Forum Discussion

vatbub's avatar
vatbub
Contributor
8 years ago
Solved

TestComplete ExitCode -805306369

Hey guys,

today, I logged in into our CI and saw that our tests were running since 157 hours (=6 days). So I logged in to the PC running TestComplete and saw that it was basically doing nothing, it opened the ProjectSuite and then nothing happened, it did not run anything. So I clicked on "Stop Test Run" and it saved the log and exited, but the CI told me that the Exit Code was -805306369. Unfortunately, this Exit Code is not documented here.

 

The testlog itself told me that it actually did something, like installing the app and running the tests except that the test device did not show anything of that :)

 

Does anyone know something about that exit code? It would be a great help.

Regards,

vat

  • vatbub's avatar
    vatbub
    8 years ago

    Hey guys,

    I just found out why this issue happens. As far as I understood it, TestComplete gets stuck for some reason and TeamCity is waiting for it endlessly and eventually ends up having no more resources in its JVM.

     

    Now that "some reason" seems to be a iOS TableViewCell. It takes TestComplete ages to find that cell (I already discussed that issue here) and apparently gets stuck sometimes on that. I'll investigate more tomorrow, until then, have a good night,

    vat

12 Replies

Sort By
  • Colin_McCrae's avatar
    Colin_McCrae
    Community Hero
    8 years ago

    Are you running against something using Java? (Or is something on the CI box using Java?)

     

    All mentions I can find about that error code seem to be related to Java JVM's?

  • sanjay0288's avatar
    sanjay0288
    Frequent Contributor
    8 years ago

    This exit code is related to out of memory exception. Can you confirm that the application was working fine in the previous releases? If so it could be a problem with the application component/corrupted workspace / JVM not being allocated enough memory. 

     

    If you have access to dev team you can double check with them to ensure the problem is not from the application side.

    • tristaanogre's avatar
      tristaanogre
      Esteemed Contributor
      8 years ago


      An out of resource error actually makes sense on running a test for 6 days, especially if TC/TE is writing out a bunch of logs.  

      Just a curiousity question: is it normal for your tests to take a long time to run?  I'd actually be concerned after 1 day if I haven't received a notification that a test run is complete.

      • vatbub's avatar
        vatbub
        Contributor
        8 years ago

        No, usually our tests take 15 Minutes, but I guess that something happened right at the beginning and I simply didn't see it as I did not log in into the CI for 6 days.

    • vatbub's avatar
      vatbub
      Contributor
      8 years ago

      Our CI is TeamCity, which is Java based, but it runs TestComplete by executing a cmd-Script in the Winows command line, so Java should not matter, in my opinion. Our Application under test is actually a Android App and we use the Visual Studio Emulator for Android to run it.

       

      And we use TestComplete 11, latest release btw

      • Colin_McCrae's avatar
        Colin_McCrae
        Community Hero
        8 years ago
        vatbub wrote:

        Our CI is TeamCity, which is Java based, but it runs TestComplete by executing a cmd-Script in the Winows command line, so Java should not matter, in my opinion.

        Well, if TeamCity is invoking TC/TE, and is Java based, and was trying to start/run/do-something with TC/TE for 6 days, I'd say there is a VERY good chance that a JVM involved could well run out of memory!

         

        After 6 days of trying, and failing, to run, it's hardly surprising that it didn't work if you just jumped in manually and tried to run it as is. Did you clear out an reset everything? All the Java pages talking about that error talked about clearing down the JVM causing the problem.

         

        But again, I'm not convinced this is really a TC/TE problem. Sounds more like the setup running it has got itself tied in knots. Clean that up first ....