ADSM-L

ADSM Crash following triggered db backup

2000-01-31 22:40:02
Subject: ADSM Crash following triggered db backup
From: Eric Winters <ewinters AT AU1.IBM DOT COM>
Date: Tue, 1 Feb 2000 14:40:02 +1100
Guys,

AIX 4.3.2
AIX 3.1.2.40
Database 6 GB
Recovery Log 5 GB
DBTrigger - 40%

I have system which triggered an incremental database backup. The backup
completed but by the time the backup finished the recovery log was still
exceeding the trigger - another database backup started only to result in
an ADSM crash shortly thereafter.

The last thing seen in the activity log was the following.

01/30/00   02:20:43  ANR9999D logread.c(179): Attempt to read LSN
                      388227.217.816 below trunc LSN 388347.115.2545.
01/30/00   02:20:43  ANR9999D pkthread.c(492): Run-time assertion failed:
                      "tmCmpTsn( ctlP->undoTsnTable[slot], logRecP->tsn )
==
                      EQUAL", Thread 74, File iclog.c, Line 771.

Can anyone advise how the crash can be avoided under these circumstances?
Obviously I could have a bigger recovery
 log or change the trigger or reschedule the backups so the log doesn't
change so fast - but all this being equal, how
can I avoid an ADSM crash if the situation happens again? Is there a fix,
or can the system be configured? Or is this to be
expected if the recovery log fills during a database backup?

Thanks,


Regards,

Eric Winters
IBM Global Services Australia
<Prev in Thread] Current Thread [Next in Thread>
  • ADSM Crash following triggered db backup, Eric Winters <=