abrt 1.0.8 detected a crash. architecture: i686 Attached file: backtrace cmdline: /usr/lib/openoffice.org3/program/swriter.bin -writer component: openoffice.org executable: /usr/lib/openoffice.org3/program/swriter.bin kernel: 2.6.32.9-70.fc12.i686.PAE package: openoffice.org-writer-1:3.1.1-19.28.fc12 rating: 4 reason: Process /usr/lib/openoffice.org3/program/swriter.bin was killed by signal 6 (SIGABRT) release: Fedora release 12 (Constantine)
Created attachment 403027 [details] File: backtrace
I feel this is a threading problem: an Invalidate() call without acquiring the SolarMutex, which incidentally clashed with repaint running from the main thread. It seems there are at least four methods of ExtensionBox_Impl calling Invalidate() and only one (checkEntries) is called with the SolarMutex acquired. But I'd be very glad to get Caolan's opinion on this. dtardon->mark: Can you reproduce the crash? If it really is a threading issue, I wouldn't expect 100 % reproducibility, but at least sometimes?
(In reply to comment #2) > I feel this is a threading problem: an Invalidate() call without acquiring the > SolarMutex, which incidentally clashed with repaint running from the main > thread. It seems there are at least four methods of ExtensionBox_Impl calling > Invalidate() and only one (checkEntries) is called with the SolarMutex > acquired. But I'd be very glad to get Caolan's opinion on this. > > dtardon->mark: Can you reproduce the crash? If it really is a threading issue, > I wouldn't expect 100 % reproducibility, but at least sometimes? If I remember correctly (and I'm not positive I do), I had the extension manager open and was looking for and applying updates. I tried to reproduce but the updates had all been installed. Is there an easy way to undo the updates so I can attempt to reproduce?
Afraid not :( The only way is to remove the extension and install older version of it.
caolanm->dtardon: Sounds plausible anyway that any routes that might end up in Invalidate should take a SolarMutex, e.g. ExtMgrDialog::addPackageToList and friends. Can you take that up with upstream and clarify if that's the case.
Well, upstream haven't said a word in over two months and there has been no other report for the same problem so far, so I'm taking the liberty to close this UPSTREAM. I'll reconsider commiting the fix to Fedora if there is another report for the same thing.
*** Bug 605931 has been marked as a duplicate of this bug. ***
*** Bug 664072 has been marked as a duplicate of this bug. ***
We have got another dupe of this and upstream has accepted the patch meanwhile, so I'm going to push it into Fedora after all.
fixed in openoffice.org-3.2.0-12.35.fc13 and openoffice.org-3.3.0-18.2.fc14
openoffice.org-3.3.0-18.2.fc14 has been submitted as an update for Fedora 14. https://admin.fedoraproject.org/updates/openoffice.org-3.3.0-18.2.fc14
openoffice.org-3.3.0-18.2.fc14 has been pushed to the Fedora 14 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update openoffice.org'. You can provide feedback for this update here: https://admin.fedoraproject.org/updates/openoffice.org-3.3.0-18.2.fc14
openoffice.org-3.3.0-18.2.fc14 has been pushed to the Fedora 14 stable repository. If problems still persist, please make note of it in this bug report.