Network Information Services (NIS and NIS+) Guide

Administering NIS+ Tables

This section describes how the following topics:

• Using the nistbladm Command
• Using the niscat Command
• Using the nismatch and nisgrep Commands
• Using the nisln Command
• Using the nissetup Command
• Using the nisaddent Command

Using the nistbladm Command

The nistbladm command is the primary NIS+ table administration command. With it, you can create, modify, and delete NIS+ tables and entries. To create a table, its directory must already exist. To add entries to the table, the table and columns must already be defined.

To create a table, you must have create rights to the directory under which you will create it. To delete a table, you must have destroy rights to the directory. To modify the contents of a table, whether to add, change, or delete entries, you must have modify rights to the table or the entries.

To create a table, use:

nistbladm -c table-type columnspec...tablename

To delete a table, use:

nistbladm -d tablename columnspec ::=  column=[CSI,rights]

To add entries, use:

nistbladm -a
nistbladm -A entry

To modify entries, use:

nistbladm -m new-entry old-entry

To remove entries, use:

nistbladm -r
nistbladm -R [entry or table] entry::=column=value ... tablename
[column=value,...],tablename

The columnspec syntax is explained in the command description for nistbladm.

Using the niscat Command

The niscat command displays the contents of an NIS+ table. However, you can also use it to display the object properties of the table. You must have read rights to the table, entries, or columns that you wish to display.

To display the contents of a table, use:

niscat [-hM] tablename

To display the object properties of a table, use:

niscat -o tablename
niscat -o entry 

For more information, see the command description for niscat.

Using the nismatch and nisgrep Commands

The nismatch and nisgrep commands search NIS+ tables for entries that match a particular string or regular expression, respectively. They display either the entries themselves or only a count of how many entries matched. The differences between the nismatch and nisgrep commands are highlighted in the following table.

The tasks and examples in this section describe the syntax for both commands.

To use either command, you must have read access to the table you are searching through.

The examples in this section are based on the values in the following table, named depts.wiz.com. Only the first two columns are searchable.

Searching with Regular Expressions

Regular expressions are combinations of text and symbols that you can use to search for special configurations of column values. For example, the regular expression 'Hello' searches for a value that begins with Hello:

rootmaster% nisgrep -h greeting='Hello' phrases.wiz.com. 

The regular expression symbols are summarized in the following table:

The following are three examples using regular expressions:

• To search through the first column, use:

  nismatch string tablename
  nisgrep reg-exp tablename
  

• To search through a particular column, use:

  nismatch column=string tablename
  nisgrep column=reg-exp tablename
  

• To search through multiple columns, use:

  nismatch column=string ... tablename
  nismatch [column=string,...],tablename
  nisgrep  column=reg-exp ... tablename
  

Searching the First Column

To search for a particular value in the first column of a table, enter the first column value and a tablename. In nismatch, the value must be a string. In nisgrep, the value must be a regular expression.

nismatch [-h] string tablename
nisgrep [-h] reg-expression tablename

This example searches through the depts table for all the entries whose first column has a value of R&D:

rootmaster% nismatch -h 'R&D' depts.wiz.com.
rootmaster% nisgrep -h 'R&D' depts.wiz.com.

Note that quotes are used around the R&D expression to prevent the shell from interpreting the ampersand (&) as a meta character.

Searching a Particular Column

To search through a particular column other than the first, use:

nismatch column=string tablename
nisgrep column=reg-expression tablename

This example searches through the depts table for all the entries whose second column has a value of Austin:

rootmaster% nismatch -h Site=Austin depts.wiz.com
rootmaster% nisgrep -h Site=Austin depts.wiz.com

Searching Multiple Columns

To search for entries with matches in two or more columns, use:

nismatch [-h] column=string ... tablename
nismatch [-h] [column=string,...],tablename
 
nisgrep [-h] column=reg-exp ... tablename

This example searches for entries whose second column has a value of Austin and whose third column has a value of jbrown:

rootmaster% nismatch -h [Site=Austin,Manager=jbrown],  depts.wiz.com. 
rootmaster% nisgrep -h Site=Austin Manager=jbrown depts.wiz.com.

