Informix

nested-group-icon.png

DB2

Expand all | Collapse all

IDS 12.10.FC12W1 Memory block header corruption

  • 1.  IDS 12.10.FC12W1 Memory block header corruption

    Posted Tue February 25, 2020 09:16 AM

    We are having more and more issues around shmem block header corruption with IDS 12.10.FC12W1. Sometimes the bad pool gets repaired, sometimes not. If not, the engine will eventually stop accepting new connections, so we bounce the hard way of course. 

    Anyone else experiencing similar? I've created more PMRs (TS) in the last 6 months than in the last few years once we moved to IDS 12.10. Platform is AIX 7.1. 

    Below is a snippet:

    01/23/20 07:02:58 Memory block header corruption detected in mt_shm_malloc_segid 9
    01/23/20 07:03:02 Found during mt_shm_free 1, error seen 1 times since last print to the log
    01/23/20 07:03:02 Pool '8350181' (0x7000002aa2b6040)
    01/23/20 07:03:02 Bad block header 0x7000002f106b860
    01/23/20 07:03:02 Assert Failed: Memory block header corruption detected in mt_shm_free 1
    01/23/20 07:03:02 IBM Informix Dynamic Server Version 12.10.FC12W1
    01/23/20 07:03:02 Who: Session(8350181, <user redacted> 62586972, 70000023904f108)
    Thread(19565549, sqlexec, 7000001a2226b58, 93)
    File: mtshpool.c Line: 2763
    01/23/20 07:03:02 Results: Unable to repair pool


    Thanks! 

    Mark Scranton
    mark@markscranton.com



    ------------------------------
    Mark Scranton
    ------------------------------


  • 2.  RE: IDS 12.10.FC12W1 Memory block header corruption

    Posted Tue February 25, 2020 09:26 AM
    Are clients shared memory or networked?

    Art

    Art S. Kagel, President and Principal Consultant
    ASK Database Management


    Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference.  Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves.








  • 3.  RE: IDS 12.10.FC12W1 Memory block header corruption

    Posted Tue February 25, 2020 02:14 PM
    All TCP. Not the shmem limit you might be thinking of ...

    ------------------------------
    Mark Scranton
    ------------------------------



  • 4.  RE: IDS 12.10.FC12W1 Memory block header corruption

    Posted Tue February 25, 2020 03:42 PM
    Wasn't thinking of limits, just that any share memory connected app running as "informix" is capable of corrupting shared memory! But since it's network, not that.

    Art

    Art S. Kagel, President and Principal Consultant
    ASK Database Management


    Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference.  Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves.








  • 5.  RE: IDS 12.10.FC12W1 Memory block header corruption

    Posted Sat February 29, 2020 01:45 AM
    Hello Mark,

    It would be more accurate to look at the stack trace together, but it seems that IT30342 is a similar error in 12.10.FC12W1 and above.
    https://www-01.ibm.com/support/docview.wss?uid=swg1IT30342

    I have not seen this case in my customers yet.
    IBM can determine the exact cause of the error, but have you already contacted IBM?

    ------------------------------
    SangGyu Jeong
    Software Engineer
    Infrasoft
    Seoul Korea, Republic of
    ------------------------------