is a read-only dynamic cache group, do not use the PARALLEL clause.

  • If the automatic refresh state of a cache group (dynamic or explicitly loaded) is PAUSED, the state is changed to ON after an unconditional REFRESH CACHE GROUP statement issued on the cache group completes.

  • If the automatic refresh state of a dynamic cache group is PAUSED, the state remains PAUSED after a REFRESH CACHE GROUP...WITH ID statement completes.

  • Generally, you do not have to fully qualify the column names in the WHERE clause of the REFRESH CACHE GROUP statement. However, since TimesTen automatically generates queries that join multiple tables in the same cache group, a column must be fully qualified if there is more than one table in the cache group that contains columns with the same name.

  • If the REFRESH CACHE GROUP statement fails when you specify COMMIT EVERY n ROWS (where n >= 0), the content of the target cache group could be in an inconsistent state since some loaded rows are already committed. Some cache instances may be partially loaded. Use the UNLOAD CACHE GROUP statement to unload the cache group, then use the LOAD CACHE GROUP statement to reload the cache group.

  • Following the execution of a REFRESH CACHE GROUP statement, the ODBC function SQLRowCount(), the JDBC method getUpdateCount(), and the OCI function OCIAttrGet() with the OCI_ATTR_ROW_COUNT argument return the number of cache instances that were refreshed.

  • Use the WITH ID clause:

    • In place of the WHERE clause for faster refreshing of the cache instance

    • To specify binding parameters

    • To roll back the refresh transaction upon failure

Restrictions

  • The REFRESH CACHE GROUP...WITH ID clause is not supported in TimesTen Scaleout.

  • Do not specify the PARALLEL clause:

    • With the WITH ID clause

    • With the COMMIT EVERY n ROWS clause

    • When database level locking is enabled (connection attribute LockLevel is set to 1)

    • For read-only dynamic cache groups

  • Do not use the WITH ID clause when refreshing these types of cache groups:

    • Static read-only cache groups

    • Static user managed cache groups with the autorefresh attribute

    • User managed cache groups with the autorefresh and propagate attributes

  • Do not use the WITH ID clause with the COMMIT EVERY n ROWS clause.

  • Do not use the WHERE clause with dynamic or read-only cache groups.

Examples

REFRESH CACHE GROUP recreation.cache COMMIT EVERY 30 ROWS;

Is equivalent to:

UNLOAD CACHE GROUP recreation.cache;
LOAD CACHE GROUP recreation.cache COMMIT EVERY 30 ROWS;

Use the HR schema to illustrate the use of the PARALLEL clause with the REFRESH CACHE GROUP statement. The COMMIT EVERY n ROWS is required. Issue the CACHEGROUPS command. You see cache group cg2 is defined and the autorefresh state is paused. Specify the REFRESH CACHE GROUP statement with the PARALLEL clause to provide parallel loading. You see 25 cache instances refreshed.

Command> CACHEGROUPS;

Cache Group SAMPLEUSER.CG2:

  Cache Group Type: Read Only
  Autorefresh: Yes
  Autorefresh Mode: Incremental
  Autorefresh State: Paused
  Autorefresh Interval: 1.5 Minutes

  Root Table: SAMPLEUSER.COUNTRIES
  Table Type: Read Only

  Child Table: SAMPLEUSER.LOCATIONS
  Table Type: Read Only

  Child Table: SAMPLEUSER.DEPARTMENTS
  Table Type: Read Only

1 cache group found.
Command> REFRESH CACHE GROUP cg2 COMMIT EVERY 20 ROWS PARALLEL 2;
25 cache instances affected.

REVOKE

The REVOKE statement removes one or more privileges from a user.

Required privilege

ADMIN to revoke system privileges.

ADMIN or object owner to revoke object privileges.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

REVOKE {SystemPrivilege [,...] | ALL [PRIVILEGES]} FROM {User |PUBLIC} [,...]

or

REVOKE {{ObjectPrivilege [,...] | ALL [PRIVILEGES]} ON {[Owner.Object}} [,...]
FROM {user | PUBLIC}[,...]

Parameters

The following parameters are for revoking system privileges:

Parameter Description

SystemPrivilege

This is the system privilege to revoke. See "System privileges" for a list of acceptable values.

ALL [PRIVILEGES]

Revokes all system privileges from the user.

User

Name of the user from whom privileges are being revoked. The user name must first have been introduced to the TimesTen database by a CREATE USER statement.

PUBLIC

Specifies that the privilege is revoked for all users.

The following parameters are for revoking object privileges:

Parameter Description

ObjectPrivilege

This is the object privilege to revoke. See "Object privileges" for a list of acceptable values.

ALL [PRIVILEGES]

Revokes all object privileges from the user.

User

Name of the user from whom privileges are to be revoked. The user name must first have been introduced to the TimesTen database through a CREATE USER statement.

[Owner.]Object

Object is the name of the object on which privileges are being revoked. Owner is the owner of the object. If Owner is not specified, then the user who is revoking the privilege is known as the owner.

PUBLIC

Specifies that the privilege is revoked for all users.

Description

  • Privileges on objects cannot be revoked from the owner of the objects.

  • Any user who can grant a privilege can revoke the privilege even if they were not the user who originally granted the privilege.

  • Privileges must be revoked at the same level they were granted. You cannot revoke an object privilege from a user who has the associated system privilege. For example, if you grant SELECT ANY TABLE to a user and then try to revoke SELECT ON BOB.TABLE1, the revoke fails unless you have specifically granted SELECT ON BOB.TABLE1 in addition to SELECT ANY TABLE.

  • If a user has been granted all system privileges, you can revoke a specific privilege. For example, you can revoke ALTER ANY TABLE from a user who has been granted all system privileges.

  • If a user has been granted all object privileges, you can revoke a specific privilege on a specific object from the user. For example, you can revoke the DELETE privilege on table CUSTOMERS from user TERRY even if TERRY has previously been granted all object privileges.

  • You can revoke all privileges from a user even if the user has not previously been granted all privileges.

  • You cannot revoke a specific privilege from a user who has not been granted the privilege.

  • You cannot revoke privileges on objects owned by a user.

  • You cannot revoke system privileges and object privileges in the same statement.

  • You can specify only one object in an object privilege statement.

  • Revoking the SELECT privilege on a detail table or a system privilege that includes the SELECT privilege from user2 on a detail table owned by user1 causes associated materialized views owned by user2 to be marked invalid.

  • When replication is configured, this statement is replicated.

Examples

Revoke the ADMIN and DDL privileges from the user terry:

REVOKE admin, ddl FROM terry;

Assuming the revoker has ADMIN privilege, revoke the UPDATE privilege from terry on the customers table owned by pat:

REVOKE update ON pat.customers FROM terry;

ROLLBACK

Use the ROLLBACK statement to undo work done in the current transaction.

Required privilege

None

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

ROLLBACK [WORK]

Parameters

The ROLLBACK statement enables the following optional keyword:

Parameter Description

[WORK]

Optional clause supported for compliance with the SQL standard. ROLLBACK and ROLLBACK WORK are equivalent.

Description

When the PassThrough connection attribute is specified with a value greater than zero, the Oracle database transaction will also be rolled back.

A rollback closes all open cursors.

Examples

Insert a row into the regions table of the HR schema and then roll back the transaction. First set AUTOCOMMIT to 0:

Command> SET AUTOCOMMIT 0;
Command> INSERT INTO regions VALUES (5,'Australia');
1 row inserted.
Command> SELECT * FROM regions;
< 1, Europe >
< 2, Americas >
< 3, Asia >
< 4, Middle East and Africa >
< 5, Australia >
5 rows found.
Command> ROLLBACK;
Command> SELECT * FROM regions;
< 1, Europe >
< 2, Americas >
< 3, Asia >
< 4, Middle East and Africa >
4 rows found.

See also

SELECT

The SELECT statement retrieves data from one or more tables. The retrieved data is presented in the form of a table that is called the result table, result set, or query result.

Required privilege

No privilege is required for the object owner.

SELECT for another user's object.

SELECT...FOR UPDATE also requires UPDATE privilege for another user's object.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

The general syntax for a SELECT statement is the following:

[WithClause] SELECT [hint][FIRST NumRows | ROWS m TO n] [ALL | DISTINCT] SelectList
  FROM TableSpec [,...]
  [WHERE SearchCondition]
  [GROUP BY GroupByClause [,...] [HAVING SearchCondition]]
  [ORDER BY OrderByClause [,...]]
  [FOR UPDATE [OF [[Owner.]TableName.]ColumnName [,...]] 
    [NOWAIT | WAIT Seconds]]

The syntax for a SELECT statement that contains the set operators UNION, UNION ALL, MINUS, or INTERSECT is as follows:

SELECT [hint] [ROWS m TO n] [ALL] SelectList
  FROM TableSpec [,...]
    [WHERE SearchCondition]
    [GROUP BY GroupByClause [,...] [HAVING SearchCondition] [,...]]
  {UNION [ALL] | MINUS | INTERSECT}
  SELECT [ROWS m TO n] [ALL] SelectList
    FROM TableSpec [,...]
     [WHERE SearchCondition]
     [GROUP BY GroupByClause [,...] [HAVING SearchCondition [,...] ] ]
     [ORDER BY OrderByClause [,...] ]

The syntax for OrderByClause is as follows:

{ColumnID|ColumnAlias|Expression} [ASC|DESC] [NULLS { FIRST|LAST }]

Parameters

Parameter Description

[WithClause]

The WITH clause, also known as subquery factoring, enables you to assign a name to a subquery block, which can subsequently be referenced multiple times within the top-level SELECT statement. See "WithClause" for information on the syntax for the WithClause.

hint

Specifies a statement level optimizer hint for the SELECT statement. See "Statement level optimizer hints" for information on statement level optimizer hints.

