Tracking down internal error

2006-10-17
2013-03-15
  • atthecorner
    atthecorner
    2006-10-17

    When I use pydev and extensions  w/ eclipse 3.2 on windows-xp. I keep encountering a pop-up saying an internal error has occurred. I restart eclipse to solve the problem right now. I suspect as a new user of both that I am hitting a key combination that hasn't seen much action so far. It happens when I use the debugger.
    It seems to be independent of the code I happen to be looking at; it has happened in unrelated modules.

    What can I do that might help someone isolate this problem?

     
    • Fabio Zadrozny
      Fabio Zadrozny
      2006-10-17

      Do you have something in your error log?

       
      • atthecorner
        atthecorner
        2006-10-17

        Error log? That would be a good place to look.
        If by error log you mean the responses on the console, then all I see is

        "pydev debugger" followed by the arguments

        after that zilch...

        Should I be looking elsewhere?

         
        • Fabio Zadrozny
          Fabio Zadrozny
          2006-10-18

          Check http://pydev.sourceforge.net/faq.html#ref_0 for details on how to get to the error log.

           
          • atthecorner
            atthecorner
            2006-10-18

            Thanks,
               Next time it happens which should be Wednesday
            at the current frequency, I'll post what I find.
            -m

             
    • atthecorner
      atthecorner
      2006-10-17

      More detail...
      The exact message is: "An internal error occurred during step over"

      Symptoms of the problem have already appeared though. I believe, the first symptom is that the window doesn't open to the point of the first breakpoint. This is why I try stepping and ultimately encounter the step over error.

      As an aside, I find that I often begin editing a source as soon as I see a problem that can be fixed. Only later do I have the presence of mind to terminate the running code. I probably shouldn't do that. However, I noticed several times today that I was able to run debug, edit, terminate, rerun debug without a problem. So at best that editing style is necessary but not sufficient to cause the error.

      Can pydev detect when a code is being changed and just terminate execution automatically?

      Other debuggers I have seen recently have a start
      option in debugging. This sets a breakpoint on the  first line of execution, starts the debugger. Consequently, it stops immediately. But I found that to be useful data point.

       
    • atthecorner
      atthecorner
      2006-10-19

      I see that the error file accumulates. The latest showing the error records:

      !ENTRY org.eclipse.ui 4 4 2006-10-19 13:29:31.937
      !MESSAGE Plugin com.skwash, extension org.eclipse.ui.newWizards
      Required attribute 'class' not defined
      !SESSION 2006-10-19 15:45:08.062 -----------------------------------------------
      eclipse.buildId=M20060629-1905
      java.version=1.5.0_06
      java.vendor=Sun Microsystems Inc.
      BootLoader constants: OS=win32, ARCH=x86, WS=win32, NL=en_US
      Command-line arguments:  -os win32 -ws win32 -arch x86

      !ENTRY org.python.pydev 4 4 2006-10-19 16:59:33.390
      !MESSAGE Terminating: No debugger in target when resuming.
      !STACK 1
      org.eclipse.core.runtime.CoreException: Terminating: No debugger in target when resuming.
          at org.python.pydev.plugin.PydevPlugin.log(PydevPlugin.java:384)
          at org.python.pydev.debug.model.PyThread.resume(PyThread.java:121)
          at org.python.pydev.debug.model.PyStackFrame.resume(PyStackFrame.java:234)
          at org.eclipse.debug.internal.ui.actions.context.SuspendResumeAdapter$4.run(SuspendResumeAdapter.java:92)
          at org.eclipse.core.internal.jobs.Worker.run(Worker.java:58)
      org.eclipse.core.runtime.CoreException[4]: java.lang.RuntimeException: Terminating: No debugger in target when resuming.
          at org.python.pydev.plugin.PydevPlugin.log(PydevPlugin.java:383)
          at org.python.pydev.debug.model.PyThread.resume(PyThread.java:121)
          at org.python.pydev.debug.model.PyStackFrame.resume(PyStackFrame.java:234)
          at org.eclipse.debug.internal.ui.actions.context.SuspendResumeAdapter$4.run(SuspendResumeAdapter.java:92)
          at org.eclipse.core.internal.jobs.Worker.run(Worker.java:58)
      !SUBENTRY 1 org.python.pydev 4 4 2006-10-19 16:59:33.390
      !MESSAGE Terminating: No debugger in target when resuming.
      !STACK 0
      java.lang.RuntimeException: Terminating: No debugger in target when resuming.
          at org.python.pydev.plugin.PydevPlugin.log(PydevPlugin.java:383)
          at org.python.pydev.debug.model.PyThread.resume(PyThread.java:121)
          at org.python.pydev.debug.model.PyStackFrame.resume(PyStackFrame.java:234)
          at org.eclipse.debug.internal.ui.actions.context.SuspendResumeAdapter$4.run(SuspendResumeAdapter.java:92)
          at org.eclipse.core.internal.jobs.Worker.run(Worker.java:58)

      !ENTRY org.eclipse.core.jobs 4 2 2006-10-19 16:59:36.046
      !MESSAGE An internal error occurred during: "stepOver".
      !STACK 0
      java.lang.NullPointerException
          at org.python.pydev.debug.model.remote.AbstractDebuggerCommand.<init>(AbstractDebuggerCommand.java:62)
          at org.python.pydev.debug.model.remote.StepCommand.<init>(StepCommand.java:21)
          at org.python.pydev.debug.model.PyThread.stepOver(PyThread.java:163)
          at org.python.pydev.debug.model.PyStackFrame.stepOver(PyStackFrame.java:214)
          at org.eclipse.debug.internal.ui.actions.context.StepAdapter$6.run(StepAdapter.java:146)
          at org.eclipse.core.internal.jobs.Worker.run(Worker.java:58)

       
    • atthecorner
      atthecorner
      2006-11-29

      FYI, I upgraded to 1.2.5. While I am no longer getting the internal error, pydev now only hangs and less  frequently, but still mysteriously.  Bottom-line the problem is only a minor aggravation. I may
      just be getting better at avoiding the key strokes that provoke the problem.

       
      • Fabio Zadrozny
        Fabio Zadrozny
        2006-11-30

        Strange... from the error log it appears that the target is dying but it is not correctly notifying the debugger... can you reproduce that in a short example?