IBM Support

PK70945: OOM ERRORS BEING THROWN WHEN PROCESSING SCHEDULED ACTIONS

 

APAR status

  • Closed as program error.

Error description

  • OOM error occurs with the following Heapdump stack:
    ===================================================
    577,416,888 [96] 9 com/ibm/icm/ci/data/impl/Content 0x109578d0
    577,412,640 [80] 9 com/ibm/icm/ci/Ticket 0x14b84010
    577,410,976 [24] 1 com/ibm/icm/ci/Ticket$TicketCallback
    0x14b82480
    577,410,952 [24] 2
    com/ibm/icm/jcr/service/access/WPSIACLFilterImpl
    577,410,928 [96] 11 com/ibm/icm/jcr/WorkspaceImpl 0x14b8d688
    575,247,976 [56] 1 java/util/HashMap 0x14b825d8
    575,247,920 [524,304] 42,336 array of java/util/HashMap$Entry
    0x6da82880
    695,528 [32] 3 java/util/HashMap$Entry 0x3b96acb0
    534,216 [32] 3 java/util/HashMap$Entry 0x302c4350
    393,848 [32] 3 java/util/HashMap$Entry 0x43838fc8
    140,184 [192] 14 com/ibm/icm/jcr/NodeImpl 0x302d5e80
    152 [32] 1 java/lang/String 0x302c4430
    161,128 [192] 15 com/ibm/icm/jcr/NodeImpl 0x3b9686d0
    152 [32] 1 java/lang/String 0x3b9698c8
    617,680 [32] 3 java/util/HashMap$Entry 0x57bbf818
    582,912 [32] 3 java/util/HashMap$Entry 0x377bbbb8
    558,000 [32] 3 java/util/HashMap$Entry 0x41bd8be0
    541,896 [32] 3 java/util/HashMap$Entry 0x39433aa0
    539,376 [32] 3 java/util/HashMap$Entry 0x36d20040
    524,912 [32] 3 java/util/HashMap$Entry 0x19f364c8
    513,192 [32] 3 java/util/HashMap$Entry 0x63947040
    472,816 [32] 3 java/util/HashMap$Entry 0x30c68f68
    
    Another symptom seen here was a relative slowness in loading
    pages whenever the WCM_EJB task was running and processing a
    scheduled action change.
    
    A lot of memory was held by the JCR Workspace
    which was used while processing scheduled actions. The fix
    decreases the number of JCR ops to 10 before clearing the
    workspace
    

Local fix

Problem summary

  • OOM errors were thrown when  a high number of content items were
    processed as a result of an action change or a scheduled move
    
    
    Here is the thread stack that showed that scheduled move actions
    were being processed :
    
    4XESTACKTRACE          at
    com.ibm.workplace.wcm.services.content.ContentServiceImpl.getImp
    l(ContentServiceImpl.java(Compiled Code))
    4XESTACKTRACE          at
    com.ibm.workplace.wcm.services.content.ContentServiceImpl.getNon
    Draft(ContentServiceImpl.java(Compiled Code))
    4XESTACKTRACE          at
    com.ibm.workplace.wcm.services.content.ContentServiceImpl.prepar
    eSaveSteps(ContentServiceImpl.java(Compiled Code))
    4XESTACKTRACE          at
    com.ibm.workplace.wcm.services.content.ContentServiceImpl.save(C
    ontentServiceImpl.java(Compiled Code))
    4XESTACKTRACE          at
    com.ibm.workplace.wcm.services.content.ContentServiceImpl.update
    (ContentServiceImpl.java(Compiled Code))
    4XESTACKTRACE          at
    com.ibm.workplace.wcm.services.content.ContentServiceImpl.update
    (ContentServiceImpl.java(Compiled Code))
    4XESTACKTRACE          at
    com.ibm.workplace.wcm.services.content.ContentServiceImpl.update
    (ContentServiceImpl.java(Compiled Code))
    4XESTACKTRACE          at
    com.aptrix.pluto.workflow.actions.ActionTracker.rescheduleAction
    (ActionTracker.java(Compiled Code))
    4XESTACKTRACE          at
    com.aptrix.pluto.workflow.actions.ScheduledActionSchedulable.pro
    cessItem(ScheduledActionSchedulable.java(Compiled Code))
    4XESTACKTRACE          at
    com.aptrix.pluto.workflow.actions.ScheduledActionSchedulable.pro
    cessItems(ScheduledActionSchedulable.java(Compiled Code))
    4XESTACKTRACE          at
    com.aptrix.pluto.workflow.actions.ScheduledActionSchedulable.run
    (ScheduledActionSchedulable.java:181)
    4XESTACKTRACE          at
    com.ibm.wps.services.workmanager.impl.WasWorkWrapper.run(WasWork
    Wrapper.java:41)
    

Problem conclusion

  • The JCR strong workspace state holds onto memory while all the
    content items are being processed. It was being cleared
    periodically but still
    was holding onto a significant amount of memory.
    
    
    A config option has been added to specify the number of JCR
    operations after which the JCR workspace will be cleared. Here
    is an example on
    how to specify the config option in the
    WCMConfigServices.propoerties file :
    
    max.jcr.ops=10
    
    A fix for this problem is available for WCM 6.0.1.3 from Fix
    Central at:
    
    http://www.ibm.com/eserver/support/fixes/fixcentral/swgquickorde
    r?brandid=2&productid=Workplace%20Web%20Content%20Management&fix
    es=6.0.1.3-WCM-PK70945
    

Temporary fix

Comments

APAR Information

  • APAR number

    PK70945

  • Reported component name

    LOTUS WEB CONT

  • Reported component ID

    5724I2901

  • Reported release

    60G

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2008-08-21

  • Closed date

    2008-08-22

  • Last modified date

    2008-12-10

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

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

Fix information

  • Fixed component name

    LOTUS WEB CONT

  • Fixed component ID

    5724I2901

Applicable component levels

  • R60C PSY

       UP

[{"Business Unit":{"code":null,"label":null},"Product":{"code":"SUPPORT","label":"IBM Worldwide Support"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"6.0.1.3","Edition":"","Line of Business":{"code":null,"label":null}}]

Document Information

Modified date:
09 September 2020