3 Replies Latest reply on Feb 7, 2002 2:49 PM by segerss

    Core-dump

    segerss

      Our JBoss-process was stopped with a core-dump on our production-machine (AIX 4.3.3). Does anyone have any idea as to what could cause something like this?
      One thing I am thinking of is that a thread we start ourselves could be causing a problem. The spec says you can't do that and we have been meaning to fix it but we figured that it would not really be a problem. Is it?


      SIGILL 4 (*) illegal instruction (not reset when caught)
      si_signo [4]: SIGILL: (*) illegal instruction (not reset when caught)
      si_errno [0]: Error 0
      si_code [30]: ILL_ILLOPC [addr: 0x0]

      stackpointer=4a423c98
      Writing java dump to /d4/usr/jboss/2.4.4/bin/javacore20096.1012826655.txt... OK

        • 1. Re: Core-dump

          Any coredump is a bug in the JVM or a native library
          you've added.

          The worst you should get for starting a new thread when
          not allowed is a SecurityException.

          I assume AIX uses an IBM VM? There might be some fixes?

          Regards,
          Adrian

          • 2. Re: Core-dump
            jacky_cheung

            It may be the JIT bug in the JVM. Try -nojit option in the java cmd to disable JIT to see whether it works.

            I will working on a project to run JBoss on AIX. Would you mind to tell me your experience in using JBoss on AIX? Thanks.

            Best regards,
            Jacky

            • 3. Re: Core-dump
              segerss

              We have been in production since the beginning of the year and have not experienced any platform-related problems (except this core-dump maybe).
              You are talking about 'the' jit-bug. Is this a well-known bug?

              Regards
              Steven