0 Replies Latest reply: Nov 26, 2012 6:50 AM by Kristian Schmidt RSS

    Reload tasks fail randomly for the same document

    Kristian Schmidt

      I'm having weird issues with publisher (QV11 SR2). Documents fail to reload on irregular intervals. The errors do not seem to follow a pattern as the same file was reloaded without problems a few hours earlier and a few hours later. It's not only one file either, it's quite a lot of them.

       

      Here is an excerpt from the log with the actual error message with stack trace (I've attached the entire log as a zip):

       

      23.11.2012 14:41:39.5810926    Error    The task "1497-214" failed. Exception:
      QDSMain.Exceptions.DistributionFailedException: Distribute failed with errors to follow. ---> QDSMain.Exceptions.ReloadFailedException: Reload failed ---> QDSMain.Exceptions.LogBucketErrorException: The Source Document was NOT reloaded successfully. DocumentPath=C:\Distribution\Standard\Standard-214.qvw. 
         at QDSMain.ReloadTask.VerifyConditions(TaskResult taskResult)
         at QDSMain.ReloadTask.Reload(String fileName, TaskResult taskResult, String sectionAccessUserName, String sectionAccessPassword, eReloadOptions reloadOption, String variableName, String variableValue, Boolean moniterCpuUsage)
         --- End of inner exception stack trace ---
         at QDSMain.ReloadTask.Reload(String fileName, TaskResult taskResult, String sectionAccessUserName, String sectionAccessPassword, eReloadOptions reloadOption, String variableName, String variableValue, Boolean moniterCpuUsage)
         at QDSMain.DistributeTask.Execute(TaskResult currentTaskResult)
         --- End of inner exception stack trace ---
         at QDSMain.DistributeTask.Execute(TaskResult currentTaskResult)
         at QDSMain.Task.AbstractTask.TaskExecution(ILogBucket logBucket, TaskResult taskResult)
      23.11.2012 14:41:42.7791746    Information    The task "40049-214" finished successfully
      23.11.2012 14:41:42.9819798    Information    The task "1205-214" finished successfully
      23.11.2012 14:41:45.2908390    Information    The task "40115-214" finished successfully
      23.11.2012 14:41:50.0177602    Information    The task "1435-214" finished successfully
      23.11.2012 14:41:50.0645614    Information    The task "40175-214" finished successfully
      

       

      We are running multiple tasks on the same document (DocumentID 214 in our system) so the document itself is OK. As you can see from the logs it reloads correctly for customers 40049,1205,40115,1435 and 40175, but not for 1497.

      Now, earlier today (at 11.08.11 o'clock) it reloaded just fine for customer 1497 as well.

       

      I suspected it might have something to do with the QlikviewDirectoryServiceConnector service as this seems to have a history of instability both according to our operations guys and QlikTech themselves, so I stopped the service to test. The result was that all tasks failed with the above exception.

       

      We are using Configurable ODBC directory connetor to authenticate users and I've tried experimenting with the cache expiry time and Service timout, but to no avail...

       

      I'd like to note that the server is a simple test server which is running at 100% CPU during reloads. A document might have 400 reload tasks scheduled to run once a day with 10min intervals between each task. It is using IIS and all services are running on the same machine.

       

      Any suggestions are welcome!

       

      Message was edited by: kschmidt