FIRST NumRows

Specifies the number of rows to retrieve. NumRows must be either a positive INTEGER value or a dynamic parameter placeholder. The syntax for a dynamic parameter placeholder is either ? or :DynamicParameter. The value of the dynamic parameter is supplied when the statement is executed.

ROWS m TO n

Specifies the range of rows to retrieve where m is the first row to be selected and n is the last row to be selected. Row counting starts at row 1. The query SELECT ROWS 1 TO n returns the same rows as SELECT FIRST NumRows assuming the queries are ordered and n and NumRows have the same value.

Use either a positive INTEGER value or a dynamic parameter placeholder for m and n values. The syntax for a dynamic parameter placeholder is either ? or :DynamicParameter. The value of the dynamic parameter is supplied when the statement is executed.

ALL

Prevents elimination of duplicate rows from the query result. If neither ALL nor DISTINCT is specified, ALL is the default.

DISTINCT

Ensures that each row in the query result is unique. All NULL values are considered equal for this comparison. Duplicate rows are not evaluated.

You cannot use SELECT ... on a LOB column.

SelectList

Specifies how the columns of the query result are to be derived. See "SelectList" for the syntax for a select list.

FROM TableSpec

Identifies the tables referenced in the SELECT statement. The maximum number of tables per query is 24.

TableSpec identifies a table from which rows are selected. The table can be a derived table, which is the result of a SELECT statement in the FROM clause. See "TableSpec" for the TableSpec syntax.

WHERE SearchCondition

The WHERE clause determines the set of rows to be retrieved. Normally, rows for which SearchCondition is FALSE or NULL are excluded from processing, but SearchCondition can be used to specify an outer join in which rows from an outer table that do not have SearchCondition evaluated to TRUE with respect to any rows from the associated inner table are also returned, with projected expressions referencing the inner table set to NULL.

The unary (+) operator may follow some column and ROWID expressions to indicate an outer join. The (+) operator must follow all column and ROWID expressions in the join conditions that refer to the inner table. There are several conditions on the placement of the (+) operator. These generally restrict the type of outer join queries that can be expressed. The (+) operator may appear in WHERE clauses but not in HAVING clauses. Two tables cannot be outer joined together. An outer join condition cannot be connected by OR.

See "Search Conditions" for more information on search conditions.

GROUP BY GroupByClause [,...]

The GROUP BY clause identifies one or more expressions to be used for grouping when aggregate functions are specified in the select list and when you want to apply the function to groups of rows. See "GROUP BY clause" for information on the syntax for the GROUP BY clause.

HAVING SearchCondition

The HAVING clause can be used in a SELECT statement to filter groups of an aggregate result. The existence of a HAVING clause in a SELECT statement turns the query into an aggregate query. All columns referenced outside the sources of aggregate functions in any clause except the WHERE clause must be included in the GROUP BY clause.

Subqueries can be specified in the HAVING clause.

(+)

A simple join (also called an inner join) returns a row for each pair of rows from the joined tables that satisfy the join condition specified in SearchCondition. Outer joins are an extension of this operator in which all rows of the outer table are returned, whether or not matching rows from the joined inner table are found. In the case no matching rows are found, any projected expressions referencing the inner table are given the value NULL.

ORDER BY OrderByClause [,...]

Sorts the query result rows in order by specified columns or expressions. Specify the sort key columns in order from major sort key to minor sort key.

The ORDER BY clause supports column aliases, which can be referenced only in an ORDER BY clause. A single query may declare several column aliases with the same name, but any reference to that alias results in an error.

ColumnID

Must correspond to a column in the select list. You can identify a column to be sorted by specifying its name or its ordinal number. The first column in the select list is column number 1. It is better to use a column number when referring to columns in the select list if they are not simple columns. Some examples are aggregate functions, arithmetic expressions, and constants.

A ColumnID in the ORDER BY clause has this syntax:

{ColumnNumber |[[Owner.]TableName.] ColumnName}

ColumnAlias

Used in an ORDER BY clause, the column alias must correspond to a column in the select list. The same alias can identify multiple columns.

{* | [Owner.]TableName.* |

{Expression | [[Owner.]TableName.]ColumnName |

[[Owner.]TableName.]ROWID

}

[[AS] ColumnAlias]} [,...]

ASC|DESC

For each column designated in the ORDER BY clause, you can specify whether the sort order is to be ascending or descending. If neither ASC (ascending) nor DESC (descending) is specified, ascending order is used. All character data types are sorted according to the current value of the NLS_SORT session parameter.

NULLS { FIRST|LAST }

Valid with ORDER BY clause and is optional. If you specify ASC or DESC, NULLS FIRST or NULLS LAST must follow ASC or DESC.

Specify NULLS FIRST to have rows with NULL values returned first in your ordered query. Specify NULLS LAST to have rows with NULL values returned last in your ordered query. NULLS LAST is the default when rows are returned in ascending order. NULLS FIRST is the default when rows are returned in descending order.

If you specify the ORDER BY clause without the ASC or DESC clause and without the NULLS FIRST or NULLS LAST clause, the default ordering sequence is ascending NULLS LAST.

FOR UPDATE

[OF [[Owner.]

TableName.]

ColumnName [,...]]

[NOWAIT | WAIT Seconds]

FOR UPDATE

  • FOR UPDATE maintains a lock on a row until the end of the current transaction, regardless of isolation. All other transactions are excluded from performing any operation on that row until the transaction is committed or rolled back.

  • FOR UPDATE may be used with joins and the ORDER BY, GROUP BY, and DISTINCT clauses. Update locks are obtained on either tables or rows, depending on the table/row locking method chosen by the optimizer.

  • Rows from all tables that satisfy the WHERE clause are locked in UPDATE mode unless the FOR UPDATE OF clause is specified. This clause specifies which tables to lock.

  • If using row locks, all qualifying rows in all tables from the table list in the FROM clause are locked in update mode. Qualifying rows are those rows that satisfy the WHERE clause. When table locks are used, the table is locked in update mode whether or not there are any qualifying rows.

  • If the serializable isolation level and row locking are enabled, nonqualifying rows are downgraded to shared mode. If a read-committed isolation level and row locking are turned on, nonqualifying rows are unlocked.

  • SELECT...FOR UPDATE locks are not blocked by SELECT locks.

FOR UPDATE [OF [[Owner.]TableName.]ColumnName [,...] ]

  • This mode optionally includes the name of the column or columns in the table to be locked for update.

[NOWAIT | WAIT Seconds ]

  • This specifies how to proceed if the selected rows are locked. It does not apply to table-level locks or database-level locks.

  • NOWAIT specifies that there is no waiting period for locks. An error is returned if the lock is not available.

  • WAIT Seconds specifies the lock timeout setting.

    An error is returned if the lock is not obtained in the specified amount of time.

    The lock timeout setting is expressed in seconds or fractions of second. The data type for Seconds is NUMBER. Values between 0.0 and 1000000.0 are valid.

  • If neither NOWAIT nor WAIT is specified, the lock timeout interval for the transaction is used.

SelectQuery1

{UNION [ALL] | MINUS | INTERSECT}

SelectQuery2

Specifies that the results of SelectQuery1 and SelectQuery2 are to be combined, where SelectQuery1 and SelectQuery2 are general SELECT statements with some restrictions.

The UNION operator combines the results of two queries where the SelectList is compatible. If UNION ALL is specified, duplicate rows from both SELECT statements are retained. Otherwise, duplicates are removed.

The MINUS operator combines rows returned by the first query but not by the second into a single result.

The INTERSECT operator combines only those rows returned by both queries into a single result.

The data type of corresponding selected entries in both SELECT statements must be compatible. One type can be converted to the other type using the CAST operator. Nullability does not need to match.

The length of a column in the result is the longer length of correspondent selected values for the column. The column names of the final result are the column names of the leftmost select.

You can combine multiple queries using the set operators UNION, UNION ALL, MINUS, and INTERSECT.

One or both operands of a set operator can be a set operator. Multiple or nested set operators are evaluated from left to right.

The set operators can be mixed in the same query.

Restrictions on the SELECT statement that specify the set operators are as follows:

  • Neither SELECT statement can specify FIRST NumRows.

  • The SELECT subquery in a UNION, UNION ALL, MINUS, or INTERSECT must have the same number of projected expressions. This is true for INSERT...SELECT as well.

  • ORDER BY can be specified to sort the final result but cannot be used with any individual operand of a set operator. Only column names of tables or column alias from the leftmost SELECT statement can be specified in the ORDER BY clause.

  • GROUP BY can be used to group an individual SELECT operand of a set operator but cannot be used to group a set operator result.

  • The set operators cannot be used in materialized view or a joined table.

Description

  • When you use a correlation name, the correlation name must conform to the syntax rules for a basic name. All correlation names within one SELECT statement must be unique. Correlation names are useful when you join a table to itself. Define multiple correlation names for the table in the FROM clause and use the correlation names in the select list and the WHERE clause to qualify columns from that table. See "TableSpec" for more information about correlation names.

