Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-8273

java.io.EOFException when running AccuRev streams command

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: Resolved (View Workflow)
    • Priority: Major
    • Resolution: Fixed
    • Component/s: accurev-plugin
    • Labels:
      None
    • Environment:
      Windows XP (32bit) and/or Windows 7 (64bit)
      AccuRev plugin 0.6.12-SNAPSHOT (private-10/27/2010 12:41-root)
    • Similar Issues:

      Description

      Sometimes, when a build (whose source code is stored in AccuRev) starts and the AccuRev plugin starts talking to the AccuRev server, one gets the following error:

      Started by user anonymous
      Building remotely on uk-hudson-slave03
      Purging workspace...
      Workspace purged.
      Authenticating with Accurev server...
      [myHudsonProject] $ "C:\Program Files\AccuRev\bin\accurev.exe" login -H accurev:5050 Build ********
      Authentication completed successfully.
      Getting a list of streams...
      [myHudsonProject] $ "C:\Program Files\AccuRev\bin\accurev.exe" show -H accurev:5050 -fx -p MyDepot_ streams
      java.io.EOFException: no more data available - expected end tags </stream></streams> to close start tag <stream> from line 4187 and start tag <streams> from line 1, parser stopped on TEXT seen ...    depotName="MyDepot_"\r\n      streamNumber="4931"\r\n      isDynami... @4193:15
      	at org.xmlpull.mxp1.MXParser.fillBuf(MXParser.java:3035)
      	at org.xmlpull.mxp1.MXParser.more(MXParser.java:3046)
      	at org.xmlpull.mxp1.MXParser.parseAttribute(MXParser.java:1944)
      	at org.xmlpull.mxp1.MXParser.parseStartTag(MXParser.java:1799)
      	at org.xmlpull.mxp1.MXParser.nextImpl(MXParser.java:1127)
      	at org.xmlpull.mxp1.MXParser.next(MXParser.java:1093)
      	at hudson.plugins.accurev.AccurevSCM.getStreams(AccurevSCM.java:807)
      	at hudson.plugins.accurev.AccurevSCM.checkout(AccurevSCM.java:307)
      	at hudson.model.AbstractProject.checkout(AbstractProject.java:1119)
      	at hudson.model.AbstractBuild$AbstractRunner.checkout(AbstractBuild.java:479)
      	at hudson.model.AbstractBuild$AbstractRunner.run(AbstractBuild.java:411)
      	at hudson.model.Run.run(Run.java:1324)
      	at hudson.model.FreeStyleBuild.run(FreeStyleBuild.java:46)
      	at hudson.model.ResourceController.execute(ResourceController.java:88)
      	at hudson.model.Executor.run(Executor.java:139)
      

      When this happens, the subsequent SCM retries in the immediate aftermath typically fail as well.
      If one waits a while before retrying, all is well (but by this time, we've had a build failure reported).

      i.e. it appears to be a spurious, unpredictable failure.
      I can't be certain it's not the AccuRev server, but the only sure-fire way of finding out would be if the plugin dumped the XML it got from the AccuRev command into a file and then logged that on failure.

        Attachments

          Issue Links

            Activity

            Hide
            pjdarton pjdarton added a comment -

            JENKINS-8742 suggests that this is a Jenkins issue, not an AccuRev issue.
            This may very well be the same issue that's causing code-coverage plugins to fail to read coverage data from slaves.

            Show
            pjdarton pjdarton added a comment - JENKINS-8742 suggests that this is a Jenkins issue, not an AccuRev issue. This may very well be the same issue that's causing code-coverage plugins to fail to read coverage data from slaves.
            Hide
            pjdarton pjdarton added a comment -

            Since upgrading to a later version of Jenkins (1.409) which has the JENKINS-7951 / JENKINS-7836 fix in it, I've not seen this problem again.

            I believe that this issue was in the Jenkins core to do with master/slave comms, and fixing JENKINS-7951 / JENKINS-7836 fixed this one.

            I would suggest that this can be closed.

            Show
            pjdarton pjdarton added a comment - Since upgrading to a later version of Jenkins (1.409) which has the JENKINS-7951 / JENKINS-7836 fix in it, I've not seen this problem again. I believe that this issue was in the Jenkins core to do with master/slave comms, and fixing JENKINS-7951 / JENKINS-7836 fixed this one. I would suggest that this can be closed.
            Hide
            pjdarton pjdarton added a comment -

            Haven't seen this again. I think it's fixed.

            Show
            pjdarton pjdarton added a comment - Haven't seen this again. I think it's fixed.

              People

              Assignee:
              statler Scott Tatum
              Reporter:
              pjdarton pjdarton
              Votes:
              1 Vote for this issue
              Watchers:
              3 Start watching this issue

                Dates

                Created:
                Updated:
                Resolved: