crashs on any quantal review

Bug #1048886 reported by Christopher Kyle Horton on 2012-09-11
This bug affects 25 people
Affects Status Importance Assigned to Milestone
software-center (Ubuntu)
Albert Astals Cid

Bug Description

Crash occurred when I tried to write a review for the enigmail package in the Software Center. The window for writing the review popped up for a moment, seemed to be loading, then promptly disappeared and I got this crash report instead.

ProblemType: Crash
DistroRelease: Ubuntu 12.10
Package: software-center 5.3.11
ProcVersionSignature: Ubuntu 3.5.0-14.15-generic 3.5.3
Uname: Linux 3.5.0-14-generic x86_64
ApportVersion: 2.5.1-0ubuntu7
Architecture: amd64
Date: Mon Sep 10 20:54:13 2012
ExecutablePath: /usr/share/software-center/
InstallationMedia: Ubuntu 12.10 "Quantal Quetzal" - Alpha amd64 (20120904.1)
InterpreterPath: /usr/bin/python2.7
PackageArchitecture: all
ProcCmdline: /usr/bin/python /usr/share/software-center/ --pkgname enigmail --iconname enigmail --parent-xid --version 2:1.4.4-0ubuntu1 --origin ubuntu --datadir /usr/share/software-center/ --appname Enigmail\ extension\ for\ Thunderbird
 PATH=(custom, no user)
 Segfault happened at: 0x48aaf8: addl $0x1,0x98(%r13)
 PC (0x0048aaf8) ok
 source "$0x1" ok
 destination "0x98(%r13)" (0x00000098) not located in a known VMA region (needed writable region)!
SegvReason: writing NULL VMA
Signal: 11
SourcePackage: software-center
 ?? ()
 ?? ()
 ?? () from /usr/lib/python2.7/dist-packages/gi/_glib/
 ?? () from /lib/x86_64-linux-gnu/
 g_main_context_dispatch () from /lib/x86_64-linux-gnu/
Title: crashed with SIGSEGV in g_main_context_dispatch()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo

Related branches

Changed in software-center (Ubuntu):
importance: Undecided → Medium
summary: - crashed with SIGSEGV in g_main_context_dispatch()
+ crashed with SIGSEGV in subtype_dealloc.25740()
tags: removed: need-amd64-retrace

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in software-center (Ubuntu):
status: New → Confirmed
Brian Murray (brian-murray) wrote :

The stack trace for this bug report has the bug reporter's email addresses in them so perhaps should not be made public.

@Chris it would be helpful if we could make this bug report public. You can do this yourself by clicking the pencil next to 'This bug report is Private'. Thanks!

@Brian Done.

visibility: private → public
Albert Astals Cid (aacid) wrote :

I'm having a look, but my software-center/python/glib foo is not huge so if anyone has a hint of where the bug may be i'll be happy to listen to it :-)

Changed in software-center (Ubuntu):
assignee: nobody → Albert Astals Cid (aacid)
Albert Astals Cid (aacid) wrote :

Found a fix/workaround, seems that some variables were going "out of scope" and being freed too early, not sure if that might be a pygobject bug or the previous usage was wrong. Anyone with more knowledge in the area?

Michael Vogt (mvo) wrote :

This is 100% reproducable in trunk, e.g. via:

$ PYTHONPATH=. gdb --args python ./utils/ --pkgname enigmail --iconname enigmail --version 2:1.4.4-0ubuntu1 --origin ubuntu --datadir ./data --debug

This appears to be crashing for any review.

Changed in software-center (Ubuntu):
importance: Medium → Critical
tags: added: ca-escalated
Michael Vogt (mvo) on 2012-10-04
summary: - crashed with SIGSEGV in subtype_dealloc.25740()
+ crashs on any quantal review
description: updated
Changed in software-center (Ubuntu):
status: Confirmed → In Progress
status: In Progress → Fix Committed
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package software-center - 5.4

software-center (5.4) quantal; urgency=low

  [ Michael Vogt ]
  * lp:~mvo/software-center/lp1044141:
    - fix unicode error crash on a transaction error
      (LP: #1044141)
  * lp:~mvo/software-center/fix-size-calc-race:
    - fix crash due to a race in the size calculation code
      in the details view (LP: #1041004)
  * lp:~mvo/software-center/fix-crash-1048886:
    - No longer call GLib.init_threads() as this will cause
      LP: #1048886 later. Many thanks to Albert Astals Cid
  * lp:~mvo/software-center/dbus-idle-timeout:
    - stop the dbus data provider after 5min of inactivity
      (LP: #1058567)
  * lp:~mvo/software-center/lp1051829:
    - Do not report deb packages with lintian errors via the
      recoverable errors mechanism. This removes 97 out of 104
      instances of my sample of the recoverable error report
      for LP: #1051829. Other types of failures will still be
      reported as recoverable failures though (as we do want
      to know about them).

  [ Gary Lakser ]
  * lp:~gary-lasker/software-center/lp1041004:
    - ensure that app_details is valid before using it (LP: #1041004)
  * lp:~gary-lasker/software-center/fix-crash-lp1021308:
    - send synthetic "style-updated" signals to ensure that the signal
      handlers are called on themes that do not send this signal (like
      oxygen-gtk3). This fixes a crash and broken drawing LP: #1021308

  [ Nicolas Delvaux ]
  * lp:~malizor/software-center/fix-lp1050036:
    - fix crash in get_label() (LP: #1050036) and drop the i18n for the
      yes/no symbols that were not translated in any language (as a result can also be dropped)
 -- Michael Vogt <email address hidden> Fri, 05 Oct 2012 10:58:20 +0200

Changed in software-center (Ubuntu):
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers