Management procedures and utilities

IBM® Content Classification provides several utilities to help you manage system activity.

Common system management activities include:

Starting, stopping, and restarting server-side components
Changing data server connection parameters
Changing the identity or password of the user account under which the system runs (Windows only)
Converting data server files for editing purposes
Backing up data server files, knowledge bases, and decision plans
Importing and exporting language customization data
Preparing saved analysis data so that it can be imported into Classification Workbench and used to improve knowledge bases and decision plans
Rebuilding index files if a program failure causes a content set file or knowledge base to become corrupted

Starting and stopping server components
IBM Content Classification server components are controlled by the Process Manager. Use this service to start, stop, and restart server components that are running on the same computer.
Enabling support for the IPv6 protocol
You can configure your IBM Content Classification system to support addresses and URLs that adhere to the IP version 6 (IPv6) protocol. After you enable support for IPv6 addresses, IBM Content Classification will no longer use IPv4 addresses for any socket communications. Support for the IPv6 protocol is certified on Windows.
Changing data server connection parameters
To view and modify the connection parameters to a remote data server, use the DBRemoteConnectionParameters command. To view and modify the connection parameters to a data server that runs on the local computer, use the DataServerParameters command.
Changing the Process Manager user account
On Windows, you can change the user account under which the Process Manager runs.
Converting data server files
If you need to edit data server files, you can convert the binary data to text, and then convert the changed text back into binary format when you return the files to the data server.
Backing up the data server
To prevent data loss, periodically back up the data server files to a backup directory.
Configuring the system to use a backup data server
If an IBM Content Classification server that hosts the data server fails and you previously set up a backup server as part of your infrastructure for high availability, you must configure the system to use the data server on the backup server.
Configuring the data server to connect to a backup remote server
If you previously set up a backup server as part of your infrastructure for high availability and you need to use that backup server in place of a failed remote IBM Content Classification server (that is, a server that connects to a data server on a separate IBM Content Classification server), you must configure the data server to refer to the backup remote server instead of the failed remote server.
Importing and exporting language customization data
If you want to customize the language processing that is done on a knowledge base, you can use the bnsConfig88 command to import and export the text analysis settings that you want to use.
Preparing saved analysis data for import into Classification Workbench
You can run a command (bnsExtractTexts88.exe on Windows, or bnsRun extract on AIX®, Linux or Solaris) to extract stored feedback and analysis data and save it in XML format. You can then import the XML file into Classification Workbench and use it to analyze and fine tune knowledge bases and decision plans.
Rebuilding index files
If a content set file or knowledge base becomes corrupted, you can use the Rebuild_Index utility to create a new index for the file.
Merging learning data (SARC) files
Available in Fix Pack 1: You can use the MergeSarc command to combine multiple learning data (SARC) files, and optionally build a knowledge base by using the combined learning data.
Bundling decision plans and knowledge bases for IBM StoredIQ
Available in Fix Pack 1: You can use the BundleDPKB command-line tool to prepare decision plan files, knowledge base files, and learning data files for use with IBM StoredIQ.