DX Application Performance Management

 View Only
  • 1.  APM 10.1 :: BUGs Fix

    Posted Mar 23, 2016 03:05 PM

    Hi All experts,

     

    I have 3 (three) customers in Brazil that ask us for update CA ISCP 9.7.1 to CA APM 10.1.

     

    In this 3 customers, we got errors with performance and estability of the tool. 1 month after upgrade, the CA ISCP does'nt work correctly and any errors appers in LOGs like:

     

    3/23/16 03:02:42.678 PM BRT [ERROR] [Thread-70] [Manager] Caught Exception while passivating bean com.ca.apm.transactiontrace.appmap.mapper.beans.MOMTransactionTraceAppMapProcessorBean@4f89fc96

     

    3/23/16 02:44:25.053 PM BRT [ERROR] [pool-7-thread-8] [Manager] Uncaught Exception in Enterprise Manager:  In thread pool-7-thread-8 and the message is com.wily.util.exception.UnexpectedExceptionError: Invalid Registry Path Exception

     

    3/23/16 02:43:48.238 PM BRT [ERROR] [MOM Collection Buffer consumer thread.] [Manager.Cluster] Exception: java.lang.NullPointerException calling method: com.wily.isengard.messageprimitives.service.CompressedMessageServiceCallMessage: {com.wily.introscope.spec.server.beans.event.IEventManager.getEventData, v1, [1458754725705:59974]}

     

    3/23/16 03:54:25.032 PM BRT [WARN] [Harvest Engine Pooled Worker] [Manager.MetricCalculatorBean] To prevent a JavaScript calculator from running in the MOM, the JavaScript should implement a runOnMOM function that returns false

     

    3/23/16 02:51:34.119 PM BRT [WARN] [MOM Collection Buffer consumer thread.] [Manager] MOM Collection Buffer: buffer consumer thread interrupted.

    3/23/16 02:51:34.121 PM BRT [WARN] [MOM Persistence Buffer consumer thread.] [Manager] MOM Persistence Buffer: buffer consumer thread interrupted.

    3/23/16 04:03:27.477 PM BRT [WARN] [Dispatcher 1] [Manager.IsengardObjectInputStream] Invalid control byte: 0

    3/23/16 04:03:27.767 PM BRT [WARN] [Dispatcher 1] [Manager.IsengardObjectInputStream] Invalid class code: 74

     

    Anyway... Somenone have the same difficulty in this upgrade?

     

    ps.: We have P1 issue at ca support for these cases.

     

    Regards!



  • 2.  Re: APM 10.1 :: BUGs Fix

    Broadcom Employee
    Posted Mar 23, 2016 03:15 PM

    For the calculator, see this page: JavaScript Calculators



  • 3.  Re: APM 10.1 :: BUGs Fix

    Posted Mar 23, 2016 03:40 PM

    Hi Haruhiko,

    tks for your quick replay. I'm just saw this page. But our problem is performance after upgrade.

    I have the same performance issue in 3 customers with 3 different environment....

    []s



  • 4.  Re: APM 10.1 :: BUGs Fix

    Broadcom Employee
    Posted Mar 23, 2016 03:44 PM

    Hi Rafael:

    Given this sounds involved, and not a simple answer, I would open a case to investigate

    Thanks

    Hal German



  • 5.  Re: APM 10.1 :: BUGs Fix

    Posted Jun 15, 2016 01:31 PM

    Hi everyone,

     

    Today I also found the same ERROR message when a collector was consuming a lot of CPU and looked like if it were Hanged.

     

    6/15/16 12:03:22.552 PM CDT [INFO] [Thread-5449] [Manager.EMWebServer] EM Web Server shutdown complete.

    6/15/16 12:03:23.029 PM CDT [INFO] [Thread-5449] [Manager] Shutting down data persistence subsystem

    6/15/16 12:03:24.445 PM CDT [ERROR] [PO:main Mailman 4] [Manager.MessageService] Exception: java.lang.NullPointerException calling method: com.wily.isengard.messageprimitives.service.CompressedMessageServiceCallMessage: {com.wily.introscope.spec.server.beans.event.IEventManager.getEventData, v1, [1466008103755:88743]}

    6/15/16 12:03:43.089 PM CDT [INFO] [Thread-5449] [Manager] Shutting down the Isengard server

    6/15/16 12:03:43.174 PM CDT [INFO] [Thread-5449] [Manager.com.wily.apm.tess.isengard.BizDefUpdatesPublisherServiceBean] Passivated BizDefUpdatesPublisherServiceBean

     

    Thanks and regards.



  • 6.  Re: APM 10.1 :: BUGs Fix

    Broadcom Employee
    Posted Jun 15, 2016 01:35 PM

    Since your problem does not have to do with a JavaScript calculator, please open a new discussion after you've done a search to see if this hasn't been already addressed.