  • SELECT...FOR UPDATE is supported in a SELECT statement that specifies a subquery, but it can be specified only in the outermost query.

  • If your query specifies either FIRST NumRows or ROWS m TO n, ROWNUM may not be used to restrict the number of rows returned.

  • FIRST NumRows and ROWS m TO n cannot be used together in the same SELECT statement.

  • Use the SELECT...INTO statement in PL/SQL. If you use the SELECT...INTO statement outside of PL/SQL, TimesTen accepts, but silently ignores, the syntax.

Examples

This example shows the use of a column alias (max_salary) in the SELECT statement:

SELECT MAX(salary) AS max_salary 
FROM employees 
WHERE employees.hire_date > '2000-01-01 00:00:00';
< 10500 >
1 row found.

This example uses two tables, orders and lineitems.

The orders table and lineitems table are created as follows:

CREATE TABLE orders(orderno INTEGER, orderdate DATE, customer CHAR(20));

CREATE TABLE lineitems(orderno INTEGER, lineno INTEGER, 
  qty INTEGER, unitprice DECIMAL(10,2));

Thus for each order, there is one record in the orders table and a record for each line of the order in lineitems.

To find the total value of all orders entered since the beginning of the year, use the HAVING clause to select only those orders that were entered on or after January 1, 2000:

SELECT o.orderno, customer, orderdate, SUM(qty * unitprice)
FROM orders o, lineitems l
WHERE o.orderno=l.orderno
GROUP BY o.orderno, customer, orderdate
HAVING orderdate >= DATE '2000-01-01';

Consider this query:

SELECT * FROM tablea, tableb
WHERE tablea.column1 = tableb.column1 AND tableb.column2 > 5
FOR UPDATE;

The query locks all rows in tablea where:

  • The value of tablea.column1 equals at least one tableb.column1 value where tableb.column2 is greater than 5.

The query also locks all rows in tableb where:

  • The value of tableb.column2 is greater than 5.

  • The value of tableb.column1 equals at least one tablea.column1 value.

If no WHERE clause is specified, all rows in both tables are locked.

This example demonstrates the (+) join operator:

SELECT * FROM t1, t2
WHERE t1.x = t2.x(+);

The following query returns an error because an outer join condition cannot be connected by OR.

SELECT * FROM t1, t2, t3
WHERE t1.x = t2.x(+) OR t3.y = 5;

The following query is valid:

SELECT * FROM t1, t2, t3
WHERE t1.x = t2.x(+) AND (t3.y = 4 OR t3.y = 5);

A condition cannot use the IN operator to compare a column marked with (+). For example, the following query returns an error.

SELECT * FROM t1, t2, t3
WHERE t1.x = t2.x(+) AND t2.y(+) IN (4,5);

The following query is valid:

SELECT * FROM t1, t2, t3
WHERE t1.x = t2.x(+) AND t1.y IN (4,5);

The following query results in an inner join. The condition without the (+) operator is treated as an inner join condition.

SELECT * FROM t1, t2
WHERE t1.x = t2.x(+) AND t1.y = t2.y;

In the following query, the WHERE clause contains a condition that compares an inner table column of an outer join with a constant. The (+) operator is not specified and hence the condition is treated as an inner join condition.

SELECT * FROM t1, t2
WHERE t1.x = t2.x(+) AND t2.y = 3;

For more join examples, see "JoinedTable".

The following example returns the current sequence value in the student table.

SELECT SEQ.CURRVAL FROM student;

The following query produces a derived table because it contains a SELECT statement in the FROM clause.

SELECT * FROM t1, (SELECT MAX(x2) maxx2 FROM t2) tab2 
WHERE t1.x1 = tab2.maxx2;

The following query joins the results of two SELECT statements.

SELECT * FROM t1 
WHERE x1 IN (SELECT x2 FROM t2) 
UNION 
SELECT * FROM t1 
WHERE x1 IN (SELECT x3 FROM t3);

In the following, select all orders that have the same price as the highest price in their category.

SELECT * FROM orders WHERE price = (SELECT MAX(price) 
FROM stock WHERE stock.cat=orders.cat);

The next example illustrates the use of the INTERSECT set operator. There is a department_id value in the employees table that is NULL. In the departments table, the department_id is defined as a NOT NULL primary key. The rows returned from using the INTERSECT set operator do not include the row in the departments table whose department_id value is NULL.

Command> SELECT department_id FROM employees INTERSECT SELECT department_id 
         FROM departments;
< 10 >
< 20 >
< 30 >
< 40 >
< 50 >
< 60 >
< 70 >
< 80 >
< 90 >
< 100 >
< 110 >
11 rows found.
Command> SELECT DISTINCT department_id FROM employees;
< 10 >
< 20 >
< 30 >
< 40 >
< 50 >
< 60 >
< 70 >
< 80 >
< 90 >
< 100 >
< 110 >
< <NULL> >
12 rows found.

The next example illustrates the use of the MINUS set operator by combining rows returned by the first query but not the second. The row containing the NULL department_id value in the employees table is the only row returned.

Command> SELECT department_id FROM employees 
         MINUS SELECT department_id FROM departments;
< <NULL> >
1 row found.

The following example illustrates the use of the SUBSTR expression in a GROUP BY clause and the use of a subquery in a HAVING clause. The first 10 rows are returned.

Command> SELECT ROWS 1 TO 10 SUBSTR (job_id, 4,10), department_id, manager_id, 
         SUM (salary) FROM employees
         GROUP BY SUBSTR (job_id,4,10),department_id, manager_id
         HAVING (department_id, manager_id) IN
          (SELECT department_id, manager_id FROM employees x
           WHERE x.department_id = employees.department_id)
         ORDER BY SUBSTR (job_id, 4,10),department_id,manager_id;
< ACCOUNT, 100, 108, 39600 >
< ACCOUNT, 110, 205, 8300 >
< ASST, 10, 101, 4400 >
< CLERK, 30, 114, 13900 >
< CLERK, 50, 120, 22100 >
< CLERK, 50, 121, 25400 >
< CLERK, 50, 122, 23600 >
< CLERK, 50, 123, 25900 >
< CLERK, 50, 124, 23000 >
< MAN, 20, 100, 13000 >
10 rows found.

The following example locks the employees table for update and waits 10 seconds for the lock to be available. An error is returned if the lock is not acquired in 10 seconds. The first five rows are selected.

Command> SELECT FIRST 5 last_name FROM employees FOR UPDATE WAIT 10;
< King >
< Kochhar >
< De Haan >
< Hunold >
< Ernst >
5 rows found.

The next example locks the departments table for update. If the selected rows are locked by another process, an error is returned if the lock is not available. This is because NOWAIT is specified.

Command> SELECT FIRST 5 last_name e FROM employees e, departments d 
         WHERE e.department_id = d.department_id 
         FOR UPDATE OF d.department_id NOWAIT;
< Whalen >
< Hartstein >
< Fay >
< Raphaely >
< Khoo >
5 rows found.

In the following, use the HR schema to illustrate the use of a subquery with the FOR UPDATE clause.

Command> SELECT employee_id, job_id FROM job_history 
         WHERE (employee_id, job_id) NOT IN (SELECT employee_id, job_id 
         FROM employees) 
         FOR UPDATE;
< 101, AC_ACCOUNT >
< 101, AC_MGR >
< 102, IT_PROG >
< 114, ST_CLERK >
< 122, ST_CLERK >
< 176, SA_MAN >
< 200, AC_ACCOUNT >
< 201, MK_REP >
8 rows found.

In the following, use a dynamic parameter placeholder for SELECT ROWS m TO n and SELECT FIRST.

Command> SELECT ROWS ? TO ? employee_id FROM employees;

Type '?' for help on entering parameter values.
Type '*' to end prompting and abort the command.
Type '-' to leave the parameter unbound.
Type '/;' to leave the remaining parameters unbound and execute the command.

Enter Parameter 1 (TT_INTEGER) > 1
Enter Parameter 2 (TT_INTEGER) > 3
< 100 >
< 101 >
< 102 >
3 rows found.
Command> SELECT ROWS :a TO :b employee_id FROM employees;

Type '?' for help on entering parameter values.
Type '*' to end prompting and abort the command.
Type '-' to leave the parameter unbound.
Type '/;' to leave the remaining parameters unbound and execute the command.

Enter Parameter 1 (TT_INTEGER) > 1
Enter Parameter 2 (TT_INTEGER) > 3
< 100 >
< 101 >
< 102 >
3 rows found.
Command> SELECT FIRST ? employee_id FROM employees;

Type '?' for help on entering parameter values.
Type '*' to end prompting and abort the command.
Type '-' to leave the parameter unbound.
Type '/;' to leave the remaining parameters unbound and execute the command.

Enter Parameter 1 (TT_INTEGER) > 3
< 100 >
< 101 >
< 102 >
3 rows found.

The following example illustrates the use of NULLS LAST in the ORDER BY clause. Query the employees table to find employees with a commission percentage greater than .30 or a commission percentage that is NULL. Select the first seven employees and order by commission_pct and last_name. Order commision_pct in descending order and use NULLS LAST to display rows with NULL values last in the query. Output commission_pct and last_name.