Using the nisln Command

The nisln command creates symbolic links between NIS+ objects and table entries. You can use it to link objects to objects or objects to table entries. (You cannot create a link that originates with a table entry.) All NIS+ administration commands accept the -L flag, which directs them to follow links between NIS+ objects.

To create a link to another object or entry, you must have modify rights to the source object; that is, the one that will point to the other object or entry.

Attention: Never link a cred table. Each org_dir directory should have its own cred table. Do not use a link to some other org_dir cred table. NIS+ cannot operate correctly with linked cred tables.

To create a link, use:

nisln source target 

The following table provides details about the nisln command options.

For more information, see the command description for nisln.

Using the nissetup Command

The nissetup command expands an existing NIS+ directory object into a domain by creating the org_dir and groups_dir directories, and a full set of NIS+ tables. It does not, however, populate the tables with data. For that, use the nisaddent command. Expanding a directory into a domain is part of the process of setting up a domain.

The nissetup command can expand a directory into a domain that supports NIS clients as well.

To use nissetup, you must have modify rights to the directory under which you will store the tables.

To expand a directory into an NIS+ domain, use:

/usr/lib/nis/nissetup
/usr/lib/nis/nissetup directory-name

To expand a directory into an NIS-compatible NIS+ domain, use:

/usr/lib/nis/nissetup -Y
/usr/lib/nis/nissetup -Y directory-name

For more information, see the command description for nissetup.

Using the nisaddent Command

The nisaddent command loads information from text files or NIS maps into NIS+ tables. It can also dump the contents of NIS+ tables back into text files.

You can use nisaddent to transfer information from one NIS+ table to another (for example, to the same type of table in another domain), but not directly. First, you need to dump the contents of the table into a file, and then load the file into the other table. Make sure that the information in the file is formatted properly. Appendix A, Information in NIS+ Tables describes the format required for each table.

When you load information into a table, you can use any of three options: replace, append, or merge. The append option simply adds the source entries to the NIS+ table. With the replace option, NIS+ first deletes all existing entries in the table and then adds the entries from the source. In a large table, this adds a large set of entries into the table's .log file (one set for removing the existing entries, another for adding the new ones), taking up space in /var/nis and making propagation to replicas time-consuming.

The merge option produces the same result as the replace option but uses a different process, one that can greatly reduce the number of operations that must be sent to the replicas. With the merge option, NIS+ handles three types of entries differently:

• Entries that exist only in the source are added to the table
• Entries that exist in both the source and the table are updated in the table
• Entries that exist only in the NIS+ table are deleted from the table

When updating a large table with a file or map whose contents are not greatly different from those of the table, the merge option can spare the server a great many operations. Because the merge option deletes only the entries that are not duplicated in the source (the replace option deletes all entries, indiscriminately), it saves one delete and one add operation for every duplicate entry.

If you are loading information into the tables for the first time, you must have create rights to the table object. If you are overwriting information in the tables, you must have modify rights to the tables.

To load information from text files, use:

/usr/lib/nis/nisaddent -f filename table-type [domain]
/usr/lib/nis/nisaddent -f filename -t tablename table-type [domain]

To load information from NIS maps, use:

/usr/lib/nis/nisaddent -y NISdomain table-type [domain]
/usr/lib/nis/nisaddent -y NISdomain -t tablename table-type [domain]
/usr/lib/nis/nisaddent -Y map table-type [domain]
/usr/lib/nis/nisaddent -Y map -t tablename table-type [domain]

To dump information from an NIS+ table to a file, use:

/usr/lib/nis/nisaddent -d [-t tablename ] tabletype > filename 

Another way to load information from files is to use stdin as the source. However, you cannot use the -m option with stdin. You can use redirect (>) or pipe (|), but you cannot pipe into another domain.

The following table provides more information.

cat filename > nisaddent table-type

cat filename > nisaddent -a table-type

cat filename > nisaddent \
  -a table-type NIS+domain

cat filename | nisaddent table-type

cat filename | nisaddent -a table-type

nisaddent -f /etc/auto_home.xfr \
  -t auto_home.org_dir.wiz.com. key-value

nisaddent -f /etc/auto_home.xfr \
  -t auto_home.org_dir.wiz.com. key-value \
  sales.wiz.com.

For more information, see the command description for nisaddent.