Qpid
  1. Qpid
  2. QPID-3820

ProducerFlowControlTest#testSendTimeout test fails sprodically on 0-10 profiles

    Details

    • Type: Bug Bug
    • Status: Closed
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: 0.15
    • Fix Version/s: 0.15
    • Component/s: Java Broker, Java Tests
    • Labels:
      None

      Description

      On the 0-10 profiles, we see ProducerFlowControlTest#testSendTimeout occasionally fail with the following stack trace:

      org.apache.qpid.server.queue.ProducerFlowControlTest.testSendTimeout
      Failing for the past 1 build (Since Failed#47 )
      Took 10 sec.
      Error Message
      
      No timeout exception on sending
      
      Stacktrace
      
      junit.framework.AssertionFailedError: No timeout exception on sending
      	at org.apache.qpid.server.queue.ProducerFlowControlTest.testSendTimeout(ProducerFlowControlTest.java:344)
      	at org.apache.qpid.test.utils.QpidBrokerTestCase.runBare(QpidBrokerTestCase.java:237)
      	at org.apache.qpid.test.utils.QpidTestCase.run(QpidTestCase.java:138)
      
      

        Activity

        Hide
        Keith Wall added a comment - - edited

        Whilst I have not been able to successfully reproduce the testSendTimeout issue reported above, I can say that the whole test suite looks very brittle: it relies heavily on Thread.sleep() in the hope that sufficient time will have passed to allow observations to be made (exception to arrive etc, messages delivery to be stopped) etc. The Thread.sleep() approach also means the testcase is very slow 120s..160s on my box. I've seen a couple of other PSFC tests fail during the course of the last few hours. It would be best if this suite were rewritten.

        Show
        Keith Wall added a comment - - edited Whilst I have not been able to successfully reproduce the testSendTimeout issue reported above, I can say that the whole test suite looks very brittle: it relies heavily on Thread.sleep() in the hope that sufficient time will have passed to allow observations to be made (exception to arrive etc, messages delivery to be stopped) etc. The Thread.sleep() approach also means the testcase is very slow 120s..160s on my box. I've seen a couple of other PSFC tests fail during the course of the last few hours. It would be best if this suite were rewritten.
        Hide
        Keith Wall added a comment -

        Patch applied to resolve defect in testcase. Robbie, can you review the change please?

        Show
        Keith Wall added a comment - Patch applied to resolve defect in testcase. Robbie, can you review the change please?
        Hide
        Robbie Gemmell added a comment -

        Looks good to me.

        Show
        Robbie Gemmell added a comment - Looks good to me.

          People

          • Assignee:
            Robbie Gemmell
            Reporter:
            Keith Wall
          • Votes:
            0 Vote for this issue
            Watchers:
            0 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development