Database Maintenance Subcommands

Description

The database subcommands allow an administrator to optimize a search database, to truncate or empty a database, to reindex a database, to delete expired RDs from a database, and recover a database.

Subcommands

The following table is a two-column table that lists the subcommand in the first column with a brief description in the second column.

-O

Optimize the database. If you are running this subcommand on any database other than the default one, you need to use the -y option. The default database is the database defined in the config/search.conf file labeled datbase-name=logicaldbname. For example, the default value is datbase-name=default and the default database directory is db/default.

Databases do not normally need to be optimized.

-X

Truncate or empty a database. If you are running this subcommand on any database other than the default one, you need to use the -y option. Disk space used for indexes is recovered, but disk space used by the main database is not recovered, instead, it is reused as new data is added to the database.

-I

Reindex a database. If you are running this subcommand on any database other than the default one, you need to use the -y option.

-E

Delete expired RDs from a database. If you are running this subcommand on any database other than the default one, you need to use the -y option.

-R

Recover all databases. This is a global command and takes no options. All database processes, including other rdmgr instances and the main search server must be stopped before running this command.

-G

Repartition the database. This command takes no options. The partitions are defined in the config/search.conf file labeled database-partitions=p1,p2,p3,.... where p1, p2, and p3 are the filenames of the partitions. The server needs to be restarted after running this command.

-B

Completely deletes the database. Recovers all the disk space. There should be no indexing happening and the Portal server has to be off when you run this subcommand.

-L

Lists selected fields from the database to stdout. Requires the -a att option. If you are running this subcommand on any database other than the default one, you need to use the -y option.

Syntax

rdmgr   [-OXIERGBL]  [-ASTDVNP]   [-a att,att,...] [-b number] [-c search.conf]
[-j number] [-l number] [-p progress]  [-r number] [-s schema] [-y dbname]

Options

The following is a two-column table that lists the options or arguments in the first column with a brief description of the corresponding option in the second column. The following options are supported:

-A

Do not use schema aliases in config/schema.rdm file in the default search directory. Use with the I subcommand.

-S

Disable schema checking. Use with the I subcommand.

-T

Operate on the taxonomy. The taxonomy is used for browsing and classifying the database contents and is in config/taxonomy.rdm file in the default search directory. Use with O, X, I, E, B, U, and L subcommands.

-D

Update the database only; do not update the index. Use with E and X subcommands.

-V

Update the index only; do not update the database. Use with E and X subcommands.

-N

The function you specified in the command works only on the non-persistent data in the resource description. RDs in the database are a merge of persistent and non-persistent data. Use with I, E, U, and L commands.

-P

The function you specified in the command works only on the persistent data in the resource description. RDs in the database are a merge of persistent and non-persistent data. Use with I, E, U, and L subcommands.

-a att, att...

Specifies attribute view list. The att names are not case sensitive and can be any attribute whether or not they are defined in the schema; for example, author or title. If you have a multi-valued att like class-1, class-2, and class-3, only enter class as the att name.

-b number

Set the indexing batch size to this number of RDs. Use with the I subcommand.

-c search.conf

Specify where the search.conf file is. If you do not use this option, the default is the config/search.conf file in the default search directory. Other wise, you have to give the full path to the file.

-j number

Limits the number of retrieved results. Use with the E subcommand. If not stated, the default value is unlimited.

-l number

Set log level to number. A setting of 1 (default) logs all the rdmgr commands. The higher the number the more detail the logfile contains. The possible levels are 1- 100. This works with all subcommands.

-p {stdout|stderr| filename)

Prints or displays progress to stdout, stderr, or filename. This works with all subcommands.

-r number

Use with the progress option. A report is generated every number of RDs. The default is 500. Use with the u, m, g, d, and U subcommands.

-s schema

Specifies the schema definition file. The default is the config/schema.rdm file in the default search directory.

-y dbname

Specifies the search database name. If you are running this command on any database other than the default one, you need to use this option. You do not need to use this option for the default database. The default database is the database defined in the config/search.conf file labeled datbase-name=filename.

Examples

Example 1

In the following example, up to 13 RDs are removed from the database if they are expired. The progress report to stdout prints the elapsed time in seconds and the number of RDs processed so far after every five resource descriptions.

In the default search directory, type:

PortalServer-base/bin/rdmgr -E -j 13 -p stdout -r 5

Example 2

The following example shows how to recover all the search databases and makes the Search engine available again. Use this command to release stale locks in the database and to roll back incomplete data transactions. Stale locks and incomplete transactions can result from a database process being abnormally terminated.

If the Search engine is ”hung’ or not responding, in the default search directory, type:

PortalServer-base/bin/rdmgr -R