Apache OpenOffice (AOO) Bugzilla – Issue 15240
sudden silent crash ... (how get error message)?
Last modified: 2013-08-07 14:41:36 UTC
Hi, only thing I can say up to now: OO crashed 2-3 times over the last week. I had always opened some OO files simultanously. OO just disappears from screen. When reopening it then it doesn't even ask me to re-open an un-saved (former) document (like OO normaly does after a crash). I think I used version 1.02 before and there it didn't happen. I am using fvwm as a window manger and a gentoo distro. I am sorry to annouce this issue without beeing able to decribe the precise circumstances when and how the crash happens. Hope you can help me to track the error somehow. How do I make OO send me error messages? Does it help starting OO from an xterm? Thanks
The thing to do is start OpenOffice under gdb from an xterm, then when the crash happens, give the 'bt' command to get a stack dump. See http://www.kegel.com/openoffice/#dump for instructions. Once you have the stack dump, upload it as an attachment to this issue, along with some comment about what you were doing at the time of the crash. Thanks so much!!
Sorry if I bother you with secondary issues but I've got some questions about the bug tracking gdb thing. This is what happens: (1) I start a script with the following content: script crash.txt bin=/opt/OpenOffice.org1.1Beta/program LD_LIBRARY_PATH=$bin gdb $bin/soffice.bin (2) It says: Script started and exits (3) I do gdm and it leads me to the gdm commandline (4) I type run (or even: run /opt/OpenOffice1.1/program/soffice) (5) it says: no executable file specified How do I make OO run in this gdb environment? Thank You
My fault! You can't put the 'script crash.txt' in the same shell script as gdb, since script starts a subshell. I've revised my instructions - can you try them again please? Apologies!
I'll do that after I'll have finished my finals in June, 23th.
How 'bout that stack trace? Finals are over :-)
Hi, finals are over, you are right, :-) since then I had some crashs with OO but in other situations than described above (without stack dump switched on). But I still try to catch it and I'll be back here when I've got it. Bye
Prio is changed in cause of Not known dramatic situation to get the crash. We need a clearly step by step description to be able to reproduce the crash.
I'm going to reinterpret this as a request for a crash logger :-) That's part of the upcoming 1.1rc1. Once you're using that, you'll be able to submit crash logs more easily. Please use a new issue when you do. Thanks! *** This issue has been marked as a duplicate of 14119 ***
.