On 29 September 2010 00:27, Martin Simmons <mar...@lispworks.com> wrote:
> >>>>> On Tue, 28 Sep 2010 06:51:44 +1000, Greg Cockburn said:
> >
> > >>>>>> On Mon, 27 Sep 2010 21:52:17 +1000, Greg Cockburn said:
> > >>
> > >> Hi all,
> > >>
> > >> I have just installed bacula 5.0.3 from src rpms, and I can't get bat
> to
> > >> start.
> > >>
> > >> I have checked bat.conf bacula-dir.conf. I can get bconsole working
> > using
> > >> bat.conf.
> > >>
> > >> This is the output I get:
> > >>
> > >> bat: lockmgr.c:911 ASSERT failed at lockmgr.c:911: lmgr_is_active()
> > >>
> > >> I ran strace and this seems to be the closest thing to the problem
> just
> > >> before the assert fails:
> > >>
> > >> clock_gettime(CLOCK_MONOTONIC, {86341, 32597622}) = 0
> > >> read(3, "\0", 16) = 1
> > >> read(3, 0x7fffb3ee6040, 16) = -1 EAGAIN (Resource
> temporarily
> > >> unavailable)
> > >> clock_gettime(CLOCK_MONOTONIC, {86341, 32872622}) = 0
> > >> clock_gettime(CLOCK_MONOTONIC, {86341, 32988622}) = 0
> > >>
> > >> FD 3 is some sort of pipe.
> > >>
> > >> Has anyone seen this or able to help as to what is going wrong?
> > >
> > >Try running bat under gdb and then use the following gdb command when it
> > hits
> > >the assertion:
> > >
> > >thread apply all where 100
> > >
> > >If possible, do this with Bacula compiled with debugging information.
> > >
> > >__Martin
> >
> > Hi Martin,
> >
> > Starting to get out of my knowledge range, so here is the gdb output
> after
> > running "thread apply all where 100"
> >
> > Detaching after fork from child process 22516.
> > bat: lockmgr.c:911 ASSERT failed at lockmgr.c:911: lmgr_is_active()
> >
> > Program received signal SIGSEGV, Segmentation fault.
> > 0x0000003d66440640 in lmgr_thread_create(unsigned long*, pthread_attr_t
> > const*, void* (*)(void*), void*) () from /usr/lib64/libbac-5.0.3.so
> > (gdb) thread apply all where 100
> >
> > Thread 2 (Thread 0x40a00940 (LWP 22515)):
> > #0 0x0000003e31eccfc2 in select () from /lib64/libc.so.6
> > #1 0x0000000000962a80 in QProcessManager::run() ()
> > #2 0x00000000008e0be5 in QThreadPrivate::start(void*) ()
> > #3 0x0000003e32a0673d in start_thread () from /lib64/libpthread.so.0
> > #4 0x0000003e31ed3d1d in clone () from /lib64/libc.so.6
> >
> > Thread 1 (Thread 0x2aaaaaacc5f0 (LWP 22512)):
> > #0 0x0000003d66440640 in lmgr_thread_create(unsigned long*,
> pthread_attr_t
> > const*, void* (*)(void*), void*) () from /usr/lib64/libbac-5.0.3.so
> > #1 0x0000003d6643bd31 in start_watchdog() () from /usr/lib64/
> > libbac-5.0.3.so
> > #2 0x0000003d6643be47 in new_watchdog() () from /usr/lib64/
> libbac-5.0.3.so
> > #3 0x0000003d6643d3df in ?? () from /usr/lib64/libbac-5.0.3.so
> > #4 0x0000003d6643d517 in start_thread_timer(JCR*, unsigned long,
> unsigned
> > int) () from /usr/lib64/libbac-5.0.3.so
> > #5 0x0000003d66418d73 in BSOCK::connect(JCR*, int, long, long, char
> const*,
> > char*, char*, int, int) () from /usr/lib64/libbac-5.0.3.so
> > #6 0x0000003d66415555 in bnet_connect(JCR*, int, long, long, char
> const*,
> > char*, char*, int, int) () from /usr/lib64/libbac-5.0.3.so
> > #7 0x00000000004295ad in DirComm::connect_dir (this=0x106b340) at
> > bcomm/dircomm.cpp:171
> > #8 0x000000000042dfcf in Console::connect_dir (this=0x106b030) at
> > console/console.cpp:132
> > #9 0x0000000000415812 in MainWin::MainWin (this=0x1054680, parent=0x0)
> at
> > mainwin.cpp:98
> > #10 0x000000000040d044 in main (argc=0, argv=0x7fffffffcdb0) at
> main.cpp:163
>
> I've submitted a bug report at http://bugs.bacula.org/view.php?id=1642which
> describes the problem.
>
> The only way to fix the rpms is to delve into the build process to make
> them
> consistent w.r.t. what diff-503.patch does in the main bacula rpms.
>
> __Martin
>
>
> ------------------------------------------------------------------------------
> Start uncovering the many advantages of virtual appliances
> and start using them to simplify application deployment and
> accelerate your shift to cloud computing.
> http://p.sf.net/sfu/novell-sfdev2dev
> _______________________________________________
> Bacula-users mailing list
> Bacula-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>
Hi Martin,
I had a look at that patch and modified to remove the define of the
DEVELOPER flag.
Recompiled and all works brilliantly.
Thanks for raising the bug.
Greg.
------------------------------------------------------------------------------
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