Command> SELECT FIRST 7 commission_pct,last_name
         FROM employees where commission_pct > .30
         OR commission_pct IS NULL
         ORDER BY commission_pct DESC NULLS LAST,last_name;
< .4, Russell >
< .35, King >
< .35, McEwen >
< .35, Sully >
< <NULL>, Atkinson >
< <NULL>, Austin >
< <NULL>, Baer >
7 rows found.

WithClause

Syntax

WithClause has the following syntax:

WITH QueryName AS ( Subquery ) [, QueryName AS ( Subquery )] ...

Parameters

WithClause has the following parameter:

Parameter Description

QueryName AS (Subquery)

Specifies an alias for a subquery that can be used multiple times within the SELECT statement.

Description

Subquery factoring provides the WITH clause that enables you to assign a name to a subquery block, which can subsequently be referenced multiple times within the main SELECT query. The query name is visible to the main query and any subquery contained in the main query.

The WITH clause can only be defined as a prefix to the main SELECT statement.

Subquery factoring is useful in simplifying complex queries that use duplicate or complex subquery blocks in one or more places. In addition, TimesTen uses subquery factoring to optimize the query by evaluating and materializing the subquery block once and providing the result for each reference in the SELECT statement.

You can specify the set operators: UNION, MINUS, INTERSECT in the main query.

Restrictions using the WITH clause:

  • Do not use the WITH clause in a view or materialized view definition.

  • Recursive subquery factoring is not supported.

  • Do not use the WITH clause in subqueries or derived tables.

  • You cannot provide a column parameter list for the query alias. For example, TimesTen does not support: WITH w1(c1,c2) AS ...

Example

The following example creates the query names dept_costs and avg_cost for the initial query block, then uses these names in the body of the main query.

Command> WITH dept_costs AS (
         SELECT department_name, SUM(salary) dept_total
         FROM employees e, departments d
         WHERE e.department_id = d.department_id
         GROUP BY department_name),
          avg_cost AS (
         SELECT SUM(dept_total)/COUNT(*) avg
         FROM dept_costs)
         SELECT * FROM dept_costs
         WHERE dept_total >
         (SELECT avg FROM avg_cost)
         ORDER BY department_name;

> DEPARTMENT_NAME DEPT_TOTAL
-------------------------------
Sales 304500
Shipping 156400

SelectList

SQL syntax

The SelectList parameter of the SELECT statement has the following syntax:

{* | [Owner.]TableName.* | 
  { Expression | [[Owner.]TableName.]ColumnName |
    [[Owner.]TableName.]ROWID | NULL
  }
  [[AS] ColumnAlias] } [,...]

Parameters

The SelectList parameter of the SELECT statement has the following parameters:

Parameter Description

*

Includes, as columns of the query result, all columns of all tables specified in the FROM clause.

[Owner.]TableName.*

Includes all columns of the specified table in the result.

Expression

An aggregate query includes a GROUP BY clause or an aggregate function.

When the select list is not an aggregate query, the column reference must reference a table in the FROM clause.

A column reference in the select list of an aggregate query must reference a column list in the GROUP BY clause. If there is no GROUP BY clause, then the column reference must reference a table in the FROM clause. See "GROUP BY clause" for more information on the GROUP BY clause.

[[Owner.]Table.] ColumnName

Includes a particular column from the named owner's indicated table. You can also specify the CURRVAL or NEXTVAL column of a sequence. See "Using CURRVAL and NEXTVAL in TimesTen Classic" for more details.

[[Owner.]Table.] ROWID

Includes the ROWID pseudocolumn from the named owner's indicated table.

NULL

When NULL is specified, the default for the resulting data type is VARCHAR(0). You can use the CAST function to convert the result to a different data type. NULL can be specified in the ORDER BY clause.

ColumnAlias

Used in an ORDER BY clause, the column alias must correspond to a column in the select list. The same alias can identify multiple columns.

{*|[Owner.]TableName.*|

{Expression |[[Owner.]TableName.]ColumnName |

[[Owner.]TableName.]ROWID

}

[[AS] ColumnAlias]} [,...]

Description

  • The clauses must be specified in the order given in the syntax.

  • TimesTen does not support subqueries in the select list.

  • A result column in the select list can be derived in any of the following ways.

    • A result column can be taken directly from one of the tables listed in the FROM clause.

    • Values in a result column can be computed, using an arithmetic expression, from values in a specified column of a table listed in the FROM clause.

    • Values in several columns of a single table can be combined in an arithmetic expression to produce the result column values.

    • Aggregate functions (AVG, MAX, MIN, SUM, and COUNT) can be used to compute result column values over groups of rows. Aggregate functions can be used alone or in an expression. You can specify aggregate functions containing the DISTINCT qualifier that operate on different columns in the same table. If the GROUP BY clause is not specified, the function is applied over all rows that satisfy the query. If the GROUP BY clause is specified, the function is applied once for each group defined by the GROUP BY clause. When you use aggregate functions with the GROUP BY clause, the select list can contain aggregate functions, arithmetic expressions, and columns in the GROUP BY clause. See "GROUP BY clause" for details on the GROUP BY clause.

    • A result column containing a fixed value can be created by specifying a constant or an expression involving only constants.

  • In addition to specifying how the result columns are derived, the select list also controls their relative position from left to right in the query result. The first result column specified by the select list becomes the leftmost column in the query result, and so on.

  • Result columns in the select list are numbered from left to right. The leftmost column is number 1. Result columns can be referred to by column number in the ORDER BY clause. This is especially useful to refer to a column defined by an arithmetic expression or an aggregate.

  • To join a table with itself, define multiple correlation names for the table in the FROM clause and use the correlation names in the select list and the WHERE clause to qualify columns from that table.

  • When you use the GROUP BY clause, one answer is returned per group in accordance with the select list, as follows:

    • The WHERE clause eliminates rows before groups are formed.

    • The GROUP BY clause groups the resulting rows. See "GROUP BY clause" for more details.

    • The select list aggregate functions are computed for each group.

Examples

In the following example, one value, the average number of days you wait for a part, is returned:

SELECT AVG(deliverydays)
FROM purchasing.supplyprice;

The part number and delivery time for all parts that take fewer than 20 days to deliver are returned by the following statement.

SELECT partnumber, deliverydays
FROM purchasing.supplyprice
WHERE deliverydays < 20;

Multiple rows may be returned for a single part.

The part number and average price of each part are returned by the following statement.

SELECT partnumber, AVG(unitprice)
FROM purchasing.supplyprice
GROUP BY partnumber;

In the following example, the join returns names and locations of California suppliers. Rows are returned in ascending order by partnumber values. Rows containing duplicate part numbers are returned in ascending order by vendorname values. The FROM clause defines two correlation names (v and s), which are used in both the select list and the WHERE clause. The vendornumber column is the only common column between vendors and supplyprice.

SELECT partnumber, vendorname, s.vendornumber,vendorcity
  FROM purchasing.supplyprice s, purchasing.vendors v
  WHERE s.vendornumber = v.vendornumber AND vendorstate = 'CA'
ORDER BY partnumber, vendorname;

The following query joins table purchasing.parts to itself to determine which parts have the same sales price as the part whose serial number is '1133-P-01'.

SELECT q.partnumber, q.salesprice
  FROM purchasing.parts p, purchasing.parts q
  WHERE p.salesprice = q.salesprice AND p.serialnumber = '1133-P-01';

The next example shows how to retrieve the rowid of a specific row. The retrieved rowid value can be used later for another SELECT, DELETE, or UPDATE statement.

SELECT rowid
FROM purchasing.vendors
WHERE vendornumber = 123;

The following example shows how to use a column alias to retrieve data from the table employees.

SELECT MAX(salary) AS max_salary FROM employees;

TableSpec

SQL syntax

The TableSpec parameter of the SELECT statement has the following syntax:

TableNameSyntax | JoinedTable | DerivedTable

TableNameSyntax::=  [Owner.]TableName  [CorrelationName] |
                   ([Owner.]TableName) [CorrelationName] |
                   ([Owner.]TableName  [CorrelationName])

A simple table specification has the following syntax:

[Owner.]TableName or ([Owner.]TableName)

Parameters

The TableSpec parameter of the SELECT statement has the following parameters:

Parameter Description

TableNameSyntax

Identifies a table to be referenced. Parentheses are optional.

CorrelationName

CorrelationName specifies an alias for the immediately preceding table. When accessing columns of that table elsewhere in the SELECT statement, use the correlation name instead of the actual table name within the statement. The scope of the correlation name is the SQL statement in which it is used. The correlation name must conform to the syntax rules for a basic name. See "Basic names" for more information.

All correlation names within one statement must be unique.

JoinedTable

Specifies the query that defines the table join. See "JoinedTable" for more information.

DerivedTable

Specifies a table derived from the evaluation of a SELECT statement. No FIRST NumRows or ROWS m TO n clauses are allowed in this SELECT statement. See "DerivedTable" for more information.

JoinedTable

The JoinedTable parameter specifies a table derived from CROSS JOIN, INNER JOIN, LEFT OUTER JOIN or RIGHT OUTER JOIN.

