shotwell crashed with SIGSEGV in publishing_accounts_uoa_publisher_authenticator_on_processed()

Bug #1225629 reported by Philip Khor on 2013-09-15
58
This bug affects 8 people
Affects Status Importance Assigned to Milestone
shotwell (Ubuntu)
High
Alberto Mardegan

Bug Description

Shotwell crashed when attempting to publish photos to Facebook.

ProblemType: Crash
DistroRelease: Ubuntu 13.10
Package: shotwell 0.15.0~pr1-0ubuntu2
ProcVersionSignature: Ubuntu 3.11.0-7.13-generic 3.11.0
Uname: Linux 3.11.0-7-generic x86_64
NonfreeKernelModules: wl
ApportVersion: 2.12.1-0ubuntu4
Architecture: amd64
Date: Sun Sep 15 16:28:25 2013
ExecutablePath: /usr/bin/shotwell
InstallationDate: Installed on 2013-08-16 (29 days ago)
InstallationMedia: Ubuntu 13.10 "Saucy Salamander" - Alpha amd64 (20130815)
MarkForUpload: True
ProcCmdline: shotwell
SegvAnalysis:
 Segfault happened at: 0x7f8481617d3a: mov (%rax),%rax
 PC (0x7f8481617d3a) ok
 source "(%rax)" (0x38000002432e) not located in a known VMA region (needed readable region)!
 destination "%rax" ok
SegvReason: reading unknown VMA
Signal: 11
SourcePackage: shotwell
StacktraceTop:
 ?? () from /usr/lib/shotwell/plugins/builtin/shotwell-publishing.so
 ?? () from /usr/lib/shotwell/plugins/builtin/shotwell-publishing.so
 ?? () from /usr/lib/x86_64-linux-gnu/libsignon-glib.so.1
 g_simple_async_result_complete () from /usr/lib/x86_64-linux-gnu/libgio-2.0.so.0
 ?? () from /usr/lib/x86_64-linux-gnu/libgio-2.0.so.0
Title: shotwell crashed with SIGSEGV in g_simple_async_result_complete()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm audio cdrom dip lpadmin plugdev sambashare sudo

Philip Khor (philip10khor) wrote :

StacktraceTop:
 publishing_accounts_uoa_publisher_authenticator_on_processed () from /tmp/apport_sandbox_ohnuln/usr/lib/shotwell/plugins/builtin/shotwell-publishing.so
 _publishing_accounts_uoa_publisher_authenticator_on_processed_signon_auth_session_process_cb () from /tmp/apport_sandbox_ohnuln/usr/lib/shotwell/plugins/builtin/shotwell-publishing.so
 process_async_cb_wrapper (object=<optimized out>, res=0x7f848c00e2e0, user_data=0x2eb8740) at signon-auth-session.c:243
 ?? ()
 ?? ()

Changed in shotwell (Ubuntu):
importance: Undecided → Medium
summary: - shotwell crashed with SIGSEGV in g_simple_async_result_complete()
+ shotwell crashed with SIGSEGV in
+ publishing_accounts_uoa_publisher_authenticator_on_processed()
tags: removed: need-amd64-retrace
Ubuntu QA Website (ubuntuqa) wrote :

This bug has been reported on the Ubuntu Package testing tracker.

A list of all reports related to this bug can be found here:
http://packages.qa.ubuntu.com/qatracker/reports/bugs/1225629

tags: added: package-qa-testing
Launchpad Janitor (janitor) wrote :

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

Changed in shotwell (Ubuntu):
status: New → Confirmed
Sebastien Bacher (seb128) wrote :

Thank you for your bug report.

Hey Mardy, do you think you could have a look? That seems an issue in uoa patch we carry for shotwell

information type: Private → Public
Changed in shotwell (Ubuntu):
assignee: nobody → Alberto Mardegan (mardy)
importance: Medium → High
Sebastien Bacher (seb128) wrote :

There is a better stacktrace on the errors.ubuntu.com report: https://errors.ubuntu.com/problem/e678dd3bb4c4325a55d018bb42eccadaf32a5633

Alberto Mardegan (mardy) wrote :

Unfortunately the most important part (the code from shotwell) is missing debugging information. I'll try to investigate anyway; it looks like Online Accounts is returning an error, and the code which I wrote in shotwell doesn't handle it well.

Alberto Mardegan (mardy) wrote :

Hi! I tried to reproduce this bug, but with no luck. Philip (and others who can reproduce this), can you please give me some more information about when this precisely happens? Is it consistently reproducible? Does it happen when you click on "Login", or before that?
If you can reproduce it reliably, can you please try reproducing it after settings the following environment variables:

export SHOTWELL_LOG=1
export G_MESSAGES_DEBUG=all

Then you'll find the logs in ~/.cache/shotwell/shotwell.log

Incidentally, I found an issue in libsignon-glib while reproducing this bug, but I don't think it's the cause of this:
http://code.google.com/p/accounts-sso/source/detail?r=cab8e7788b3f2fccad97144d0603938720a25d0e&name=unref-critical&repo=libsignon-glib#

Changed in shotwell (Ubuntu):
status: Confirmed → Incomplete
Philip Khor (philip10khor) wrote :

Hi,
Strangely I can't reproduce it even by publishing the same photos (though using a different FB account). Unfortunately I can't remember precisely how this happened. My apologies.

I'll try again ASAP to see if I can reproduce this bug using other methods.

Simon Lyall (simon-darkmere) wrote :

I just had a crash. Happened the instant I tried to publish to flickr. I upgraded to 14.10 yesterday and have had several crashes since, uploads seem to work about 50% of the time.

I ran with:

export SHOTWELL_LOG=1
export G_MESSAGES_DEBUG=all
shotwell &

attached shotwell.log but not all details seem to be there.

Simon Lyall (simon-darkmere) wrote :

Forgot to post crash message.

shotwell:21178): Gtk-DEBUG: Connecting to session manager
*** Error in `shotwell': corrupted double-linked list: 0x000000000890d7d0 ***

Jim Nelson (yorba-jim) wrote :

The Flickr double-free crash is reported at bug #1382843 (and easily reproduced). I don't know that it's the same as the bug reported for this ticket, so I'm not marking as a duplicate.

On Tue, 2 Dec 2014, Jim Nelson wrote:
> The Flickr double-free crash is reported at bug #1382843 (and easily
> reproduced). I don't know that it's the same as the bug reported for
> this ticket, so I'm not marking as a duplicate.

Looking at #1382843 this is almost certainly what I am seeing rather than
this one.

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Duplicates of this bug

Other bug subscribers