Results 1 to 3 of 3

Thread: Reorg problem

  1. #1
    Join Date
    Sep 2002
    Posts
    456

    Arrow Unanswered: Reorg problem

    Database DB2 UDB 7.1
    Platform Linux

    Last week I ran the reorg on the tables and was consistenly getting this message logged in diag file:

    -----
    DIA1000C An unexpected end of file was reached "".

    ZRC=FFFF8401
    ------

    Then, when the reorg was going against one of the biggest table (no. of rows :: 70000, row length :: 160 byte) in rebuilding phase of index, things went from bad to worse:

    --
    2003-05-02-14.39.04.321883 Instance:db2inst1 Node:000
    PID:31905(db2agent (xxxxxxxxx))
    Appid:C0A86466.9189.030502030400
    base_sys_utilities sqlesrsp Probe:9 Database:xxxxxxxxx

    RIDS/sqlesrsp MS/DB_BAD_FAT PRCS/9f6fb ffff


    2003-05-02-14.39.04.331524 Instance:db2inst1 Node:000
    PID:31905(db2agent (xxxxxxxxx)) Appid:C0A86466.9189.030502030400
    buffer_pool_services sqlbTerminateAgent Probe:837 Database:xxxxxxxxx
    DIA3727C The database has been marked as being damaged.

    ZRC=FFFFE11A

    2003-05-02-14.39.04.353956 Instance:db2inst1 Node:000
    PID:31905(db2agent (xxxxxxxxx)) Appid:C0A86466.9189.030502030400
    base_sys_utilities sqleTermDbConnect Probe:5 Database:xxxxxxxxx
    DIA3727C The database has been marked as being damaged.

    ZRC=FFFFE11A

    2003-05-02-14.39.04.409451 Instance:db2inst1 Node:000
    PID:31905(db2agent (xxxxxxxxx)) Appid:C0A86466.9189.030502030400
    buffer_pool_services sqlbStopPools Probe:0 Database:xxxxxxxxx

    Stopping the database.
    --

    This whole thing also created my trap files. Any idea guys? what could have caused this...

    dollar

  2. #2
    Join Date
    Nov 2002
    Location
    Delaware
    Posts
    186

    Re: Reorg problem

    There's a bug in DB2. Its supposed to be fixed in FP8. We ran into this almost a year ago, and since it was a prod system they logged in and manually updated the files to fix it. What happens when a re-org finished with an error, it does not write that info out to the log files and if the db crashes, upon startup it cannot find that info that it needs to bring the tablespace online. We had 2 choices, take and old backup and run it forward to the point before you started the restore. But since ours was production we had IBM edit the control files to bring the tablespace online. The tablespace doesn't loose the data, it just doesn't get logged.

  3. #3
    Join Date
    Sep 2002
    Posts
    456

    Re: Reorg problem

    Thanks!!!!

    But in our case database is allright, database got corrupted and then there was a crash recovery. I am going to open the ticket with IBM sometime this week, because I want to find out why it happened and what can I do to prevent this in future.

    dollar

    Originally posted by quigleyd
    There's a bug in DB2. Its supposed to be fixed in FP8. We ran into this almost a year ago, and since it was a prod system they logged in and manually updated the files to fix it. What happens when a re-org finished with an error, it does not write that info out to the log files and if the db crashes, upon startup it cannot find that info that it needs to bring the tablespace online. We had 2 choices, take and old backup and run it forward to the point before you started the restore. But since ours was production we had IBM edit the control files to bring the tablespace online. The tablespace doesn't loose the data, it just doesn't get logged.

Posting Permissions

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