Kbase P134752: Database fail to restart going through logical undo phase of crash recovery with error 11105 after n
| Autor |
Progress Software Corporation - Progress |
| Acesso |
Público |
| Publicação |
10/27/2009 |
|
Status: Verified
SYMPTOM(s):
Database fail to restart going through logical undo phase of crash recovery with error 11105 after normal shutdown
SYSTEM ERROR: rmUndoLogicalChange: recid area txn updctr rmFetchRecord returns expected (11105)
Cannot truncate the database before image file with error 11105
Stack trace from _proutil reads:
rmUndoLogicalChange
rllbk
warmstrt
The following errors can be found in the log fiile prior to the normal shutdown
SYSTEM ERROR: lkrels record <recid> not locked. (435)
SYSTEM ERROR: lkrend called with active transaction (5142)
SYSTEM ERROR: lkTableLockUpdate: Can't find table lock for table number rowid owner , chain (11697)
FACT(s) (Environment):
All Supported Operating Systems
Progress/OpenEdge Product Family
CAUSE:
The exact cause is unknown at the time of this writing. Prior to normal shutdown, the administrator issued multiple kill -9 (endtask would be similar on Windows) to java processes related to AppServers, which are further indication that there were other system/environment abnormalities that were not investigated at the time.
FIX:
Restore database from a backup.
If restoring the database from a backup is not possible, then force truncate the BI file to bypass crash recovery.
For additional information regarding the Perils of Forcing into a database, please reference solution: P24330, "Consequences of using the -F startup option"
Command: proutil dbname -C truncate bi -F