FS#19070 - [drivel] crash on connecting to blogger.com service
Attached to Project:
Community Packages
Opened by Victor (DELIRIUM) - Tuesday, 13 April 2010, 06:14 GMT
Last edited by Sergej Pupykin (sergej) - Monday, 01 November 2010, 10:24 GMT
Opened by Victor (DELIRIUM) - Tuesday, 13 April 2010, 06:14 GMT
Last edited by Sergej Pupykin (sergej) - Monday, 01 November 2010, 10:24 GMT
|
Details
Description:
On trying to log in to blogger.com service (either blogger 1.0 or Atom/Blogger 2.0) the application crashes with floating point exception. Additional info: * package version(s) $ pacman -Qi drivel Name : drivel Version : 3.0.0-5 URL : http://www.dropline.net/drivel/index.php Licenses : GPL Groups : None Provides : None Depends On : curl gtksourceview2 gtkspell libgnomeui>=2.0.0 libsoup Optional Deps : None Required By : None Conflicts With : None Replaces : None Installed Size : 1328.00 K Packager : Sergej Pupykin <pupykin.s+arch@gmail.com> Architecture : x86_64 Build Date : Thu 05 Nov 2009 05:26:24 PM MSK Install Date : Tue 13 Apr 2010 09:57:56 AM MSD Install Reason : Explicitly installed Install Script : Yes Description : GTK client for working with online journals (blogs). * config and/or log files etc. Two log files (output of -v option) and core dumps attached Steps to reproduce: 1) Launch drivel. 2) Type in username and password 3) Select "Blogger 1.0" or "Atom/Blogger 2.0" in "Journal type" field. 4) Click "Log in" Application crashes with floating point exception. The following appears in dmesg: drivel[7808] trap divide error ip:42d471 sp:7fff277929a0 error:0 in drivel[400000+3d000] Note: when selecting "Blogger 1.0" it is possible to either enter valid or invalid login/pass - the crash occurs in both cases. When entering invalid credentials for Blogger 2.0 the message about invalid login appears but nothing crashes, with valid login/pass it crashes. System info: list of currently installed packages attached. |
This task depends upon
Closed by Sergej Pupykin (sergej)
Monday, 01 November 2010, 10:24 GMT
Reason for closing: Fixed
Additional comments about closing: it looks working somehow...
Monday, 01 November 2010, 10:24 GMT
Reason for closing: Fixed
Additional comments about closing: it looks working somehow...
Starting program: /usr/bin/drivel
[Thread debugging using libthread_db enabled]
Program received signal SIGSEGV, Segmentation fault.
0xb7eb9de9 in soup_uri_set_port () from /usr/lib/libsoup-2.4.so.1
(gdb) bt
#0 0xb7eb9de9 in soup_uri_set_port () from /usr/lib/libsoup-2.4.so.1
#1 0x0807486e in net_start_session ()
#2 0x08073088 in main ()
I think you should report to upstream
> I have recently faced a problem with running drivel 3.0.0.5 on Arch
Linux: on trying to log in to blogger.com service (either blogger 1.0 or
Atom/Blogger 2.0) the application crashes with floating point exception.
>
> I have posted a bug report to Arch Linux bug-tracker:
>
> http://bugs.archlinux.org/task/19070
The Arch Linux data for the package is out of date.
Homepage: http://drivel.sourceforge.net/
Bug tracker: http://bugzilla.gnome.org/browse.cgi?product=drivel
SourceForge: http://sourceforge.net/projects/drivel/
Freshmeat: http://freshmeat.net/projects/drivel/
Other info: http://linux.codehelp.co.uk/?q=drivel
> On trying to log in to blogger.com service (either blogger 1.0 or
Atom/Blogger 2.0) the application crashes with floating point exception.
Would appear to be a repeat of this bug:
https://bugzilla.gnome.org/show_bug.cgi?id=449198
> URL : http://www.dropline.net/drivel/index.php
Please update that URL.
> * config and/or log files etc.
> Two log files (output of -v option) and core dumps attached
Attachments are stripped from the mailing list. If you have a
backtrace, attach it to the bug report.
http://wiki.debian.org/HowToGetABacktrace
> I didn't find any drivel bug tracker and decided to send a message to
this list.
The package maintainer needs to update the package information.
Drivel 3.0.0 works fine on other systems, so it isn't necessarily an
upstream bug that causing the failure to start reported but this
blogger issue does seem to be upstream. I just don't have a fix.
Patches welcome...