Results 1 to 5 of 5

Thread: db down

  1. #1
    Join Date
    Jul 2012
    Posts
    3

    Unanswered: db down

    Hello forummers , i have extracted this from online log;

    16:17:55 Event alarms enabled. ALARMPROG = '/usr/informix/etc/log_full.sh'
    16:18:00 DR: DRAUTO is 0 (Off)
    16:18:00 Informix Dynamic Server Version 7.31.UC7 Software Serial Number AAC#J948419
    16:18:00 Informix Dynamic Server Initialized -- Shared Memory Initialized.
    16:18:00 Physical Recovery Started.
    16:18:00 Physical Recovery Complete: 9 Pages Restored.
    16:18:00 Logical Recovery Started.
    16:18:00 Assert Failed: Error during recovery left index inconsistent.
    16:18:00 Informix Dynamic Server Version 7.31.UC7
    16:18:00 Who: Session(10, informix@hpbaan, 0, -957630352)
    Thread(24, xchg_1.6, c6e97014, 4)
    File: rskey.c Line: 1373
    16:18:00 Results: Index 'baan:"bsp".ttfcmg101800#tfcmg1018002absp' is now unusable
    16:18:00 Action: Run 'oncheck -cI baan:"bsp".ttfcmg101800#tfcmg1018002absp'
    16:18:01 Assert Failed: Error during recovery left index inconsistent.
    16:18:01 Informix Dynamic Server Version 7.31.UC7
    16:18:01 Who: Session(10, informix@hpbaan, 0, -957630352)
    Thread(27, xchg_1.9, c6e97ed8, 5)
    File: rskey.c Line: 1373
    16:18:01 Results: Index 'baan:"bsp".ttfcmg101800#tfcmg1018005absp' is now unusable
    16:18:01 Action: Run 'oncheck -cI baan:"bsp".ttfcmg101800#tfcmg1018005absp'
    16:18:01 Rollforward of log record failed. iserrno = 126
    16:18:01 Log Record: log = 155245, pos = 11a6ec, type = OLDRSAM:HINSERT(40), trans = 31
    16:18:01 Assert Failed: Error during recovery left index inconsistent.
    16:18:01 Informix Dynamic Server Version 7.31.UC7
    16:18:01 Who: Session(10, informix@hpbaan, 0, -957630352)
    Thread(26, xchg_1.8, c6e979ec, 3)
    File: rskey.c Line: 1373
    16:18:01 Results: Index 'baan:"bsp".ttfcmg101800#tfcmg1018004absp' is now unusable
    16:18:01 Action: Run 'oncheck -cI baan:"bsp".ttfcmg101800#tfcmg1018004absp'
    16:18:01 Assert Failed: Error during recovery left index inconsistent.
    16:18:01 Informix Dynamic Server Version 7.31.UC7
    16:18:01 Who: Session(10, informix@hpbaan, 0, -957630352)
    Thread(25, xchg_1.7, c6e97500, 1)
    File: rskey.c Line: 1373
    16:18:01 Results: Index 'baan:"bsp".ttfcmg101800#tfcmg1018003absp' is now unusable
    16:18:01 Action: Run 'oncheck -cI baan:"bsp".ttfcmg101800#tfcmg1018003absp'
    16:18:03 See Also: /tmp/af.4012438
    16:18:03 Rollforward of log record failed. iserrno = 105
    16:18:03 Log Record: log = 155245, pos = 11b1a4, type = OLDRSAM:ADDITEM(28), trans = 31
    16:18:03 Assert Failed: Chunk 5 is being taken OFFLINE.
    16:18:03 Informix Dynamic Server Version 7.31.UC7
    16:18:03 Who: Session(10, informix@hpbaan, 0, -957630352)
    Thread(23, xchg_1.5, c6e96b28, 1)
    File: rsmirror.c Line: 1710
    16:18:03 Results: Dynamic Server must abort
    16:18:03 Action: Reinitialize shared memory
    16:18:04 See Also: /tmp/af.4032438
    16:18:04 Rollforward of log record failed. iserrno = 105
    16:18:04 Log Record: log = 155245, pos = 11b318, type = OLDRSAM:ADDITEM(28), trans = 31
    16:18:04 See Also: /tmp/af.4022439
    16:18:04 Rollforward of log record failed. iserrno = 105
    16:18:04 Log Record: log = 155245, pos = 11b234, type = OLDRSAM:ADDITEM(28), trans = 31
    16:18:04 See Also: /tmp/af.4002438
    16:18:04 Rollforward of log record failed. iserrno = 105
    16:18:04 Log Record: log = 155245, pos = 11b114, type = OLDRSAM:ADDITEM(28), trans = 31
    16:18:05 See Also: /tmp/af.3ff2438
    16:18:05 Assert Failed: INFORMIX-OnLine Must ABORT
    Critical media failure.
    16:18:05 Informix Dynamic Server Version 7.31.UC7
    16:18:05 Who: Session(10, informix@hpbaan, 0, -957630352)
    Thread(23, xchg_1.5, c6e96b28, 1)
    File: rsmirror.c Line: 1714
    16:18:05 stack trace for pid 2924 written to /tmp/af.3ff2438
    16:18:10 shmget: [EEXIST][17]: key 52564801: shared memory already exists
    16:18:10 Could not create single shared memory segment with resident and non-resident par

