Signature mgMouseService.so@0x2ddd UUID 34623a32-2aa7-4512-b7b3-7cfde2081204 Time 2008-12-04 16:07:34-08 Uptime 39 Last Crash 42 seconds before submission Product Firefox Version 3.2a1pre Build ID 20081204021237 Branch 1.9.2 OS Linux OS Version 0.0.0 Linux 2.6.27-9-generic #1 SMP Thu Nov 20 21:57:00 UTC 2008 i686 GNU/Linux CPU x86 CPU Info GenuineIntel family 10 model 22 stepping 1 Crash Reason SIGSEGV Crash Address 0xb376fddd Comments Crashing Thread Frame Module Signature [Expand] Source 0 mgMouseService.so mgMouseService.so@0x2ddd 1 libgdk-x11-2.0.so.0.1400.4 libgdk-x11-2.0.so.0.1400.4@0x47165 2 libgdk-x11-2.0.so.0.1400.4 libgdk-x11-2.0.so.0.1400.4@0x490b2 3 libgdk-x11-2.0.so.0.1400.4 libgdk-x11-2.0.so.0.1400.4@0x494de 4 libglib-2.0.so.0.1800.2 libglib-2.0.so.0.1800.2@0x396f7 5 libglib-2.0.so.0.1800.2 libglib-2.0.so.0.1800.2@0x3cda2 6 libglib-2.0.so.0.1800.2 libglib-2.0.so.0.1800.2@0x3cf60 7 libxul.so nsAppShell::ProcessNextNativeEvent widget/src/gtk2/nsAppShell.cpp:144 8 libxul.so nsBaseAppShell::DoProcessNextNativeEvent widget/src/xpwidgets/nsBaseAppShell.cpp:151 9 libxul.so nsBaseAppShell::OnProcessNextEvent widget/src/xpwidgets/nsBaseAppShell.cpp:278 10 libxul.so nsThread::ProcessNextEvent xpcom/threads/nsThread.cpp:497 11 libxul.so NS_ProcessNextEvent_P nsThreadUtils.cpp:227 12 libxul.so nsBaseAppShell::Run widget/src/xpwidgets/nsBaseAppShell.cpp:170 13 libxul.so nsAppStartup::Run toolkit/components/startup/src/nsAppStartup.cpp:192 14 libxul.so XRE_main toolkit/xre/nsAppRunner.cpp:3268 15 firefox-bin main browser/app/nsBrowserApp.cpp:156 16 libc-2.8.90.so libc-2.8.90.so@0x16684
I'm not able to reproduce the crash. (Ubuntu 8.10, 2.6.27-9-generic) BTW: There is no MozGest version compatible with Fx 3.2a1pre! ...at least I can't remember that I've released such a version.
yeah, well... please see the reports it's clear that this library (which i take it you're confirming is probably yours) is hurting someone. actually, it looks like precisely one person: Linux 0.0.0 Linux 2.6.27-9-generic #1 SMP Thu Nov 20 21:57:00 UTC 2008 i686 GNU/Linux can crash reporter send people messages? i'd really love to send this guy one :(
If the user submitted an email, you could file a server ops bug to ask aravind to query the db directly for his email. Socorro doesn't have any functionality to do anything with email addresses as it stands.
One question: What's the meaning of "Uptime 39"? Minutes, seconds, milliseconds? And, yes "mgMouseService.so" is my library. I still can't reproduce the crash though.
Uptime is seconds since the program started to the time it crashed.
i recently had a crash with this stacktrace, so I think this may be me. What do you need?
well, a couple of things would help. 1. you could load about:crashes and figure out which one is yours so that i know which one. 2. install -dbg / -debuginfo / symbols for at least libgdk-x11 libgdk-x11, ideally for everything loaded in gecko. w/ a stack trace, it should be possible to try to figure out which functions were being called. jochen: do you have debuginfo for your extension?
here's one I just got: http://crash-stats.mozilla.com/report/index/c747d524-7b1f-4132-9c04-697972081209 also: I'm not quite sure what you want me to install. care to explain what you want?
Ok, I can reproduce the crash. http://crash-stats.mozilla.com/report/index/a8d93bd6-db8c-4ce6-9d59-123f52081210 There is no reliable way to reproduce the crash, but closing a tab will sooner or later reproduce the crash after clicking on content. In MozGest 2.0.3 mgMouseService.so is calling GDK_IS_WINDOW(xyz) and it's not sure xyz still exists or is currently destroyed. https://addons.mozilla.org/en-US/firefox/addons/versions/39#version-2.1.0pre2 or http://www.mousegestures.org/download/release/2.1.X/mozgest_2_1_0pre2.xpi should work, because the crashing code was removed several weeks ago. Both links should serve identical builds. I wish there would be a simple (and frozen) way to get the window handle of the current content... @Shayne Jewers: Can you please download the above mentioned build and test it? Thanks! Of course you have to bump compatibility to Fx 3.2a1pre... ;)
http://library.gnome.org/devel/gtk/unstable/GtkWidget.html "delete-event" : Run Last "destroy-event" : Run Last try listening for these?
(In reply to comment #10) > http://library.gnome.org/devel/gtk/unstable/GtkWidget.html > "delete-event" : Run Last > "destroy-event" : Run Last > > try listening for these? Well, to be honest, I wasn't aware of that. The crashing code was removed weeks ago, because it wasn't needed. It wasn't removed because it was crashing. Is there a "lean and mean" way to get the window handle of the currently displayed content window? Maybe via "nsIWindowWatcher"?
fwiw, i had to ask a coworker, and i know for a fact that our (corporate) codebase crashes because of stale pointers such as these, so even though we know of it, it doesn't seem to mean we properly use it.
What a mess! From my point of view the bug is already fixed because the upcoming version of MozGest comes without the crashing code. It's just a matter of time for review (AMO) though... The main problem for an update (AMO) is that mgMouseService is closed source...
I haven't had crashing in a while, so it looks like this is fixed.
Crash Signature: [@ mgMouseService.so@0x2ddd]
Portable Nightly 37.0a1 (2014-12-21) Have been using the Portable Nightly the past few days and trying to get mouse gestures to work, between All-In-One Gestures and FireGestures. Have had some little successes. I think the gestures interface is very important. I have gotten used to using it. No crashes at all since I've been using the Portable Nightly.
This is about a legacy add-on which is no longer compatible with Firefox 57, so closing.
Status: ASSIGNED → RESOLVED
Last Resolved: 7 months ago
Resolution: --- → INVALID
You need to log in before you can comment on or make changes to this bug.