LGWR failure and is causing instance shutdown.

Hi All,

I experienced a problem where oracle instance shuts down,
and the last entry on alert_sid.log file:

:
Thread 1 advanced to log sequence 22412
  Current log# 2 seq# 22412 mem# 0: /opt/dbfiles1/dev1/oradata/dev1/redo02.log
LGWR: terminating instance due to error 27152
Instance terminated by LGWR, pid = 13562
Wed Jul 25 14:16:28 2001
Errors in file /oracle/admin/dev1/bdump/dev1_lgwr_13562.trc:
ORA-27300: OS system dependent operation:semctl failed with status: 22
ORA-27301: OS failure message: Invalid argument
ORA-27302: failure occurred at: sskgpwrm1
ORA-27152: attempt to post process failed
Additional information: 1
Additional information: -2090048448
ORA-27300: OS system dependent operation:semctl failed with status: 22
ORA-27301: OS failure message: Invalid argument
ORA-27302: failure occurred at: sskgpwpost1
ORA-27303: additional information: semid = 12648451

Some additional information:
From init.ora: processes = 50
Oracle version is 8.1.6

I'm quite at lost at how to start investigating this problem.
The problem has occurred three times in the span of the last 10 days.

Any help, hints would be appreciated.

Thanks heaps.
Lucy
lusianaAsked:
Who is Participating?
 
ianmcgrathConnect With a Mentor Commented:
Some info that might help

The ORA-27152 indicates a problem at the Network layer

Bug 1228104:
The customer runs batch jobs from an MVS system against an 8.1.5 database on
Windows NT, when they have forty running concurrently the instance terminates
with :
  ORA-27152 attempt to post process failed
  OSD-19013670: Message 19013670 not found;  product=RDBMS; facility=SOSD
  (OS 170258008)
.
This problem seems to occur when Log Writer (LGWR) is busy and many session
are waiting for 'log buffer space'. If they reduce the number of concurrent
batch jobs the problem goes away, but reducing the number prevents them from
achieving the necessary data throughput.
.
The problem seems to be exacerbated by running hot backups, so they are trying
to reduce the number of waiters by not running these during the high load and
spreading the redo logs across more devices.
.
.
Workaround
----------
Reduce the number of waiters for a particular event.
.
Related Bugs
------------
Bug:1132689


Hope this helps

Ian
0
 
KongCommented:
This is an Oracle bug, fixed in 8.1.7 and 8.1.6.3 (Solaris).

It occurs when two of your ORACLE_SIDs are similarly named. Rename one of your ORACLE_SIDs - this is relatively easy since your instance ID (ORACLE_SID) can be different from db_name.
0
 
KongCommented:
Here is some more info. Good luck!

Bug 1262942 is an issue where the hash function used to generate the key for semaphore set allocation can generate duplicates for similar ORACLE_SID/ORACLE_HOME combinations. This can cause problems when there are several databases on the same server using similar ORACLE_SIDs as if the keys are the same only one of the instances can be running at a time.

The bug is fixed in 8.1.7 and included in the 8.1.6.3 patchset release (for Solaris only). The workaround is to rename the ORACLE_SID of one of the clashing instances.

You can check for duplicate keys using 'ipcs -s':

o startup instance 1
o ipcs -s
o shutdown instance 1
o startup instance 2
o ipcs -s

0
 
lusianaAuthor Commented:
Thanks for your response.
At the moment, our database has only one instance.
The ORACLE_SID is defined to be the same as db_name.
Thanks again.
0
 
lusianaAuthor Commented:
Thanks heaps for the responses, everyone.
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.