Topic
IC4NOTICE: developerWorks Community will be offline May 29-30, 2015 while we upgrade to the latest version of IBM Connections. For more information, read our upgrade FAQ.
2 replies Latest Post - ‏2009-11-29T14:55:14Z by SystemAdmin
SystemAdmin
SystemAdmin
37422 Posts
ACCEPTED ANSWER

Pinned topic Attempts to flush session locks out websphere application

‏2009-11-29T00:43:15Z |
I am running Websphere 6.1.0.25 with Hibernate 3.2.7 and Spring 2.5. We have a scenario where a browser request to the application is preventing the entire application from processing additional request--this lasts for around 2 minutes, then the application behaves normal and starts processing again. This occurs using different application request so we cannot establish a specific bug in the app causing this issue-- we think its a cacheing / flushing of the session issue but cannot figure it out.

This is running on linux.

Here is the Hibernate traces. We have seen that application die and restart but not everytime.
11/23/09 9:15:03:847 EST 00000051 Printer 1 org.hibernate.pretty.Printer toString com.cleartricity.kermit.opscenter.business.domain.Device{measuredPower=200, ulLabelImage=null, serialNumber=5433-3, year=2009, description=tstat, id=1945, energyUsage=null, activeInd=Y, measuredVoltage=1160, deviceController=com.cleartricity.kermit.opscenter.business.domain.DeviceController#1545, currentTempMode=null, deviceType=com.cleartricity.kermit.opscenter.business.domain.DeviceType#5, devicePower=400, make=unknown, physicalLocation=tstat, prevTempMode=null, model=unknown, deviceVoltage=1200, hvacDeviceId=2, deviceName=tstat, gtwyDeviceId=3}
11/23/09 9:15:03:850 EST 00000051 Printer 1 org.hibernate.pretty.Printer toString com.cleartricity.kermit.opscenter.business.domain.Device{measuredPower=200, ulLabelImage=null, serialNumber=901002, year=2009, description=hvac, id=1944, energyUsage=null, activeInd=Y, measuredVoltage=1160, deviceController=com.cleartricity.kermit.opscenter.business.domain.DeviceController#1544, currentTempMode=com.cleartricity.kermit.opscenter.business.domain.TempMode#2, deviceType=com.cleartricity.kermit.opscenter.business.domain.DeviceType#2, devicePower=400, make=unknown, physicalLocation=hvac, prevTempMode=com.cleartricity.kermit.opscenter.business.domain.TempMode#2, model=unknown, deviceVoltage=1200, hvacDeviceId=0, deviceName=hvac, gtwyDeviceId=2}
11/23/09 9:15:03:852 EST 00000051 Printer 1 org.hibernate.pretty.Printer toString more......
11/23/09 9:15:03:855 EST 00000051 AbstractFlush 3 org.hibernate.event.def.AbstractFlushingEventListener performExecutions executing flush
11/23/09 9:15:03:858 EST 00000051 ConnectionMan 3 org.hibernate.jdbc.ConnectionManager flushBeginning registering flush begin
11/23/09 9:15:03:860 EST 00000051 ConnectionMan 3 org.hibernate.jdbc.ConnectionManager flushEnding registering flush end
11/23/09 9:15:03:863 EST 00000051 AbstractFlush 3 org.hibernate.event.def.AbstractFlushingEventListener postFlush post flush
11/23/09 9:17:51:908 EST 0000000f HAManagerImpl I HMGR0131I: The local member of group IBM_hc=Consert01OpsCenterCluster,WSAF_SIB_BUS=BusCluster1,WSAF_SIB_MESSAGING_ENGINE=Consert01OpsCenterCluster.000-BusCluster1,type=WSAF_SIB has not responded to a health check. The JVM will be terminated.
11/23/09 9:17:51:912 EST 0000000f SystemOut O Panic:isAlive panic
11/23/09 9:17:51:913 EST 0000000f SystemErr R java.lang.Throwable
11/23/09 9:17:51:920 EST 0000000f SystemErr R at java.lang.Thread.dumpStack(Thread.java:441)
11/23/09 9:17:51:921 EST 0000000f SystemErr R at com.ibm.ws.hamanager.coordinator.impl.JVMControllerImpl.panicJVM(JVMControllerImpl.java:49)
11/23/09 9:17:51:921 EST 0000000f SystemErr R at com.ibm.ws.hamanager.impl.HAManagerImpl$IsAliveBucket.sendPings(HAManagerImpl.java:979)
11/23/09 9:17:51:921 EST 0000000f SystemErr R at com.ibm.ws.hamanager.impl.HAManagerImpl$IsAliveManager.alarm(HAManagerImpl.java:767)
11/23/09 9:17:51:921 EST 0000000f SystemErr R at com.ibm.ejs.util.am._Alarm.run(_Alarm.java:90)
11/23/09 9:17:51:922 EST 0000000f SystemErr R at com.ibm.ws.util.ThreadPool$Worker.run(ThreadPool.java:1497)

Notice the approx. 2 minute gap from 9:15:03:863--9:17:51:908 nothing happens in the app- and in fact we get a jvm panic--

We do not have second level caching turned on.

We are stuck-- Any direction anyone can provide would be appreciated.

Thanks
Bill
brodriguez@consert.com
Updated on 2009-11-29T14:55:14Z at 2009-11-29T14:55:14Z by SystemAdmin
  • SystemAdmin
    SystemAdmin
    37422 Posts
    ACCEPTED ANSWER

    Re: Attempts to flush session locks out websphere application

    ‏2009-11-29T02:52:25Z  in response to SystemAdmin
    Bill,

    Sounds like this issue:
    http://www-01.ibm.com/support/docview.wss?rs=180&uid=swg24021309
    http://www-01.ibm.com/support/docview.wss?rs=180&&uid=swg1PK59051

    There are some properties there that you may want to try.

    thanks,
    dims
    • SystemAdmin
      SystemAdmin
      37422 Posts
      ACCEPTED ANSWER

      Re: Attempts to flush session locks out websphere application

      ‏2009-11-29T14:55:14Z  in response to SystemAdmin
      Dims,

      Thanks for the quick reply. We are running 6.1.0.25 version of websphere and the APAR you reference has a potential fix has been delivered in 6.1.0.19 --so we have the fix already installed.

      I did an analysis of all the apars fixed in 6.1.0.27 (we have .25) and there are some potential issues which have been fixed which might solve our issue--we will be upgrading our system to .27 just to be safe.

      Thanks again for the reply!!

      Bill