    titions. Proceeding to create 2 shared memory segments instead.
    16:18:10 shmget: [EEXIST][17]: key 52564801: shared memory already exists
    16:18:10 mt_shm_init: can't create resident segment

    16:18:17 See Also: /tmp/af.3ff2438, shmem.3ff2438.1
    16:18:17 Error writing '/tmp/shmem.3ff2438.1' errno = 28
    16:18:17 rsmirror.c, line 1714, thread 23, proc id 2924, INFORMIX-OnLine Must ABORT
    Critical media failure..
    16:18:17 PANIC: Attempting to bring system down
    16:18:17 semctl: errno = 22

    16:18:17 semctl: errno = 22

    16:18:17 semctl: errno = 22

    16:18:17 semctl: errno = 22

    16:18:17 semctl: errno = 22

    16:18:17 semctl: errno = 22

    16:18:17 semctl: errno = 22

    16:18:17 semctl: errno = 22

    16:18:17 semctl: errno = 22

    Pls help as the db has not up for 3 days. Fyi, i have a very little experience in Informix. Could anyone give any suggestions/solutions? Thank you.

  2. #2
    Join Date
    Sep 2011
    Location
    Pont l'Abbé, Brittany, France
    Posts
    381
    Provided Answers: 1
    Hi,

    the original problem seems to be a corruption in the first index indicated by the "run oncheck -cI" message.

    You are using a version that is no more under regular support, meaning you will have no patch BUT,

    you seem not to be able to restart probably due to remaining shared memory segments that prevent IDS from restarting. You should delete with the ipcrm -s command
    to list list, run "ipcs -s" and look for address looking like "0x52......" owned by informix or root
    Another simple way, is to reboot the server, if any possible.

    Then, I would attempt to delete the index pointed by the "oncheck -cI" message, and recreate them.

    Getting assert failed is probably a bug, but your version is too old to be patched.

    Finally, I would strongly recommand to migrate to 11.70, which is an in-place migration ( no need to unload/load the data. You are likely to get more issues like this, with chances to have no patch to resolve them.

  3. #3
    Join Date
    Jul 2012
    Posts
    3
    TQ Begooden-IT, will try ur method.

  4. #4
    Join Date
    Jul 2012
    Posts
    3
    Hi Begooden-IT,

    Have tried ur method but still cannot work. anyone have suggestions?

    Last choice have to restore last year backup tape since the data in current tape is corrupted.

    Pls anyone. Thanks.

  5. #5
    Join Date
    Sep 2011
    Location
    Pont l'Abbé, Brittany, France
    Posts
    381
    Provided Answers: 1
    HI

    I you confirm that after rebooting or deleting shared memory segments, you cannot proceed to logical recovery, best solution is to restore the last backup.
    There seems to be a corruption in chunk 5 that stop you from proceeding to logical restore till completion.

    It would be a good idea to implement a backup strategy based on more than one backup device ( ie 1 different tape per day of the week + one per week , cycling on one month )

    It would also be a good idea to seriously consider migration to 11.70. You have absolutely no chance to have support on 7.31, nor patches because this version is no more supported.

Posting Permissions

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