GNOME Bugzilla – Bug 356808
crash in Evolution: Deleted the current mess...
Last modified: 2006-09-25 13:30:59 UTC
What were you doing when the application crashed? Deleted the current message, then closed the blank message window it left. Then minimised evolution and continued doing other things. Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.0 2006-09-04 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: 256942080 vsize: 0 resident: 256942080 share: 0 rss: 96800768 rss_rlim: 0 CPU usage: start_time: 1158659249 rtime: 0 utime: 20967 stime: 0 cutime:19917 cstime: 0 timeout: 1050 it_real_value: 0 frequency: 17634 Backtrace was generated from '/usr/bin/evolution-2.8' (no debugging symbols found) Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1232976208 (LWP 4794)] [New Thread -1325933664 (LWP 6988)] [New Thread -1321890912 (LWP 5252)] [New Thread -1313059936 (LWP 5007)] [New Thread -1304667232 (LWP 5006)] [New Thread -1296274528 (LWP 4963)] [New Thread -1276154976 (LWP 4932)] [New Thread -1267369056 (LWP 4930)] [New Thread -1258939488 (LWP 4910)] [New Thread -1250546784 (LWP 4909)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 72783
Thanks for the bug report. Unfortunately, that stack trace is not very useful in determining the cause of the crash. Can you get us one with debugging symbols? Please see http://live.gnome.org/GettingTraces for more information on how to do so. Also, can you reproduce this issue, Adam?
Can't reproduce it unfortunately; not even sure what exactly caused it. Probably wouldn't have even noticed the crash without Bug Buddy. I've installed the evolution-dbg package, so if it happens again the stack trace will be more useful.
(In reply to comment #2) > Can't reproduce it unfortunately; not even sure what exactly caused it. Thanks for the fast response, Adam. :) Already suspected the crash isn't reproducible, based on your description. Closing INCOMPLETE, since the issue is not reproducible. Unfortunately the stacktrace does not have any hints about the real cause of the crash and does not help in debugging this issue. For some yet unknown reason, such stacktraces are pretty common currently on certain distros. However, almost all of them do turn out to be filed already, once we managed to gather a useful stacktrace. To do so, simply install the corresponding debugging package and reproduce the issue, if possible. The resulting stacktrace (see bug-buddy details) will help us to identify and fix the issue. Please feel free to report any further bugs you find. Thanks!
This just happened again. Doing exactly the same thing as before; deleted a message, closed the window, minimised evolution and switched back to Firefox to do other stuff. Stack trace with debugging info: Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.0 2006-09-04 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: 222638080 vsize: 0 resident: 222638080 share: 0 rss: 75780096 rss_rlim: 0 CPU usage: start_time: 1158750090 rtime: 0 utime: 10259 stime: 0 cutime:9575 cstime: 0 timeout: 684 it_real_value: 0 frequency: 4411 Backtrace was generated from '/usr/bin/evolution-2.8' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread -1232468304 (LWP 4783)] [New Thread -1354794080 (LWP 6956)] [New Thread -1321383008 (LWP 4996)] [New Thread -1312552032 (LWP 4987)] [New Thread -1304159328 (LWP 4986)] [New Thread -1295766624 (LWP 4944)] [New Thread -1275647072 (LWP 4930)] [New Thread -1266861152 (LWP 4928)] [New Thread -1258431584 (LWP 4923)] [New Thread -1250038880 (LWP 4922)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 72842
Thread 1 (Thread -1232468304 (LWP 4783))
Stacktrace looks kind of unusual, though fine and with debugging symbols. Unique trace. Already reopened. Thanks, Adam. :) Could you please also install the debugging package for evolution-data-server and the basic GNOME libs? Just in case, if you get that same crash again -- and if so, please add that trace to this bug report, too. Details can be found here: http://live.gnome.org/GettingTraces/DistroSpecificInstructions Also, are you able to see any aditional pattern to reproduce this? Like on deleting the first or last message only, etc.
I've installed the additional debug packages as requested. I guess we'll just wait and see. There's no pattern that I can distinguish, except that the two times this has happened I deleted the current message, closed the message window and minimised the evolution main window (bringing me back to Firefox both times) all in the time it took evolution to delete the message -- I use IMAP, so some times it takes a few seconds. It might also be worth mentioning Bug 356533 -- because of that, the next message is never loaded when deleting a message. This means that in relation to this bug, evolution would have been in the process of drawing an empty window (which I closed) when this happened. I have no idea if that makes any difference, but it can't hurt to mention it anyway.
Just happened again; similar to before, deleted a message and closed the window, then alt-tabbed to firefox. Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.0 2006-09-04 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: 161886208 vsize: 0 resident: 161886208 share: 0 rss: 39960576 rss_rlim: 0 CPU usage: start_time: 1158779140 rtime: 0 utime: 12113 stime: 0 cutime:11467 cstime: 0 timeout: 646 it_real_value: 0 frequency: 12360 Backtrace was generated from '/usr/bin/evolution-2.8' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread -1232243024 (LWP 6811)] [New Thread -1324299360 (LWP 12247)] [New Thread -1323742304 (LWP 7249)] [New Thread -1313526880 (LWP 6982)] [New Thread -1305134176 (LWP 6981)] [New Thread -1275421792 (LWP 6935)] [New Thread -1283814496 (LWP 6929)] [New Thread -1266635872 (LWP 6927)] [New Thread -1258206304 (LWP 6925)] [New Thread -1249813600 (LWP 6924)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 72873
Thread 1 (Thread -1232243024 (LWP 6811))
Stacktrace actually looks similar to bug 239441. However, that one did not have any duplicate in 3.5 years. Also, recurse and sort_info values are totally different.
Crashed it again :-D This time was slightly different. Rather than deleting a message then closing the window, I opened a new message then decided I didn't want to read it after all and immediately closed the window (before the message was done rendering). At this point it crashed. I'm guessing this has something to do with the fact that the message window was busy doing something when it was closed. I have a stack trace, but it looks pretty much the same to the last one to me (but then, I don't really know what I'm looking at). I'll attach it rather than pasting it, since this page is getting quite long.
Created attachment 73207 [details] Stack trace
(In reply to comment #9) > I have a stack trace, but it looks pretty much the same to the last one to me > (but then, I don't really know what I'm looking at). I'll attach it rather > than pasting it, since this page is getting quite long. True. ;) Indeed, that's the same crash. In case you are curious: Just have a look at the top-most crashing thread, the one with the <signal handler called> line. Compare the functions called, downward. resort_node() is the last function called, that crashed. It has been called by etta_sort_info_changed(), and so on. In this case, you can see, that values like recurse and sort_info are identical. Most others are pointers anyway, addresses of the data in memory -- so they of course differ.
Thanks for the bug report. This particular bug has already been reported into our bug tracking system, but please feel free to report any further bugs you find. I have also updated bug 239441 that this bug here provides an up-to-date stacktrace. *** This bug has been marked as a duplicate of 239441 ***