IBM Support

LI72282: DB2 PANICS WHEN EAGAIN (FILE SHARING VIOLATION) IS RECEIVED FROM SQLOSEEKWRITE64()

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • Note that these symptoms are very similar to those of IY85772.
    .
    During high workload, DB2 UDB may sporadically receive the
    EAGAIN error from the operating system when writing data to
    disk. This typically happens due to filesystem contention
    related to tablespace container files. Under the current logic
    DB2 will panic, i.e. shut down the instance programmatically,
    as such a disk write problem is considered serious.
    .
    In reality, EAGAIN during disk write is often a non-serious
    intermittent problem. The purpose of this APAR is to add retry
    logic into DB2, so that every time we run into this scenario
    DB2 will try to recover, preventing the engine shutdown.
    .
    The problem may manifest itself by various symptoms. The key
    is to look for file sharing violations from sqloseekwrite64().
    .
    Example:
    .
    db2diag.log
    ===========
    2007-04-03-07.20.49.865314+240 I3381743G710       LEVEL: Error
    PID     : 12168                TID  : 3086034624  PROC : db2pcln
    INSTANCE: db2inst1             NODE : 000
    FUNCTION: DB2 UDB, os services, sqloseekwrite64, probe: 100
    MESSAGE : ZRC=0x870F0016=-2029060074=SQLO_SHAR "File sharing vio
              DIA8519C A file sharing violation has occurred
    .
    t12168.000
    ==========
    sqlbLogErr + 0x5e
    sqlbClnrAsyncWriteCompletion + 0x77c
    SQLO_LIO_HANDLE_DATA::sqloLioDeQueueNBlocks + 0xba
    sqloLioCollectNBlocks + 0x4a
    sqlbClnrCollectSomeAIO + 0x88
    sqlbClnrDispatchSomeAIO + 0xcc
    sqlbClnrWriteBatchIOList + 0x9b
    sqlbClnrWritePages + 0x8e
    sqlbClnrFindWork + 0x8cb
    sqlbClnrEntryPoint + 0x7b
    sqloCreateEDU + 0x28e
    

Local fix

  • Reduce the filesystem contention, if possible. For example, if
    the EAGAIN error occurs when writing into a tablespace
    container, add new containers to the tablespace and redestribute
    the tablespace data evenly across all the containers.
    

Problem summary

  • Users affected: see PMR text
    Problem description: see PMR text
    Problem summary: see PMR text
    

Problem conclusion

  • First fixed in DB2 UDB Version 8.1, FixPak 16
    

Temporary fix

  • see PMR text
    

Comments

APAR Information

  • APAR number

    LI72282

  • Reported component name

    DB2 UDB WSE LIN

  • Reported component ID

    5765F3504

  • Reported release

    820

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2007-04-19

  • Closed date

    2008-02-13

  • Last modified date

    2008-02-13

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

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

    LI72283

Modules/Macros

  • ENGN_SQO
    

Fix information

  • Fixed component name

    DB2 UDB WSE LIN

  • Fixed component ID

    5765F3504

Applicable component levels

  • R810 PSN

       UP

[{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSEPGG","label":"DB2 for Linux- UNIX and Windows"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"820","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Document Information

Modified date:
16 October 2021