IBM Support

IC78410: DB2 INSTANCE CRASHED WITH SIGSEGV calling aicTaskProcessor

Subscribe

You can track all active APARs for this component.

APAR status

  • Closed as fixed if next.

Error description

  • This problem occurs when a reorg operation on altered
    column is performed on a partitioned table which has outstanding
    asynchronized Index Cleaning (AIC) jobs. DB2 instance crashed
    with the following in db2diag.log:
    
    2011-05-09-00.15.50.106430-300 E8796E1368         LEVEL: Severe
    PID     : 21982                TID  : 47712760678304PROC :
    db2taskp (TWHPR1) 0
    INSTANCE: twhpr00              NODE : 000         DB   : TWHPR1
    APPHDL  : 0-20                 APPID: *LOCAL.DB2.110509051557
    AUTHID  : TWHPR00
    FUNCTION: DB2 UDB, SQO Memory Management,
    sqloDiagnoseFreeBlockFailure, probe:999
    MESSAGE : Memory validation failure, diagnostic file dumped.
    DATA #1 : String, 28 bytes
    Corrupt pool free tree node.
    DATA #2 : File name, 24 bytes
    p21982t0_diagnostics.mem
    CALLSTCK:
      [0] 0x00002B64F715C0DF
    _ZN8SMemPool32diagnoseMemoryCorruptionAndCrashEmPKc + 0x1EF
      [1] 0x00002B64F8466AE3
    _ZN8SMemPool10MemTreeGetEmmPPvPP17SqloChunkSubgroupPP12SMemLostN
    ode
    + 0x57B
      [2] 0x00002B64F84672E6
    _ZN8SMemPool19allocateMemoryBlockEmmmjmPPvPP17SqloChunkSubgroupP
    P12SMemLostNodeP12SMemLogEvent
    + 0xCE
      [3] 0x00002B64F84653BE sqlogmblkEx + 0x3DE
      [4] 0x00002B64F6B9B6DA
    _Z18sqliGetKeyWorkAreaP17SQLI_KEY_WORKAREAP10SQLI_IDXCBm + 0x10A
      [5] 0x00002B64F7E2B7B3 _Z20sqliReorgCleanupKeysP7SQLI_CB +
    0x2C3
      [6] 0x00002B64F7E833E7
    _Z21sqliAsyncIndexCleanupP20sqle_agent_privatecbP6AIC_CBm +
    0x1B7
      [7] 0x00002B64F826E971 aicTaskProcessor + 0x1B9
      [8] 0x00002B64F8262A6D
    _ZN8ABPAgent19taskProcessorDriverEP17ABPTaskProContext + 0x1ED
      [9] 0x00002B64F8262EA9 _ZN8ABPAgent4mainEP17ABPTaskProContext
    + 0xC9
    

Local fix

  • When encountering the problem, after crash, use the following
    to by pass AIC in crash recovery:
    1) set DB2_DISABLE_ABP=yes
    2) drop index and recreate the index
    3) unset DB2_DISABLE_ABP
    
    Use the following to avoid the problem:
    Check if there are outstanding index cleanup jobs from SYSTASKS
    before ALTER a column.
    

Problem summary

  • Users Effected:
    This problem effects all users that issue reorg command on DPF
    system.
    
    Problem Description:
    This problem occurs when a reorg operation on a altered column
    is performed on a partitioned table which has outstanding
    asynchronized Index Cleaning (AIC) jobs. DB2 instance crashed.
    
    Problem Summary:
    DB2 INSTANCE CRASHED WITH SIGSEGV CALLING aicTaskProcessor.
    

Problem conclusion

Temporary fix

  • Upgrade to DB2 Version Version 9.7, FixPak 6.
    

Comments

APAR Information

  • APAR number

    IC78410

  • Reported component name

    DB2 FOR LUW

  • Reported component ID

    DB2FORLUW

  • Reported release

    970

  • Status

    CLOSED FIN

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2011-08-30

  • Closed date

    2013-01-17

  • Last modified date

    2013-01-17

  • APAR is sysrouted FROM one or more of the following:

    IC77128

  • APAR is sysrouted TO one or more of the following:

Fix information

Applicable component levels

  • R970 PSY

       UP



Document information

More support for: DB2 for Linux, UNIX and Windows

Software version: 9.7

Reference #: IC78410

Modified date: 17 January 2013