The nistbladm command is used to administer NIS+ tables. There are five primary operations that it performs: creating and deleting tables, adding entries to, modifying entries within, and removing entries from tables.
Though NIS+ does not place restrictions on the size of tables or entries, the size of data has an impact on the performance and the disk space requirements of the NIS+ server. NIS+ is not designed to store huge pieces of data, such as files; instead, pointers to files should be stored in NIS+.
NIS+ design is optimized to support 10,000 objects with a total size of 10M bytes. If the requirements exceed the above, it is suggested that the domain hierarchy be created, or the data stored in the tables be pointers to the actual data, instead of the data itself.
When creating tables, a table type, type, and a list of column definitions must be provided.
type is a string that is stored in the table and later used by the service to verify that entries being added to it are of the correct type.
Syntax for column definitions is:
flags is a combination of:
Searchable. Specifies that searches can be done on the column's values (see nismatch(1)).
Case-insensitive (only makes sense in combination with S). Specifies that searches should ignore case.
Crypt. Specifies that the column's values should be encrypted.
Binary data (does not make sense in combination with S). If not set, the column's values are expected to be null terminated ASCII strings.
XDR encoded data (only makes sense in combination with B).
access is specified in the format as defined by the nischmod(1) command.
When manipulating entries, this command takes two forms of entry name. The first uses a series of space separated colname=value pairs that specify column values in the entry. The second is a NIS+ indexed name, indexedname, of the form:
[ colname=value, . . . ],tablename
The following options are supported:
Adds entries to a NIS+ table. The difference between the lowercase `a' and the uppercase `A' is in the treatment of preexisting entries. The entry's contents are specified by the column=value pairs on the command line. Values for all columns must be specified when adding entries to a table.
Normally, NIS+ reports an error if an attempt is made to add an entry to a table that would overwrite an entry that already exists. This prevents multiple parties from adding duplicate entries and having one of them get overwritten. If you wish to force the add, the uppercase `A' specifies that the entry is to be added, even if it already exists. This is analogous to a modify operation on the entry.
Creates a table named tablename in the namespace. The table that is created must have at least one column and at least one column must be searchable.
Destroys the table named tablename. The table that is being destroyed must be empty. The table's contents can be deleted with the -R option below.
Edits the entry in the table that is specified by indexdname. indexdname must uniquely identify a single entry. It is possible to edit the value in a column that would change the indexed name of an entry.
The change (colname=value) may affect other entries in the table if the change results in an entry whose indexed name is different from indexedname and which matches that of another existing entry. In this case, the -e option will fail and an error will be reported. The -E option will force the replacement of the existing entry by the new entry (effectively removing two old entries and adding a new one).
A synonym for -E. This option has been superseded by the -E option.
Removes entries from a table. The xentry is specified by either a series of column=value pairs on the command line, or an indexed name that is specified as entryname. The difference between the interpretation of the lowercase `r' versus the uppercase `R' is in the treatment of non-unique entry specifications. Normally the NIS+ server will disallow an attempt to remove an entry when the search criterion specified for that entry resolves to more than one entry in the table. However, it is sometimes desirable to remove more than one entry, as when you are attempting to remove all of the entries from a table. In this case, using the uppercase `R' will force the NIS+ server to remove all entries matching the passed search criterion. If that criterion is null and no column values specified, then all entries in the table will be removed.
Updates attributes of a table. This allows the concatenation path (-p), separation character (specified with the (-s)), column access rights, and table type string (-t) of a table to be changed. Neither the number of columns, nor the columns that are searchable may be changed.
When creating objects, this option specifies a different set of defaults to be used during this operation. The defaults string is a series of tokens separated by colons. These tokens represent the default values to be used for the generic object properties. All of the legal tokens are described below.
This token sets the default time to live for objects that are created by this command. The value time is specified in the format as defined by the nischttl(1) command. The default value is 12 hours.
This token specifies that the NIS+ principal ownername should own the created object. Normally this value is the same as the principal who is executing the command.
This token specifies that the group groupname should be the group owner for the object that is created. The default value is NULL.
This token specifies the set of access rights that are to be granted for the given object. The value rights is specified in the format as defined by the nischmod(1) command. The default value is - - - -rmcdr- - -r- - -.
When creating or updating a table, this option specifies the table's search path. When a nis_list() function is invoked, the user can specify the flag FOLLOW_PATH to tell the client library to continue searching tables in the table's path if the search criteria used does not yield any entries. The path consists of an ordered list of table names, separated by colons. The names in the path must be fully qualified.
When creating or updating a table, this option specifies the table's separator character. The separator character is used by niscat(1) when displaying tables on the standard output. Its purpose is to separate column data when the table is in ASCII form. The default value is a space.
When updating a table, this option specifies the table's type string.
This example creates a table named hobbies in the directory foo.com. of the type hobby_tbl with two searchable columns, name and hobby.
example% nistbladm -c hobby_tbl name=S,\ a+r,o+m hobby=S,a+r hobbies.foo.com.
The column name has read access for all (that is, owner, group, and world) and modify access for only the owner. The column hobby is readable by all, but not modifiable by anyone.
In this example, if the access rights had not been specified, the table's access rights would have come from either the standard defaults or the NIS_DEFAULTS variable (see below).
To add entries to this table:
example% nistbladm -a name=bob hobby=skiing hobbies.foo.com. example% nistbladm -a name=sue hobby=skiing hobbies.foo.com. example% nistbladm -a name=ted hobby=swimming hobbies.foo.com.
In the following example, the common root domain is foo.com (NIS+ requires at least two components to define the root domain) and the concatenation path for the subdomains bar and baz are added:
example% nistbladm -u -p hobbies.bar.foo.com.:hobbies.baz.foo.com. hobbies
To delete the skiers from our list:
example% nistbladm -R hobby=skiing hobbies.foo.com.
Note: The use of the -r option would fail because there are two entries with the value of skiing.
To create a table with a column that is named with no flags set, you supply only the name and the equals (=) sign as follows:
example% nistbladm -c notes_tbl name=S,a+r,o+m note= notes.foo.com.
This example created a table, named notes.foo.com., of type notes_tbl with two columns name and note. The note column is not searchable.
When entering data for columns in the form of a value string, it is essential that terminal characters be protected by single or double quotes. These are the characters equals (=), comma (,), left bracket ([), right bracket (]), and space ( ). These characters are parsed by NIS+ within an indexed name. These characters are protected by enclosing the entire value in double quote (") characters as follows:
example% nistbladm -a fullname="Joe User" nickname=Joe nicknames
If there is any doubt about how the string will be parsed, it is better to enclose it in quotes.
This variable contains a defaults string that will be override the NIS+ standard defaults. If the -D switch is used those values will then override both the NIS_DEFAULTS variable and the standard defaults.
If this variable is set, and the NIS+ table name is not fully qualified, each directory specified will be searched until the table is found. See nisdefaults(1).
See attributes(5) for descriptions of the following attributes:
NIS+ might not be supported in future releases of the SolarisTM Operating Environment. Tools to aid the migration from NIS+ to LDAP are available in the Solaris 9 operating environment. For more information, visit http://www.sun.com/directory/nisplus/transition.html.
To modify one of the entries, say, for example, from “bob” to “robert”:
example% nistbladm -m name=robert [name=bob],hobbies
Notice that “[name=bob],hobbies” is an indexed name, and that the characters `[' (open bracket) and `]' (close bracket) are interpreted by the shell. When typing entry names in the form of NIS+ indexed names, the name must be protected by using single quotes.
It is possible to specify a set of defaults such that you cannot read or modify the table object later.