KB migration issues
This topic contains issues that you might encounter when migrating a Knowledge Base.
Failure of the cell backup
If the backup of the cell fails, verify:
- Whether sufficient disk space is available on the Integration Service computer to download the KB from the BMC TrueSight Infrastructure Management Server
- That the cell directory installationDirectory\server\etc\BPPM_presentationcellName can be copied and its files are not locked
- That you can write to the backup directory
Ensure that the trace messages of the mccomp command are not redirected to a log file. To check the trace messages, perform the following steps:
- Open the installationDirectory\server\etc\mccomp.trace file.
- Verify that all trace messages in the mccomp.trace file are commented out.
- If the mcclient.trace file is used instead of the mccomp.trace file, ensure that the trace messages are commented out in the mcclient.trace file.
Upgrade log files and reports
To verify or troubleshoot the KB migration process, you can review log files and reports.
You can find the log files under installationDirectory\server\SIEMupgrade\log. Look for the me.log and the navi.log files. The report file is located at installationDirectory\server\SIEMupgrade\report. Report names follow this syntax: PHASE_ID_TIME_STAMP, where PHASE_ID refers to the upgrade operation. For example, the report for the pre-upgrade phase would be named PreUpgradeChecks_ TIME_STAMP.
KB Migration Wizard issues
If you cannot execute the KB Migration Wizard because of connectivity issues, you can manually execute the mmigrate utility. The mmigrate utility is a BMC component that migrates older KBs to the latest KB formats.
BMC recommends that only experienced and knowledgeable users execute mmigrate utility.
For information about the mmigrate utility, see Using mmigrate utility.
Oracle Database issues
You find the following error in the \pw\pronto\logs\ProactiveNet.log file:
BPPM:9.6:PSR:Data PSR:upgrade:Oracle backend:Error "Unable to delete instance with MOTypeId = 501071 and instance ID = 82665 java.sql.SQLException: ORA-01502: index 'PROACT.SYS_C0038380' or partition of such index is in unusable state
You can encounter this entry in the log file if the index pointing to the primary key is corrupted. To overcome this problem, you must drop the index and re-create it.
To drop the index and re-create it
On the computer that has Oracle Database, run the following query:
alter table tableName drop primary key
tableName is the name of the table that has the primary key.
The primary key is dropped.
Run the following query:
alter table tableName add primary key (columnName1, columnName2)
tableName is the name of the table that has the primary key, and columnName is the name of the column to which the primary key is created.
You find the following error in the \pw\pronto\logs\ProactiveNetPRUNER.log file:
ERROR 03/19 20:02:03 Msg_Svr [Thread-1] 344001 Exception inside JMSConnector constructor. javax.naming.CommunicationException: Failed to retrieve stub from server 10.128.32.152:9376 [Root exception is java.io.StreamCorruptedException: unexpected block data]
You can encounter this entry in the log file if JServer has not initialized completely after the upgrade. It does not cause any problem for the pruner. The table partitions that are not removed get pruned in the next run of the pruner. You can safely ignore this error.