SQL syntax

The syntax for JoinedTable is as follows:

{CrossJoin | QualifiedJoin}

Where CrossJoin is:

TableSpec1 CROSS JOIN TableSpec2

And QualifiedJoin is:

TableSpec1 [JoinType] JOIN TableSpec2 ON SearchCondition

In the QualifiedJoin parameter, JoinType syntax is as follows:

{INNER | LEFT [OUTER] | RIGHT [OUTER]}

Parameters

The JoinedTable parameter of the TableSpec clause of a SELECT statement has the following parameters:

Parameter Description

CrossJoin

Performs a cross join on two tables. A cross join returns a result table that is the cartesian product of the input tables. The result is the same as that of a query with the following syntax:

SELECT Selectlist FROM Table1, Table2

QualifiedJoin

Specifies that the join is of type JoinType.

TableSpec1

Specifies the first table of the JOIN clause.

TableSpec2

Specifies the second table of the JOIN clause.

JoinType JOIN

Specifies the type of join to perform. These are the supported join types:

  • INNER

  • LEFT [OUTER]

  • RIGHT [OUTER]

INNER JOIN returns a result table that combines the rows from two tables that meet SearchCondition.

LEFT OUTER JOIN returns join rows that match SearchCondition and rows from the first table that do not have SearchCondition evaluated as true with any row from the second table.

RIGHT OUTER JOIN returns join rows that match SearchCondition and rows from the second table that do not have SearchCondition evaluated as true with any row from the first table.

ON SearchCondition

Specifies the search criteria to be used in a JOIN parameter. SearchCondition can refer only to tables referenced in the current qualified join.

Description

  • FULL OUTER JOIN is not supported.

  • A joined table can be used to replace a table in a FROM clause anywhere except in a statement that defines a materialized view. Thus, a joined table can be used in UNION, MINUS, INTERSECT, a subquery, a nonmaterialized view, or a derived table.

  • A subquery cannot be specified in the operand of a joined table. For example, the following statement is not supported:

    SELECT * FROM
      regions INNER JOIN (SELECT * FROM countries) table2
      ON regions.region_id=table2.region_id;
    
  • A view can be specified as an operand of a joined table.

  • A temporary table cannot be specified as an operand of a joined table.

  • OUTER JOIN can be specified in two ways, either using the (+) operator in SearchCondition of the WHERE clause or using a JOIN table operation. The two specification methods cannot coexist in the same statement.

  • Join order and grouping can be specified with a JoinedTable operation, but they cannot be specified with the (+) operator. For example, the following operation cannot be specified with the (+) operator:

    t LEFT JOIN (t2 INNER JOIN t3 ON x2=x3) ON (x1 = x2 - x3)

Examples

These examples use the regions and countries tables from the HR schema.

The following performs a left outer join.

SELECT * FROM regions LEFT JOIN countries
  ON regions.region_id=countries.region_id
  WHERE regions.region_id=3;

< 3, Asia, JP, Japan, 3 >
< 3, Asia, CN, China, 3 >
< 3, Asia, IN, India, 3 >
< 3, Asia, AU, Australia, 3 >
< 3, Asia, SG, Singapore, 3 >
< 3, Asia, HK, HongKong, 3 >
6 rows found.

You can also perform a left outer join with the (+) operator, as follows.

SELECT * FROM regions, countries
  WHERE regions.region_id=countries.region_id (+)
    AND regions.region_id=3;

The following performs a right outer join.

SELECT * FROM regions RIGHT JOIN countries
  ON regions.region_id=wountries.region_id
  WHERE regions.region_id=3;

< AU, Australia, 3, 3, Asia >
< CN, China, 3, 3, Asia >
< HK, HongKong, 3, 3, Asia >
< IN, India, 3, 3, Asia >
< JP, Japan, 3, 3, Asia >
< SG, Singapore, 3, 3, Asia >
6 rows found.

The next example performs a right outer join with the (+) operator.

SELECT * FROM countries, regions
       WHERE regions.region_id (+)=countries.region_id
       AND countries.region_id=3;
< JP, Japan, 3, 3, Asia >
< CN, China, 3, 3, Asia >
< IN, India, 3, 3, Asia >
< AU, Australia, 3, 3, Asia >
< SG, Singapore, 3, 3, Asia >
< HK, HongKong, 3, 3, Asia >
6 rows found.

Note that the right join methods produce the same rows but in a different display order. There should be no expectation of row order for join results.

The following performs an inner join.

SELECT * FROM regions INNER JOIN countries
  ON regions.region_id=countries.region_id
  WHERE regions.region_id=2;

< 2, Americas, US, United States of America, 2 >
< 2, Americas, CA, Canada, 2 >
< 2, Americas, BR, Brazil, 2 >
< 2, Americas, MX, Mexico, 2 >
< 2, Americas, AR, Argentina, 2 >
5 rows found.

The next example performs a cross join.

SELECT * FROM regions CROSS JOIN countries 
  WHERE regions.region_id=1;

< 1, Europe, AR, Argentina, 2 >
< 1, Europe, AU, Australia, 3 >
< 1, Europe, BE, Belgium, 1 >
< 1, Europe, BR, Brazil, 2 >
...
< 1, Europe, SG, Singapore, 3 >
< 1, Europe, UK, United Kingdom, 1 >
< 1, Europe, US, United States of America, 2 >
< 1, Europe, ZM, Zambia, 4 >
< 1, Europe, ZW, Zimbabwe, 4 >
25 rows found.

DerivedTable

A derived table is the result of a SELECT statement in the FROM clause, with an alias.

SQL syntax

The syntax for DerivedTable is as follows:

(Subquery) [CorrelationName]

Parameters

The DerivedTable parameter of the TableSpec clause of a SELECT statement has the following parameters:

Parameter Description

Subquery

See "Subqueries" for information on subqueries.

CorrelationName

Optionally use CorrelationName to specify an alias for the derived table. It must be different from any table name referenced in the query.

Description

When using a derived table, these restrictions apply:

  • The DUAL table can be used in a SELECT statement that references no other tables, but needs to return at least one row. Selecting from DUAL is useful for computing a constant expression (an expression that is evaluated to a constant value) with the SELECT statement. Because DUAL has only one row, the constant is returned only once.

  • Subquery cannot refer to a column from another derived table.

  • A derived table cannot be used as a source of a joined table.

  • A derived table cannot be used as a target of a DELETE or UPDATE statement.

GROUP BY clause

Specify the GROUP BY clause if you want the database to group the selected rows based on the value of expressions for each row and return a single row of summary information for each group. If the GROUP BY clause is omitted, the entire query result is treated as one group. If this clause contains CUBE or ROLLUP, the results contain superaggregate groupings in addition to the regular groupings.

The expressions in the GROUP BY clause can do the following:

  • Designate single or multiple columns.

  • Include arithmetic operations, the ROWID pseudocolumn, or NULL.

  • Include a date, a constant, or a dynamic parameter.

  • Include ROLLUP or CUBE clauses, where the results produce superaggregate groupings in addition to the regular groupings. Superaggregate groupings are calculated subtotals and totals returned with the regular groupings in the GROUP BY clause.

  • Include GROUPING SETS clause to distinguish which superaggregate groupings to produce.

When you use the GROUP BY clause, the select list can contain only aggregate functions and columns referenced in the GROUP BY clause. If the select list contains the construct *, TableName.*, or Owner.TableName.*, the GROUP BY clause must contain all columns that the * includes. NULL values are considered equivalent in grouping rows. If all other columns are equal, all NULL values in a column are placed in a single group.

Note:

To identify and potentially eliminate NULL groupings from the superaggregate groupings, use the GROUPING function. See "GROUPING" for information.

SQL syntax

The general syntax for the GROUP BY clause is the following:

GROUP BY
 {Expression | RollupCubeClause | GroupingSetsClause }[,...]
 
GroupingSetsClause::=  GROUPING SETS 
 GroupingExpressionList | RollupCubeClause [,...]

RollupCubeClause
{ ROLLUP | CUBE } ( GroupingExpressionList ) }

GroupingExpressionList::=   
{ Expression | ExpressionList [, { Expression | ExpressionList } ] ...}
 
ExpressionList :: = ( Expression [, Expression ] ...)
Parameters
Parameter Description

Expression

Valid expression syntax. See Expressions for more information.

RollupCubeClause

The GROUP BY clause may include one or more ROLLUP or CUBE clauses.

GroupingSetsClause

The GROUP BY clause may include one or more GROUPING SETS clauses. The GROUPING SETS clause enables you to explicitly specify which groupings of data that the database returns. See "GROUPING SETS" for more information.

GroupingExpressionList

The GROUP BY clause can contain multiple expressions or expression lists.

ROLLUP GroupingExpressionList

The ROLLUP clause is used to generate super aggregate rows from groups. See "ROLLUP" for more information.

CUBE GroupingExpressionList

The CUBE clause groups selected rows based on the values of all possible combinations of the grouping columns in the CUBE clause. See "CUBE" for more information.

ExpressionList

A list of one or more expressions, each separated by a comma.

Examples

The following GROUP BY example sums the salaries for employees in the employees table and uses the SUBSTR expression to group the data by job function.

Command> SELECT SUBSTR (job_id, 4,10), SUM (salary) FROM employees
         GROUP BY SUBSTR (job_id,4,10);
< PRES, 24000 >
< VP, 34000 >
< PROG, 28800 >
< MGR, 24000 >
< ACCOUNT, 47900 >
< MAN, 121400 >
< CLERK, 133900 >
< REP, 273000 >
< ASST, 4400 >
9 rows found.

