Crashes on shutdown if libevent loop doesn't exit soon enough

[Till Ulen]

Binary package hint: transmission

The Transmission BitTorrent client crashed after a quit command.

ProblemType: Crash
Architecture: i386
Date: Thu Jan 22 09:29:16 2009
DistroRelease: Ubuntu 8.04
ExecutablePath: /usr/bin/transmission
Package: transmission-gtk 1.06-0ubuntu6
PackageArchitecture: i386
ProcCmdline: transmission
ProcEnviron:
 PATH=/home/username/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
 LANG=en_US.UTF-8
 SHELL=/bin/bash
Signal: 11
SourcePackage: transmission
StacktraceTop:
 ?? () from /usr/lib/libgobject-2.0.so.0
 g_closure_invoke () from /usr/lib/libgobject-2.0.so.0
 ?? () from /usr/lib/libgobject-2.0.so.0
 g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0
 g_signal_emit () from /usr/lib/libgobject-2.0.so.0
Title: transmission crashed with SIGSEGV in g_closure_invoke()
Uname: Linux 2.6.24-23-generic i686
UserGroups: adm admin audio cdrom floppy fuse lpadmin plugdev sambashare video

[Apport retracing service (apport)]

StacktraceTop:IA__g_closure_invoke (closure=0x81c1dc8, return_value=0x0,
signal_emit_unlocked_R (node=0x81c2a70, detail=0,
IA__g_signal_emit_valist (instance=0xbfb78194,
IA__g_signal_emit (instance=0x85d00c8, signal_id=109,
handle_keys_changed (data=0x85d00c8)

[Hew (hew)]

Thanks for your report. Does the problem still exist with Ubuntu Jaunty and transmission 1.51-0ubuntu1?

[Till Ulen (tillulen)]

I do not know. I cannot reproduce the problem.

[Hew (hew)]

We are closing this bug report because it lacks the information we need to investigate the problem, as described in the previous comments. Please reopen it if you can give us the missing information, and don't hesitate to submit bug reports in the future. To reopen the bug report you can click on the current status, under the Status column, and change the Status back to "New". Thanks again!

[Till Ulen (tillulen)]

Hew, I don't understand your decision. We've got an unreproducible crash here that has debug information attached. Is it Ubuntu policy to ignore such bugs? Is that policy documented?

Can the debug information help? I imagine someone knowledgeable enough could use the backtrace to find the specific problem that caused the crash. Or would that be too hard, a waste of time or something?

I really do not understand why you are closing this bug.

[Hew (hew)]

There have been a large number of changes between the version currently in the development release of Ubuntu, and the version the crash was reported with. For this reason it is usually helpful to test with the latest release to see if the problem still exists before it is looked at by a developer. Upstream will not look at bug reports such as these, as they will ask to first test with the latest version. While a crash is definitely a problem, it's best if we can avoid developers investigating reports of bugs that no longer exist in the latest release.

I have reopened this bug. It would help a great deal if you could test with the version in Jaunty at some stage.

[Till Ulen (tillulen)]

Thanks for reopening this bug. The explanation makes sense to me.

Could you please clarify what kind of testing do you expect to be performed on Jaunty? I know of no way to trigger this crash again even on my Hardy. Let's assume that I have installed Jaunty and the new version of Transmission does not crash there. Does it mean that this bug has been fixed in it? No. It might be still in, but I just haven't created appropriate conditions for it to show itself.

[Hew (hew)]

You mentioned the crash occurred after a "quit command". You could try reproducing the same action to see if the crash occurs again, or even just use Transmission normally for a while to see if it's now stable.

There are similarities between this and bug 195686, which was fixed with transmission 1.10. Apart from that, you appear to be the only reporter of this bug, and that version of Transmission has been around for a year. Under these circumstances, if a bug cannot be reproduced anymore in the latest version, it's reasonable to close it. It can always be reopened if the problem reappears in the future.

If you're unable to test with Jaunty, you may be able to test with transmission 1.22 in hardy-backports, which would be an improvement. I have updated the backport request at bug 304252 to ask for 1.51 to be backported.

Thanks again for your help!

[Robert Hollencamp (rhollencamp)]

I just received this problem today when I quit Transmission. Using Jaunty, fully updated as of today

[Fernando Miguel (fernandomiguel)]

i saw a similar traceback on karmic today, but didnt copy the crash lines.
if i see it again, i'll paste it here

[Charles Kerr (charlesk)]

There are several ways for Transmission to crash during shutdown, by the way, and they can probably all go under the same "umbrella" ticket because they have the same root cause, even though the backtraces are in wildly different places.

[Peter B P (peterbp)]

I'm getting this regularly, after Transmission having been active for at least several hours, seeding large amounts of data, and it's memory footprint having grown to over 200 MB in size.

I was unable to upload the crash report, as the 300+ megabyutes choked the server. :/

[Charles Kerr (charlesk)]

I think this is fixed in 1.80 beta 1. The fix passes the sniff test and has been confirmed by several Mac users who had similar behavior. There are several people subscribed to this ticket; could someone give 1.80 beta 1 a spin on Ubuntu and report back whether or not the problem persists?

[Krzysztof Klimonda (kklimonda)]

I'm closing this bug as we have had no reports from people experiencing the crash since 1.80 beta1 release.