|Applies To||Federated Identity Management Module (FIM) 4.x|
|Issue||What is the procedure for backing up and restoring the FIM database in clustered mode|
org.apache.derby.client.am.DisconnectException: java.security.PrivilegedActionException : Error opening socket to server fim on port 1527 with message : null
at org.apache.derby.client.net.NetAgent.<init>(Unknown Source)
SQL Exception: Failed to start database 'C:\bea\user_projects\Domains\FIM/rsa-fim-config/demoDB', see the next exception for details.
Another instance of Derby may have already booted the database C:\bea\user_projects\Domains\FIM\rsa-fim-config\demoDB.
ERROR XSDB6: Another instance of Derby may have already booted the database C:\b
|Cause||In a clustered deployment the derby database needs to be running because the restore script connects using JDBC. This is different than the procedure described in the installation guide for the standalone backup.|
For a backup on a clustered deployment the derby database must be running on the administration server as the backup script used the configured JDBC connection to read from the database in order to create the backup XML file. To ensure data integrity during the backup you should disable the admin.ear file.
For a restore on a clustered deployment the existing database is destroyed and a new one is created. The restore script access the database directly on the filesystem and needs an exclusive lock to do the restore. The admin.ear file should not be running during the restore or else the exclusive lock will fail. The web tier and business tiers also read the new database on startup so they should be disabled during the restore so that they will get the new database startup.
|Workaround||Error when trying to run the backupconfig.sh or backupconfig.cmd file:|
|Legacy Article ID||a46849|