Query emp_details_view to select the first 10 departments and managers within the department and count the number of employees in the department with the same manager. Use the GROUP BY clause to group the result by department and manager.

Command> columnlabels on;
Command> SELECT first 10 department_id AS DEPT, manager_id AS MGR,
         COUNT(employee_id) AS NUM_EMP
         FROM emp_details_view
         GROUP BY (department_id, manager_id)
         ORDER BY department_id, manager_id;

DEPT, MGR, NUM_EMP
< 10, 101, 1 >
< 20, 100, 1 >
< 20, 201, 1 >
< 30, 100, 1 >
< 30, 114, 5 >
< 40, 101, 1 >
< 50, 100, 5 >
< 50, 120, 8 >
< 50, 121, 8 >
< 50, 122, 8 >
10 rows found.
ROLLUP, CUBE and GROUPING SETS clauses

The following definitions describe how columns can be grouped within the ROLLUP, CUBE, and GROUPING SETS clauses:

  • Grouping column: A single column used in a GROUP BY clause. For example, in the following GROUP BY clause, X, Y, and Z are group columns.

    GROUP BY X, GROUPING SETS(Y, Z)
    
  • Composite Column: A list of grouping columns inside parentheses. For example, in the following clause, (C1, C2) and (C3, C4) are composite columns.

    GROUP BY ROLLUP( (C1,C2), (C3,C4), C5);
    
  • Grouping: Grouping is a single level of aggregation from within a grouping set. For example, in the following statement, (C1) and (C2, C3) are individual groupings.

    GROUP BY GROUPING SETS(C1, (C2,C3));
    
  • Grouping Set: A collection of groupings inside parentheses. For example, in the following statement, (C1, (C2, C3)) and (C2, (C4, C5)) are two individual grouping sets.

    GROUP BY GROUPING SETS(C1, (C2,C3)), GROUPING SETS(C2, (C4, C5));
    
  • Concatenated grouping sets: Separate multiple grouping sets with commas. The result is a cross-product of groupings from each grouping set.

  • Grand Total or Empty set column: A grand total or empty set grouping computes aggregation by considering all rows as one group. Grand totals are automatically provided in the results for ROLLUP and CUBE clauses; however, you request the grand total in the GROUPING SETS clause by providing empty parentheses, ( ).

Duplicate grouping columns can be used in ROLLUP, CUBE or GROUPING SETS. However, it does result in duplicated result rows.

The ROLLUP, CUBE and GROUPING SETS clauses are not supported in a materialized view definition.

The following sections describe the GROUPING SETS, ROLLUP, and CUBE clauses:

GROUPING SETS

The GROUPING SETS clause enables you to explicitly specify which groupings of data that the database returns. You specify only the desired groups by enclosing them within parentheses, so the database only generates the superaggregate summaries in which you are interested.

The following statement produces three groups: one group returns results for each gender and year columns, a second for a summary superaggregate for each of the months and the last result for the grand total.

SELECT GENDER, YEAR, MONTH,
   SUM (NUM_OF_STUDENTS) AS TOTAL
   FROM INSTRUCTOR_SUMMARY
   GROUP BY GROUPING SETS ((GENDER, YEAR),  -- 1ST GROUP
                           (MONTH), -- 2ND GROUP
                           ()); -- 3RD GROUP
 

You can combine multiple GROUPING SETS to generate specific combinations between the multiple GROUPING SETS. The following statement contains two GROUPING SETS clauses:

GROUP BY GROUPING SETS (YEAR, MONTH),
         GROUPING SETS (WEEK, DAY);

This is equivalent to the following GROUPING SETS statement:

GROUP BY GROUPING SETS (YEAR, WEEK), 
 (YEAR, DAY), 
 (MONTH, WEEK), 
 (MONTH, DAY);

When a GROUP BY clause has both regular grouping columns and a GROUPING SETS clause, the results are grouped by the regular grouping column as follows:

GROUP BY a, b GROUPING SETS(c, d);

This is equivalent to the following:

GROUP BY GROUPING SETS((a, b, c), (a, b, d));

The following example specifies the grouping sets of (region_name, country_name), state_province, and grand totals.

Command> SELECT region_name AS Region,
           country_name AS Country,
           state_province AS State,
            COUNT(employee_id) AS "Total Emp"
         FROM regions r, countries c, locations l, departments d, employees e
         WHERE r.region_id = c.region_id AND
           l.country_id = c.country_id AND
           d.location_id = l.location_id AND
           d.department_id = e.department_id
         GROUP BY grouping sets((region_name, country_name), state_province, ())
         ORDER BY region_name, state_province;
 
REGION, COUNTRY, STATE, TOTAL EMP
< Americas, Canada, <NULL>, 2 >
< Americas, United States of America, <NULL>, 68 >
< Europe, Germany, <NULL>, 1 >
< Europe, United Kingdom, <NULL>, 35 >
< <NULL>, <NULL>, Bavaria, 1 >
< <NULL>, <NULL>, California, 45 >
< <NULL>, <NULL>, Ontario, 2 >
< <NULL>, <NULL>, Oxford, 34 >
< <NULL>, <NULL>, Texas, 5 >
< <NULL>, <NULL>, Washington, 18 >
< <NULL>, <NULL>, <NULL>, 106 >
< <NULL>, <NULL>, <NULL>, 1 >
12 rows found.

ROLLUP

ROLLUP is used within the GROUP BY clause. When used with SUM, ROLLUP generates subtotals from most detailed level (all columns specified in the ROLLUP clause) to the grand total level, by removing one column at each level. These are known as superaggregate rows.

The ROLLUP clause returns the following:

  • Regular aggregate rows that would be produced by GROUP BY without using ROLLUP.

  • Subtotals following the grouping list specified in the ROLLUP clause. ROLLUP takes as its argument an ordered list of grouping columns. Each subtotal is created for the ordered list of grouping columns dropping the right-most grouping column until it reaches the grand total. For instance, if you specify GROUP BY ROLLUP(x, y, z), the returned superaggregate groups would be as follows: (x,y,z), (x,y), (x), ( ).

    The number of subtotals created is n+1 aggregate levels, where n is the number of grouping columns. For example, if there are three expressions (n=3) in the ROLLUP clause, then n+1 = 3+1, resulting in four groupings.

  • Grand total row.

You can group columns using composite columns inside parentheses. For example, in the following statement:

GROUP BY ROLLUP( (a, b), (c, d), e);

The (a, b) and (c, d) composite columns are treated as a unit when the database produces the ROLLUP results. In this example, the grouping sets returned are as follows: ((a, b), (c, d), e ), ((a, b), (c, d)), (a, b) and ().

You can execute several ROLLUP clauses within your SELECT statement, as follows:

SELECT C1, COUNT(*)
FROM T
GROUP BY ROLLUP(a, b), ROLLUP(c, d);

This is equivalent to the following statement:

SELECT C1, COUNT(*)
FROM T
GROUP BY GROUPING SETS((a, b),(a),()), 
 GROUPING SETS((c, d),(c), ());

This example queries the employees table to select the first 10 departments and return the number of employees under each manager in each department. Use ROLLUP to subtotal the number of employees in each department and return a grand total of all employees in the company.

Command> SELECT first 10 department_id AS Dept,
          manager_id AS Mgr,
         COUNT(employee_id) AS "Total emp"
         FROM employees
         GROUP BY ROLLUP(department_id, manager_id)
         ORDER BY department_id, manager_id;
 
DEPT, MGR, TOTAL EMP
< 10, 101, 1 >
< 10, <NULL>, 1 >
< 20, 100, 1 >
< 20, 201, 1 >
< 20, <NULL>, 2 >
< 30, 100, 1 >
< 30, 114, 5 >
< 30, <NULL>, 6 >
< 40, 101, 1 >
< 40, <NULL>, 1 >
10 rows found.

The following query returns the number of employees in each region, country and state or province. The rollup returns superaggregate rows for subtotals of all employees in each state or province and in each country and a grand total for all employees in the company. By combining the region and country as its own unit (within parentheses), the rollup does not return all employees for each region.

Command> SELECT region_name AS Region,
          country_name AS Country,
          state_province AS State,
          COUNT(employee_id) AS "Total Emp"
         FROM regions r, countries c, locations l, departments d, employees e
         WHERE r.region_id = c.region_id
         AND l.country_id = c.country_id
         AND d.location_id = l.location_id
         AND d.department_id = e.department_id
         GROUP BY rollup((region_name, country_name), state_province)
         ORDER BY region_name;

REGION, COUNTRY, STATE, TOTAL EMP
< Americas, Canada, Ontario, 2 > 
< Americas, United States of America, Texas, 5 > 
< Americas, United States of America, California, 45 > 
< Americas, United States of America, Washington, 18 > 
< Americas, Canada, <NULL>, 2 > 
< Americas, United States of America, <NULL>, 68 > 
< Europe, Germany, Bavaria, 1 > 
< Europe, United Kingdom, <NULL>, 1 > 
< Europe, United Kingdom, Oxford, 34 > 
< Europe, Germany, <NULL>, 1 > 
< Europe, United Kingdom, <NULL>, 35 > 
< <NULL>, <NULL>, <NULL>, 106 > 
12 rows found. 

