IT08757: EXTERNAL PURGE ERROR WHILE GENERATING DOCUMENT ON DISK LIST JAVA.LANG.ILLEGALSTATEEXCEPTION: QUEUE FULL.

APAR status

• Closed as program error.

Error description

• External purge returns the following error in purge logs:
  
  [2015-04-09 11:40:02.848] ERROR 000000000000 GLOBAL_SCOPE
  [ConsumerTask] PURGE ERROR While generating Document On Disk
  List
  java.lang.IllegalStateException: Queue full
      at java.util.AbstractQueue.add(AbstractQueue.java:109)
      at
  com.sterlingcommerce.woodstock.services.archive.purge.hpp.AsyncD
  odFileDelete.deferredDeleteFile(AsyncDodFileDelete.java:48)
      at
  com.sterlingcommerce.woodstock.services.archive.purge.hpp.Consum
  erTask.writeDataOnDiskInfo(ConsumerTask.java:452)
      at
  com.sterlingcommerce.woodstock.services.archive.purge.hpp.Consum
  erTask.taskLooper(ConsumerTask.java:731)
      at
  com.sterlingcommerce.woodstock.services.archive.purge.hpp.Consum
  erTask.run(ConsumerTask.java:678)
      at
  java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExec
  utor.java:1156)
      at
  java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExe
  cutor.java:626)
      at java.lang.Thread.run(Thread.java:804)
  [2015-04-09 11:40:02.849] ERROR The following error occurred:
  Queue full
  null
  [2015-04-09 11:40:02.849] ERROR Attempting to recover
  [2015-04-09 11:40:02.849] ERROR
  
  The Schedule_PurgeService is unable to keep up with the
  workload. External purge is not deleting documents on disk.
  
  Tried tuning external purge properties in
  archivethread.properties_hpp_ext file:
  
  purge_maxPoolSize = 10
  purge_keepAliveTime = 30000
  purge_maxQueueDepth = 100000
  purge_debug=true
  purge_scavengerSleepTime=20000
  purge_batchSize=500
  
  No impact. Still getting "queue full" error.
  
  Customer requires external purge to maintain database size and
  delete documents on disk. Schedule_PurgeService is unable to
  keep up with workload.
  

Local fix

• STRRTC - 462556
  PC / PC
  Circumvention: None
  

Problem summary

• Problem Description:
  A capacity-restricted queue is used to store deferred delete
  file requests, but the capacity is hard-coded. The number of
  threads allocated for these deferred delete file requests is
  also hard-coded. This makes it impossible for the customer to
  configure these two values to avoid a queue full error.
  
  Platforms Affected:
  All
  
  Users Affected:
  External purge users
  

Problem conclusion

• Resolution Summary:
  Two new properties have been added to archivethread.properties
  to give customers more visibility and control over external
  purge configurations. The "purge_taskListCapacity" property can
  be used to set the capacity of the deferred delete file requests
  queue. The "purge_numberDeleteThreads" property can be used to
  set the number of threads allocated for processing deferred
  delete file requests.
  
  Delivered In:
  5020500_5
  

Temporary fix

Comments

APAR Information

• 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

  STR B2B INTEGRA

• Fixed component ID

  5725D0600

Applicable component levels

• R525 PSY

     UP