DBAsupport.com Forums - Powered by vBulletin
Results 1 to 9 of 9

Thread: Instance terminating during Backup

  1. #1
    Join Date
    Oct 2006
    Posts
    4

    Angry Instance terminating during Backup

    I'm admnistering a 350 GB Oracle 8 database. During backup datafiles ,control files & redologs are copied from one disk array to another.

    Of late during backup DB writer & log writer processess are terminating the instance.

    below is a part of my alert log file



    Tue Sep 5 19:07:12 2006
    alter tablespace MOVIMIENTOS12 begin backup
    Completed: alter tablespace MOVIMIENTOS12 begin backup
    Tue Sep 5 19:07:14 2006
    alter tablespace IMAGENES11 begin backup
    Completed: alter tablespace IMAGENES11 begin backup
    Tue Sep 5 19:27:49 2006
    DBW0: terminating instance due to error 27062
    Instance terminated by DBW0, pid = 13928





    Completed: alter tablespace INDICES77A begin backup
    Mon Sep 18 19:27:08 2006
    alter tablespace INDICES88 begin backup
    Mon Sep 18 19:27:12 2006
    Thread 1 advanced to log sequence 114798
    Current log# 2 seq# 114798 mem# 0: /data/sgc/dbics/dbics06/ics.log2A
    Current log# 2 seq# 114798 mem# 1: /data/sgc/dbics/admin/safety_mirror/ics.log2B
    Mon Sep 18 19:27:13 2006
    Completed: alter tablespace INDICES88 begin backup
    Mon Sep 18 19:27:13 2006
    alter tablespace MOVIMIENTOS12 begin backup
    Completed: alter tablespace MOVIMIENTOS12 begin backup
    Mon Sep 18 19:27:16 2006
    alter tablespace IMAGENES11 begin backup
    Completed: alter tablespace IMAGENES11 begin backup
    Mon Sep 18 20:49:58 2006
    Thread 1 advanced to log sequence 114799
    Current log# 3 seq# 114799 mem# 0: /data/sgc/dbics/dbics06/ics.log3A
    Current log# 3 seq# 114799 mem# 1: /data/sgc/dbics/admin/safety_mirror/ics.log3B
    Mon Sep 18 21:11:28 2006
    Thread 1 advanced to log sequence 114800
    Current log# 1 seq# 114800 mem# 0: /data/sgc/dbics/dbics06/ics.log1A
    Current log# 1 seq# 114800 mem# 1: /data/sgc/dbics/admin/safety_mirror/ics.log1B
    Mon Sep 18 21:28:27 2006
    Thread 1 advanced to log sequence 114801
    Current log# 2 seq# 114801 mem# 0: /data/sgc/dbics/dbics06/ics.log2A
    Current log# 2 seq# 114801 mem# 1: /data/sgc/dbics/admin/safety_mirror/ics.log2B
    LGWR: terminating instance due to error 27062
    Instance terminated by LGWR, pid = 17663




    kindly advice how to deal with the problem once & for all


    Regards

    Msororaji

  2. #2
    Join Date
    Apr 2003
    Location
    Gourock, Scotland
    Posts
    102
    If I have to choose between two evils, I always like to choose the one I haven't tried yet.

  3. #3
    Join Date
    Apr 2001
    Location
    Bangalore, India
    Posts
    727
    I saw some where in metalink (unable to find it now..!), it is an internal error because of DB_ASYNCH_IO=TRUE. It was suggesting some OS patch to resove the problem and till then convert DB_ASYNCH_IO=FALSE. Is there any associated error message in OS log?
    Thomas Saviour(royxavier@yahoo.com)
    Technical Lead (Databases)
    Thomson Reuters (Markets)

    http://ora600tom.wordpress.com/

  4. #4
    Join Date
    Oct 2006
    Posts
    4
    thanks for your assistance

    I checked all the OS log files & nothing was written when the instance was crashing

    Regards

  5. #5
    Join Date
    Apr 2001
    Location
    Bangalore, India
    Posts
    727
    Found in metalink few I/O realted bugs ( specially in Sun Solaris OS) and crashing database. Did you try the backup with DB_ASYNCH_IO=FALSE? Anyway time to think about an upgarde to get support from Oracle. Raise an SR with Metalink.
    Thomas Saviour(royxavier@yahoo.com)
    Technical Lead (Databases)
    Thomson Reuters (Markets)

    http://ora600tom.wordpress.com/

  6. #6
    Join Date
    Oct 2006
    Posts
    4
    Yes my OS plaftform is Solaris 5.8. Will be upgrading to Solaris 9 early next yr & later on to Oracle 9i

    My init.ora file dosent have parameter DB_ASYNCH_IO. I've inserted (ie DB_ASYNCH_IO=FALSE) but since my Database is a 24X7 I guess I'll have to wait for the next crash so that the change takes effect.

    Thanks

  7. #7
    Join Date
    Mar 2002
    Posts
    534
    Will be upgrading to Solaris 9 early next yr & later on to Oracle 9i
    wouldn't it, in summer 2007, not make more sense to directly upgrade to 10gR2 (10.2.0.4) ?

  8. #8
    Join Date
    Jul 2002
    Location
    Northampton, England
    Posts
    612
    Quote Originally Posted by Msororaji
    thanks for your assistance

    I checked all the OS log files & nothing was written when the instance was crashing

    Regards
    What OS logs did you look at? /var/adm/messages?

    Definitely raise this one with Oracle Support.
    Assistance is Futile...

  9. #9
    Join Date
    Oct 2006
    Posts
    4
    My ITM insists I upgrade to 9i and then 10g after a year. Can't convince him otherwise. Seems to be a sceptic.

    checked for logs in /var/adm/messages & for trace files in udest & bdest directories but there was non when the instance was terminating

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


Click Here to Expand Forum to Full Width