CUBE

The CUBE clause groups the selected rows based on the values of all possible combinations of the grouping columns in the CUBE clause. It returns a single row of summary information for each group. For example, if there are three expressions (n=3) in the CUBE clause, then 2n = 23, resulting in eight groupings. Rows grouped on the values of n expressions are called regular rows; all others are called superaggregate rows. You can group using composite columns. For example, a commonly requested CUBE operation is for state sales subtotals on all combinations of month, state, and product sold.

If you specify GROUP BY CUBE(a, b, c), the resulting aggregate groupings generated are as follows: (a,b,c), (a,b), (a,c), (b,c), a, b, c, ( ).

This example returns the number of employees for each region and country, issue the following query.

Command> SELECT region_name AS Region,
          country_name AS Country,
          COUNT(employee_id) AS "Total Emp"
       FROM regions r, countries c, locations l, departments d, employees e
       WHERE r.region_id = c.region_id
       AND l.country_id = c.country_id
       AND d.location_id = l.location_id
       AND d.department_id = e.department_id
       GROUP BY CUBE(region_name, country_name)
       ORDER BY region_name;
 
REGION, COUNTRY, TOTAL EMP
< Americas, Canada, 2 >
< Americas, United States of America, 68 >
< Americas, <NULL>, 70 >
< Europe, Germany, 1 >
< Europe, United Kingdom, 35 >
< Europe, <NULL>, 36 >
< <NULL>, Canada, 2 >
< <NULL>, Germany, 1 >
< <NULL>, United Kingdom, 35 >
< <NULL>, United States of America, 68 >
< <NULL>, <NULL>, 106 >
11 rows found.

TRUNCATE TABLE

The TRUNCATE TABLE statement is similar to a DELETE statement that deletes all rows.

In TimesTen Classic, the TRUNCATE operation is faster than the DELETE operation in most circumstances, as DELETE removes each row individually.

In TimesTen Scaleout, TRUNCATE TABLE is similar to a DDL statement that invalidates all commands that depend on the table being truncated. It is preferable to use the DELETE statement rather than the TRUNCATE statement to delete all rows in a table.

Required privilege

No privilege is required for the table owner.

DELETE for another user's table.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

TRUNCATE TABLE [Owner.]TableName

Parameters

Parameter Description

[Owner.]TableName

Identifies the table to be truncated.

Description

  • TRUNCATE is a DDL statement. A commit is performed before and after execution of the TRUNCATE statement.

  • If your table has out of line columns and there are millions of rows to truncate, consider calling the ttCompact built-in procedure to free memory.

  • Concurrent read committed read operations are allowed, and semantics of the reads are the same as for read committed reads in presence of DELETE statements.

  • TRUNCATE is allowed even when there are child tables. However, child tables need to be empty for TRUNCATE to proceed. If any of the child tables have any rows in them, TimesTen returns an error indicating that a child table is not empty.

  • TRUNCATE is not supported with any detail table of a materialized view, table that is a part of a cache group, or temporary table.

  • When a table contains out of line varying-length data, the performance of TRUNCATE TABLE is similar to that of DELETE statement that deletes all rows in a table. For more details on out-of line data, see "Numeric data types".

  • Where tables are being replicated, the TRUNCATE statement replicates to the subscriber, even when no rows are operated upon.

  • When tables are being replicated with timestamp conflict checking enabled, conflicts are not reported.

  • DROP TABLE and ALTER TABLE operations cannot be used to change hash pages on uncommitted truncated tables.

Examples

To delete all the rows from the recreation.clubs table, use:

TRUNCATE TABLE recreation.clubs;

UNLOAD CACHE GROUP

The UNLOAD CACHE GROUP statement removes data from the cache group.

Required privilege

No privilege is required for the cache group owner.

UNLOAD or UNLOAD ANY CACHE GROUP for another user's cache group.

INSERT, DELETE, UPDATE privileges on underlying tables.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

UNLOAD CACHE GROUP [Owner.]GroupName 
  [WHERE ConditionalExpression]
  [COMMIT EVERY n ROWS]

or

UNLOAD CACHE GROUP [Owner.]GroupName 
WITH ID (ColumnValueList)

Note:

The WITH ID clause is not supported in TimesTen Scaleout.

Parameters

Parameter Description

[Owner.]GroupName

Name assigned to the cache group.

WHERE ConditionalExpression

Use the WHERE clause to specify a search condition to qualify the target rows of the cache operation. If you use more than one table in the WHERE clause and the tables have columns with the same names, fully qualify the table names.

COMMIT EVERY n ROWS

Use the COMMIT EVERY n ROWS clause to indicate the frequency (based on the number of rows that are unloaded) at which a commit is issued during the unload operation.

ROWS refers to the number of rows that are deleted from the cache group. For example, if your cache group has one cache instance and the cache instance consists of 1 parent row and 10 child rows, and you issue COMMIT EVERY 2 ROWS, TimesTen issues one commit after the entire cache instance is deleted. TimesTen does not commit in the middle of deleting an cache instance. So once the unload operation reaches its threshold (2 rows in this case), TimesTen issues a commit after all rows are deleted for that cache instance.

If you specify this clause, the cache agent must be running and the unload must be the only operation in the transaction.

Express n as an integer where (n >= 0). If you specify 0 for n, the UNLOAD CACHE GROUP statement is executed as one transaction and the cache agent does the delete.

To improve performance, use this clause when you are performing operations on cache groups that affect large amounts of data.

Do not use this clause when you have cache groups with a small amount of data.

WITH ID ColumnValueList

The WITH ID clauses enables you to use primary key values to unload the cache instance. Specify ColumnValueList as either a list of literals or binding parameters to represent the primary key values.

The WITH ID clause is not supported in TimesTen Scaleout.

Description

  • The UNLOAD CACHE GROUP statement deletes rows from the TimesTen cache tables without affecting the data in the Oracle database tables.

  • If your table has out of line columns and there are millions of rows, consider calling the ttCompact built-in procedure to free memory.

  • If the cache group is replicated, an UNLOAD CACHE GROUP statement deletes the entire contents of any replicated cache group as well.

  • Execution of the UNLOAD CACHE GROUP statement for an AWT cache group waits until updates on the rows have been propagated to the Oracle database.

  • The UNLOAD CACHE GROUP statement can be used for any type of cache group. See "CREATE CACHE GROUP" for information on cache groups.

  • Use the UNLOAD CACHE GROUP statement carefully with cache groups that have the AUTOREFRESH attribute. A row that is unloaded can reappear in the cache group as the result of an autorefresh operation if the row or its child rows are updated in the Oracle database.

  • Following the execution of an UNLOAD CACHE GROUP statement, the ODBC function SQLRowCount(), the JDBC method getUpdateCount(), and the OCI function OCIAttrGet() with the OCI_ATTR_ROW_COUNT argument return the number of cache instances that were unloaded.

  • If you specify the COMMIT EVERY n ROWS clause, the cache agent performs the unload operation and commits the transaction after unloading the data. Make sure the cache agent is up and running. If you do not specify the COMMIT EVERY n ROWS clause, the unload operation is executed by the application.

  • If you specify the COMMIT EVERY n ROWS clause, you cannot rollback the unload operation. If the unload operation fails when you specify the COMMIT EVERY n ROWS clause (where n >= 0), the cache group could be in an inconsistent state since some unloaded rows are already committed. Therefore, some cache instances may be partially unloaded. If this occurs, unload the cache group again.

  • Use the WITH ID clause to specify binding parameters.

  • The UNLOAD CACHE GROUP operation is executed in its own transaction.

Restrictions

  • The UNLOAD CACHE GROUP...WITH ID clause is not supported in TimesTen Scaleout.

  • Do not reference child tables in the WHERE clause.

  • Do not user the WITH ID clause on static read-only cache groups, or static user-managed cache groups with the autorefresh attribute.

  • Do not use the WITH ID clause with the COMMIT EVERY n ROWS clause.

Examples

Use the UNLOAD CACHE GROUP... COMMIT EVERY n ROWS to unload data from cached tables. The cache agent unloads the data because the COMMIT EVERY n ROWS clause is used.

Command> UNLOAD CACHE GROUP testcache WHERE sampleuser.orders.order_id > 100
         COMMIT EVERY 100 ROWS;
2 cache instances affected.

CREATE and UNLOAD a cache group. The application performs the unload operation because the COMMIT EVERY n ROWS clause is not used.

CREATE CACHE GROUP recreation.cache
    FROM recreation.clubs (
        clubname CHAR(15) NOT NULL,
        clubphone SMALLINT,
        activity CHAR(18),
        PRIMARY KEY(clubname))
    WHERE (recreation.clubs.activity IS NOT NULL);
UNLOAD CACHE GROUP recreation.cache;

UPDATE

The UPDATE statement updates the values of one or more columns in all rows of a table or in rows that satisfy a search condition.

Required privilege

No privilege is required for the table owner.

UPDATE for another user's table.

Usage with TimesTen Scaleout

This statement is supported with TimesTen Scaleout.

SQL syntax

UPDATE [hint] [FIRST NumRows] 
{[Owner.]TableName [CorrelationName]}
SET {ColumnName =
{Expression1 | NULL | DEFAULT}} [,...]
[ WHERE SearchCondition ]
RETURNING|RETURN Expression2[,...] INTO DataItem[,...]

Parameters

Parameter Description

hint

Specifies a statement level optimizer hint for the UPDATE statement. See "Statement level optimizer hints" for information on statement level optimizer hints.

