Administration

Starting and Stopping Components

You can only start and stop Knowledge Management (KM) together with the portal. For more information, see Starting and Stopping the Portal .

The stand-alone engine Search and Classification (TREX) is started and stopped independently of KM and the portal. For more information, see Starting and stopping TREX .

If you want TREX to be temporarily unavailable, you do not need to stop KM or the portal. However, you will not be able to use searching and indexing functions during this time. These functions are available again as soon as TREX is restarted.

Logon and Load Balancing

You can use the load balancing features of the Application Server Java (AS Java) to enhance the performance of the cluster. For more information, see AS Java Cluster Architecture .

User Management

Knowledge Management uses the user management functions of SAP NetWeaver AS for Java as provided by the User Management Engine (UME). For more information, see User Management of the Application Server Java .

Data Backup and Restore

AS Java-Based Data Backup and Restore

The procedures for backing up and restoring KM are the same as those described in the portal technical operations documentation. For more information, see Backing Up and Restoring the Portal .

Special Features of Data Backup and Restore in Knowledge Management

Backing Up and Restoring Internal Repositories and CM Repositories in DB Mode

KM content in internal repositories and CM repositories in DB persistence mode is stored in the database. To back up KM content, you have to create a backup copy of the database. For more information about backing up and restoring the database, see the administration documentation of the relevant database.
Backing Up and Restoring Objects Outside the Database

As well as backing up the database, you have to back up the directory hierarchy of the \etc file system repository. This contains system and configuration data. By default, this repository is located at ...\usr\sap\ <SAP System ID> \SYS\global\config\cm\etc.

Include this directory in the data backup and restore process for KM. Use standard mechanisms for this.

Partial Restore of Documents From a Database Backup

If you only want to restore individual files in your productive system (with their properties and ACLs) instead of restoring the entire database, proceed as follows:
1. Import the backed up database into a second portal installation (backup system). This can be a test system, for example.
2. Create a WebDAV repository manager to cater for the CM repository in question in the backup system. For more information, see WebDAV Repository Manager .
3. You can now use the WebDAV repository manager in the productive system to access the required files and copy them to the required folders in the productive system, thereby restoring the files in question.
  
  Note
  
  When using WebDAV to restore files, you can only include the properties and ACLs of the documents in question. You can only restore additional metadata such as feedback, ratings, and read flags by restoring the entire database.
Backing Up and Restoring CM Repositories in DBFS Mode

If you are operating CM repositories in DBFS persistence mode, you must use an adapted procedure for backing up and restoring data. For more information, see Backing Up and Restoring CM Repositories in DBFS Mode .
Backing Up Content on Remote Servers

Content on remote servers that you have included in KM using repository managers is not normally backed up using the procedures for backup and restore described by SAP. Take this into account when planning your data backup and restore. Use the usual procedures of the IT department in your company.

Recommendation

We recommend storing an overview of all connected remote servers with your project documents. To subsequently create this overview, you can find the required information (address of the remote server and path details) in the KM configuration. To call up the configuration, choose System Administration System Configuration Knowledge Management Content Management Repository Managers in the portal. Note the information relevant for the server (for example, the entries in the Root Directory parameter) for the CM, file system, and WebDAV repository managers, as well as all other repository managers used in your scenario. You can then use this overview when backing up the data

Data Backup and Restore for TREX

For more information about the data backup and restore procedures for TREX, see Data Backup and Restore for TREX .

You should note that you must suspend all KM crawling tasks in the indexing monitor before starting a complete data backup (offline) for TREX. After restoring the data, you must restart the crawling tasks. For more information, see Indexing Monitor .