Skip Navigation Links | |
Exit Print View | |
Oracle Fusion Middleware Command-Line Usage Guide for Oracle Unified Directory 11g Release 1 (11.1.1) |
1. Server Administration Commands
2. Data Administration Commands
Task Back End Connection Options
Task Back End Connection Options
Task Back End Connection Options
Task Back End Connection Options
Task Back End Connection Options
The split-ldif command splits an LDIF file into multiple LDIF files according to a given distribution workflow element. The generated LDIF files are used to populate the partitions of a distribution deployment.
This command is supported only for the proxy.
split-ldif options
The split-ldif command splits an LDIF file into multiple LDIF files according to a given distribution workflow element. The data in the LDIF file is split based on the attributes indicated and based on the distribution type defined. The generated LDIF files are then used to populate the partitions. For each partition the split-ldif command creates a partition file as follows:
outputDirectory/outputFilenamePrefix-partitionID.ldif
Sometimes, the distribution algorithm is not able to determine the partition to which an entry should be sent, either because the entry does not contain all the parameters required by the algorithm, or the required parameters are present but they match no partition. In such a scenario, the output is written to an error file.
All the entries that do not have all the required parameters are written to the following error file:
outputDirectory/outputFilenamePrefix-missingrequired-param.ldif
All the entries that have the required parameters but whose parameters do not match any configured partition are written to the following error file:
outputDirectory/outputFilenamePrefix-partition-not-found.ldif
However, for the global index initialization you use the directory containing the files compatible with the global index format. The split-ldif command creates one directory per attribute to be indexed, and each directory contains files for initializing the global index.
The global index catalog is populated using the files in the directory created, which do not have a LDIF format. For more information, see gicadm.
The split-ldif command accepts an option in either its short form (for example, -i ldifFile) or its long form equivalent (for example, --ldifFile ldifFile).
Name of the LDIF file to split. Global Index Options and Split Options can be used to customize the behavior.
Lists the available distribution workflow elements from the directory server's configuration.
Generates an index file to be used for the global index catalog, for the listed attribute type.
Generates only the index file.
Name of the distribution workflow element to split the data.
Generates an index file where all the entries are distributed to the same single partition having the listed partitionId.
The directory where output LDIF files will be generated.
The prefix of the filename to generate (will contain the partition ID and the .ldif extension).
Overwrites generated files that may already exist from previous use.
Display the version information for the directory server.
Display examples of the usage.
Display command-line usage information for the command and exit without making any attempt to stop or restart the directory server.
Example 2-42 Using split-ldif to Populate a Global Index with One Indexed Attribute
The following command uses an existing database file (-i) which it splits into several files, based on the distribution information already defined in the Oracle Unified Directory proxy deployment. The command defines the distribution workflow element name (-d), the database file (-i) to be split, and the attribute to be indexed in the global index files (-x). Indicating -f will overwrite any existing LDIF files.
You must have deployed a Oracle Unified Directory proxy with distribution before running this command.
$ split-ldif -d "distrib-we" -i database.ldif -x employeenumber -f
Assuming, for this example, that your distribution algorithm was numeric, and that you set two partitions with boundaries 1–1000 and 1000–2000. When you run the command above, the following directory and LDIF files are created:
This file contains all the entries from database with employee numbers from 1–999, which will be used to populate partition 1.
This file contains all the entries from database with employee numbers from 1000–1999, which will be used to populate partition 2.
This directory contains the global index files for the employee number attribute.
Example 2-43 Using split-ldif to Populate a Global Index with Several Indexed Attributes
The following command uses an existing database file (-i) which it splits into several files, based on the distribution information already defined in the Oracle Unified Directory proxy deployment. The command defines the distribution workflow element name (-d), the database file (-i) to be split, and the attributes to be indexed in the global index files (-x). Indicating -f will overwrite any existing LDIF files.
You must have deployed a Oracle Unified Directory proxy with distribution before running this command.
$ split-ldif -d "distrib-we" -i database.ldif \ -x employeenumber -x uid -f
Assuming, for this example, that your distribution algorithm was numeric, and that you set two partitions with boundaries 1–50000 and 50000–100001. When you run the command above, the following LDIF files and directories are created:
This file contains all the entries from database with employee numbers from 1–49999, which will be used to populate partition 1.
This file contains all the entries from database with employee numbers from 50000–100000, which will be used to populate partition 2.
This directory contains the global index files for the employee number attribute.
This directory contains the global index files for the uid attribute.
UNIX and Linux: instance-dir/OUD/bin/split-ldif
Windows: instance-dir\OUD\bat\split-ldif.bat
gicadm