FIRST NumRows

Specifies the number of rows to update. FIRST NumRows is not supported in subquery statements. NumRows must be either a positive INTEGER value or a dynamic parameter placeholder. The syntax for a dynamic parameter placeholder is either ? or :DynamicParameter. The value of the dynamic parameter is supplied when the statement is executed.

[Owner.]TableName [CorrelationName]

[Owner.]TableName identifies the table to be updated.

CorrelationName specifies an alias for the table and must conform to the syntax rules for a basic name. See "Basic names" for details. When accessing columns of that table elsewhere in the UPDATE statement, use the correlation name instead of the actual table name. The scope of the correlation name is the SQL statement in which it is used.

All correlation names within one statement must be unique.

SET ColumnName

ColumnName specifies a column to be updated. You can update several columns of the same table with a single UPDATE statement. Primary key columns can be included in the list of columns to be updated as long as the values of the primary key columns are not changed.

Expression1

Any expression that does not contain an aggregate function. The expression is evaluated for each row qualifying for the update operation. The data type of the expression must be compatible with the data type of the updated column. Expression1 can specify a column or sequence CURRVAL or NEXTVAL reference when updating values. See "Using CURRVAL and NEXTVAL in TimesTen Classic" for more details.

NULL

Puts a NULL value in the specified column of each row satisfying the WHERE clause. The column must allow NULL values.

DEFAULT

Specifies that the column should be updated with the default value.

WHERE SearchCondition

The search condition can contain a subquery. All rows for which the search condition is true are updated as specified in the SET clause. Rows that do not satisfy the search condition are not affected. If no rows satisfy the search condition, the table is not changed.

Expression2

Valid expression syntax. See Expressions for information.

DataItem

Host variable or PL/SQL variable that stores the retrieved Expression2 value.

Description

  • For TimesTen Scaleout, you cannot update distribution key column(s) unless you update the column(s) to the same value.

  • You cannot update primary key column(s) unless you update the column(s) to the original value.

  • If the WHERE clause is omitted, all rows of the table are updated as specified by the SET clause.

  • TimesTen generates a warning when a character or binary string is truncated during an UPDATE operation.

  • Constraint violations (UNIQUE, FOREIGN KEY, NOT NULL) result in the failure of the UPDATE statement.

  • The UPDATE operation fails if it violates any foreign key constraint. See "CREATE TABLE" for a description of foreign key constraints.

  • Restrictions on the RETURNING clause:

    • Each Expression2 must be a simple expression. Aggregate functions are not supported.

    • You cannot return a sequence number into an OUT parameter.

    • ROWNUM and subqueries cannot be used in the RETURNING clause.

    • Parameters in the RETURNING clause cannot be duplicated anywhere in the UPDATE statement.

    • Using the RETURNING clause to return multiple rows requires PL/SQL BULK COLLECT functionality. See "FORALL and BULK COLLECT operations" in Oracle TimesTen In-Memory Database PL/SQL Developer's Guide.

    • In PL/SQL, you cannot use a RETURNING clause with a WHERE CURRENT operation.

Examples

Use the UPDATE statement to update employees with department_id = 110. For employees with department_id = 110, update the manager_id to the manager_id of employees with job_id = 'FI_ACCOUNT'. Use the DISTINCT qualifier in the subquery of the SET clause.

First find the manager_id of employees with job_id = 'FI_ACCOUNT.'

Command> SELECT manager_id FROM employees WHERE job_id = 'FI_ACCOUNT';
< 108 >
< 108 >
< 108 >
< 108 >
< 108 >
5 rows found.

Next find the manager_id of employees with department_id = 110.

Command> SELECT manager_id FROM employees WHERE department_id = 110;
< 101 >
< 205 >
2 rows found.

Now update the manager_id of employees with department_id = 110. Use SELECT DISTINCT in the subquery of the SET clause. After the UPDATE, verify the manager_id for employees with department_id = 110 was updated.

Command> UPDATE employees SET manager_id =
          (SELECT DISTINCT employees.manager_id
           FROM employees
           WHERE employees.job_id = 'FI_ACCOUNT')
         WHERE employees.department_id = 110;
2 rows updated.

Command> SELECT manager_id FROM employees WHERE department_id = 110;
< 108 >
< 108 >
2 rows found.

Use subqueries in the SET clause of the UPDATE statement. Update employees with location_id = 1700 or location_id = 2400. Set department_id for these employees to the department_id of location_id = 2500. (This is department_id 80). Set salary for these employees to the maximum salary of their department.

First query the first 5 employees to check their department_id and salary.

Command> SELECT FIRST 5 employee_id, department_id, salary
         FROM employees
         ORDER BY employee_id, department_id, salary;
< 100, 90, 24000 >
< 101, 90, 17000 >
< 102, 90, 17000 >
< 103, 60, 9000 >
< 104, 60, 6000 >
5 rows found.

Now use the UPDATE statement to update employees.

Command> UPDATE employees e1
         SET department_id =
              (SELECT department_id FROM departments
               WHERE location_id = 2500),
             salary =
              (SELECT MAX(salary) FROM employees e2
                WHERE e1.department_id = e2.department_id)
         WHERE department_id IN
           (SELECT department_id FROM departments
            WHERE location_id = 2400 OR location_id = 1700);
19 rows updated.

Query the first five employees again to check that employees with the original department_id of 90 have been updated. The department_id is now 80 and the salary is 24000.

Command> SELECT FIRST 5 employee_id, department_id, salary
         FROM employees
         ORDER BY employee_id, department_id, salary;
< 100, 80, 24000 >
< 101, 80, 24000 >
< 102, 80, 24000 >
< 103, 60, 9000 >
< 104, 60, 6000 >
5 rows found.

The following example increases the price of parts costing more than $500 by 25 percent.

UPDATE purchasing.parts
SET salesprice = salesprice * 1.25
WHERE salesprice > 500.00;

This next example updates the column with the NEXTVAL value from sequence seq.

UPDATE student SET studentno = seq.NEXTVAL WHERE name = 'Sally';

The following query updates the status of all the customers who have at least one unshipped order.

UPDATE customers SET customers.status = 'unshipped'
WHERE customers.id = ANY
    (SELECT orders.custid FROM orders
        WHERE orders.status = 'unshipped');

The following statement updates all the duplicate orders, assuming id is not a primary key.

UPDATE orders a SET orders.status = 'shipped'
    WHERE EXISTS (SELECT 1 FROM orders b
        WHERE a.id = b.id AND a.rowid < b.rowid);

This next example updates job_id, salary and department_id for an employee whose last name is'Jones' in the employees table. The values of salary, last_name and department_id are returned into variables.

Command> VARIABLE bnd1 NUMBER(8,2);
Command> VARIABLE bnd2 VARCHAR2(25) INLINE NOT NULL;
Command> VARIABLE bnd3 NUMBER(4);
Command> UPDATE employees SET job_id='SA_MAN', salary=salary+1000,
          department_id=140 WHERE last_name='Jones'
         RETURNING salary*0.25, last_name, department_id
         INTO :bnd1, :bnd2, :bnd3;
1 row updated.
Command> PRINT bnd1 bnd2 bnd3;
BND1                 : 950
BND2                 : Jones
BND3                 : 140

Join update

TimesTen supports join update statements. A join update can be used to update one or more columns of a table using the result of a subquery.

Syntax

UPDATE [Owner.]TableName
SET ColumnName=Subquery
 [WHERE SearchCondition]

or

UPDATE [Owner.]TableName
SET (ColumnName[,...])=Subquery
 [WHERE SearchCondition]

Parameters

A join update statement has the following parameters:

Parameter Description

[Owner.]TableName

Identifies the table to be updated.

SET (ColumnName[,...])= Subquery

Specifies the column to be updated. You can update several columns of the same table with a single UPDATE statement. The SET clause can contain only one subquery, although this subquery can be nested.

The number of values in the select list of the subquery must be the same as the number of columns specified in the SET clause. An error is returned if the subquery returns more than one row for any updated row.

WHERE SearchCondition

The search condition can contain a subquery. All rows for which the search condition is true are updated as specified in the SET clause. Rows that do not satisfy the search condition are not affected. If no rows satisfy the search condition, the table is not changed.

Description

The subquery in the SET clause of a join update does not reduce the number of rows from the target table that are to be updated. The reduction must be specified using the WHERE clause. Thus if a row from the target table qualifies the WHERE clause but the subquery returns no rows for this row, this row is updated with a NULL value in the updated column.

Examples

In this example, if a row from t1 has no match in t2, then its x1 value in the first SELECT and its x1 and y1 values in the second SELECT are set to NULL.

UPDATE t1 SET x1=(SELECT x2 FROM t2 WHERE id1=id2);
UPDATE t1 SET (x1,y1)=(SELECT x2,y2 FROM t2 WHERE id1=id2);

In order to restrict the UPDATE statement to update only rows from t1 that have a match in t2, a WHERE clause with a subquery has to be provided as follows.

UPDATE t1 SET x1=(SELECT x2 FROM t2 WHERE id1=id2) 
  WHERE id1 IN (SELECT id2 FROM t2);
UPDATE t1 SET (x1,y1)=(SELECT x2,y2 FROM t2 WHERE id1=id2) 
  WHERE id1 IN (SELECT id2 FROM t2);

See also

SELECT