gnome-terminal crashed with SIGSEGV in type_get_qdata_L()

[Pedro Villavicencio]

Binary package hint: gnome-terminal

crashed after disabling the desktop effects, meaning launching metacity --replace

ProblemType: Crash
DistroRelease: Ubuntu 10.10
Package: gnome-terminal 2.31.90-0ubuntu1
ProcVersionSignature: Ubuntu 2.6.35-17.23-generic 2.6.35.2
Uname: Linux 2.6.35-17-generic i686
Architecture: i386
CheckboxSubmission: 4afb8c68a8fa5938a80183a271a939ea
CheckboxSystem: c460f5739f7c8446c3d18c3e0e7c5745
CrashCounter: 1
Date: Mon Aug 23 16:11:13 2010
ExecutablePath: /usr/bin/gnome-terminal
ProcCmdline: gnome-terminal
ProcEnviron:
 LANG=es_CL.utf8
 SHELL=/bin/bash
SegvAnalysis:
 Segfault happened at: 0xf597bc <type_get_qdata_L+76>: mov (%edi),%esi
 PC (0x00f597bc) ok
 source "(%edi)" (0x2da67bf8) not located in a known VMA region (needed readable region)!
 destination "%esi" ok
SegvReason: reading unknown VMA
Signal: 11
SourcePackage: gnome-terminal
StacktraceTop:
 type_get_qdata_L (node=<value optimized out>, quark=1)
 type_check_is_value_type_U (type=10978944)
 g_type_check_is_value_type (type=10978944)
 g_value_type_compatible (src_type=10978944,
 g_value_object_collect_value (value=0x920e08c,
Title: gnome-terminal crashed with SIGSEGV in type_get_qdata_L()
UserGroups: adm admin audio cdrom dialout fuse lpadmin netdev plugdev sambashare vboxusers video

[Pedro Villavicencio (pedro)]

Backtrace is completed, no need to retrace it, deleting the coredump and untagging it.

[Pedro Villavicencio (pedro)]

I've sent this upstream at https://bugzilla.gnome.org/show_bug.cgi?id=627764

[Dennis Sheil (dennis-sheil)]

If you look at the robust traceroute of #620522 against this one, you can see the bugs are the same. They also both happen when disabling desktop visual effects.

[Pedro Villavicencio (pedro)]

Dennis, this was marked as Triaged, sent upstream and if you look at the upstream report the developers are working on it, so why marking it as a duplicate of a non triaged one? that doesn't make sense please do not do it. The other bugs should be marked as a duplicate of this one. doing that now. thanks all.

[Dennis Sheil (dennis-sheil)]

> Dennis, this was marked as Triaged, sent upstream and if you look at the upstream report the developers are working on it, so why [are you] marking it as a duplicate of a non triaged one? [T]hat doesn't make sense[,] please do not do it. The other bugs should be marked as a duplicate of this one. [D]oing that now. [T]hanks all.

I'm sorry, I did not realize that what made sense was for you to create a new bug report when two earlier ones existed. With the original report having a full backtrace with debug symbols before your third report was created. I did not know that it made sense to instead make a redundant third bug report, triage that and send that one upstream, and that people should somehow intuit that this was now the report that earlier bugs should be made a duplicate of. I had been under the impression that the earliest bug report would be the one later bug reports would be made duplicates of. I was under the impression that when a bug report with full backtraces with debug symbols exists, that you do not make a redundant bug report and decide to triage that one and send that one upstream.

I can not find reference to your system of bug reporting in Ubuntu's documentation, could you kindly point me to where it is referenced? If it is not there, could you please update Ubuntu's documentation with regards to your system of bug reporting, because you are right, I can not make sense of it.

[Pedro Villavicencio (pedro)]

Could somebody try to reproduce this on Natty and comment back if they're able/unable to reproduce the issue? Thanks.

[dino99 (9d9)]

This version has expired long time ago, and is no more supported