GNOME Bugzilla – Bug 663084
[abrt] evolution-data-server-3.2.1-1.fc16: io_error: Process /usr/libexec/e-addressbook-factory was killed by signal 11 (SIGSEGV)
Last modified: 2012-07-19 13:07:22 UTC
Moving this from a downstream bug report: https://bugzilla.redhat.com/show_bug.cgi?id=750036 libreport version: 2.0.6 abrt_version: 2.0.4.981 backtrace_rating: 4 cmdline: /usr/libexec/e-addressbook-factory comment: I don't know. crash_function: io_error executable: /usr/libexec/e-addressbook-factory kernel: 3.1.0-1.fc16.x86_64 reason: Process /usr/libexec/e-addressbook-factory was killed by signal 11 (SIGSEGV) time: Sun Oct 30 04:15:18 2011 Core was generated by `/usr/libexec/e-addressbook-factory'. Program terminated with signal 11, Segmentation fault.
+ Trace 228950
Thread 7 (Thread 0x7f7fed697700 (LWP 2034))
Thread 2 (Thread 0x7f7ff23577c0 (LWP 2011))
Thread 1 (Thread 0x7f7fdde9f700 (LWP 23411))
This occurred to me as well.
Dear Evolution developers, even though you don't see it because ABRT goes to RedHat bugzilla, I'm having this crash on an almost daily basis. Could anyone care to take a look at it, please?
The crash happens below libgdata, which doesn't necessary mean that it's not evolution's fault, but it would be better to ask there first. Philip, did you see anything similar in 3.2.x+, please?
Could this be bug #637741? What version of libsoup was being used?
(In reply to comment #4) > Could this be bug #637741? What version of libsoup was being used? The one bundled with Fedora 16, which would be 2.36.1.
(In reply to comment #5) > (In reply to comment #4) > > Could this be bug #637741? What version of libsoup was being used? > > The one bundled with Fedora 16, which would be 2.36.1. Probably not bug #637741 then, although it looks similar. I can’t see what in libgdata/EDS is causing this, and suspect it might be a libsoup problem. It looks like the SoupMessage in thread 1 (0x7f7fe000e2e0) is getting destroyed between entering read_metadata(), since http://git.gnome.org/browse/libsoup/tree/libsoup/soup-message-io.c?id=LIBSOUP_2_36_1#n272 succeeds, and SOUP_SOCKET_ERROR being handled, since http://git.gnome.org/browse/libsoup/tree/libsoup/soup-message-io.c?id=LIBSOUP_2_36_1#n179 fails. I don’t know enough about the SoupMessage lifecycle in libsoup to work out what could be destroying it. libgdata certainly holds a reference to the SoupMessage while sending it: http://git.gnome.org/browse/libgdata/tree/gdata/gdata-service.c?id=LIBGDATA_0_10_1#n618.
Similar downstream bug report from 3.4.2: https://bugzilla.redhat.com/show_bug.cgi?id=833459
The refcount of the message should be at least 3 at that point... there's no way it could be getting destroyed. Is there any stderr / .xsession-errors associated with the crash?
(In reply to comment #8) > Is there any stderr / .xsession-errors associated with the crash? Have you checked the downstream RH bug and duplicates? There's quite some ABRT info there.
There's lots of abrt info, but no .xsession-errors, which is the thing (other than the backtrace) that would be most useful here.
I just got similar downstream bug report: https://bugzilla.redhat.com/show_bug.cgi?id=839614 It also contains .xsession_errors, though nothing interesting at the end of it. Dan, would you mind to look at it and write your thoughts there, please?
This bug no longer exists in git master. I've fixed it on the gnome-3-4 branch (http://git.gnome.org/browse/libsoup/commit/?h=gnome-3-4&id=ae6bc194), and I'm building new f17 packages now.