Bug 1075580

+++ This bug was initially created as a clone of Bug #1058082 +++

Version-Release number of selected component:
gnome-boxes-3.11.4-1.fc21

Additional info:
reporter:       libreport-2.1.11
backtrace_rating: 4
cmdline:        gnome-boxes
crash_function: boxes_installer_media_get_decoded_udev_property
executable:     /usr/bin/gnome-boxes
kernel:         3.13.0-1.fc21.x86_64
runlevel:       N 5
type:           CCpp
uid:            1001

Truncated backtrace:
Thread no. 1 (10 frames)
 #5 boxes_installer_media_get_decoded_udev_property at installer-media.c:3017
 #6 boxes_installer_media_get_decoded_udev_properties_for_media at /home/zeenix/checkout/gnome/gnome-boxes/src/installer-media.vala:224
 #7 boxes_installer_media_get_media_info_from_device_co at /home/zeenix/checkout/gnome/gnome-boxes/src/installer-media.vala:207
 #8 boxes_installer_media_get_media_info_from_device at /home/zeenix/checkout/gnome/gnome-boxes/src/installer-media.vala:7
 #9 boxes_installer_media_construct_for_path_co at /home/zeenix/checkout/gnome/gnome-boxes/src/installer-media.vala:69
 #10 g_simple_async_result_complete at gsimpleasyncresult.c:770
 #11 boxes_installer_media_get_device_from_path_co at /home/zeenix/checkout/gnome/gnome-boxes/src/installer-media.vala:185
 #12 g_task_return_now at gtask.c:1109
 #13 complete_in_idle_cb at gtask.c:1118
 #17 g_main_context_iteration at gmain.c:3775

--- Additional comment from Christoph Frieben on 2014-01-26 17:10:52 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:10:54 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:10:57 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:11:00 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:11:02 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:11:04 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:11:08 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:11:10 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:11:12 EST ---



--- Additional comment from Christoph Frieben on 2014-01-26 17:11:14 EST ---



--- Additional comment from Christophe Fergeau on 2014-01-27 05:39:02 EST ---

This fails on a fortify check in get_decoded_udev_property()

Looking at the code, aren't we going one byte past the end of the string in
if (encoded[i:encoded.length].scanf ("\\x%02x", out x) > 0) {
?

--- Additional comment from Zeeshan Ali on 2014-01-27 08:31:23 EST ---

(In reply to Christophe Fergeau from comment #11)
> This fails on a fortify check in get_decoded_udev_property()
> 
> Looking at the code, aren't we going one byte past the end of the string in
> if (encoded[i:encoded.length].scanf ("\\x%02x", out x) > 0) {

No, the M is exclusive in [N:M] syntax: https://wiki.gnome.org/Projects/Vala/Tutorial#Arrays

--- Additional comment from Igor Gnatenko on 2014-02-04 05:55:07 EST ---

Got today

--- Additional comment from Zeeshan Ali on 2014-02-12 07:58:41 EST ---

Would be much easier to debug this if I could reproduce it so I need some info: Is there any ISOs mounted or USB/ISO device inserted when you reproduce this? If so, can you please provide output of ` udevadm info --query=property DEVICE_PATH` ?

--- Additional comment from Christoph Frieben on 2014-02-12 15:26:55 EST ---



--- Additional comment from Christoph Frieben on 2014-02-12 15:32:53 EST ---

Issue has occurred just half an hour ago for installed packages:

  kernel-3.14.0-0.rc2.git0.2.fc21
  gnome-boxes-3.11.5.1-2.fc21

Two removable devices were atteched to the system: a USB stick listed as /dev/sdb and an SD card listed as /dev/mmcblk0p1.

After removing both of them, gnome-boxes still crashed at start-up. After tunrning the system off and on again, gnome-boxes did not crash anymore regardless of whether any of the two devices was present at boot time after a warm boot of the system or not.

--- Additional comment from Christoph Frieben on 2014-02-12 21:34:40 EST ---

Even if gnome-boxes does not crash any more, this observation is of little interest given that the application is now stuck at the source selection stage.

--- Additional comment from Zeeshan Ali on 2014-02-17 09:16:08 EST ---

(In reply to Christoph Frieben from comment #17)
> Created attachment 862609 [details]
> Source selection of GNOME Boxes
> 
> Even if gnome-boxes does not crash any more, this observation is of little
> interest given that the application is now stuck at the source selection
> stage.

Its still very relevant and this new issue could be an entirely different bug. Could you please get a backtrace of all threads when its stuck?

--- Additional comment from Zeeshan Ali on 2014-02-17 09:18:51 EST ---

(In reply to Zeeshan Ali from comment #18)
> (In reply to Christoph Frieben from comment #17)
> > Created attachment 862609 [details]
> > Source selection of GNOME Boxes
> > 
> > Even if gnome-boxes does not crash any more, this observation is of little
> > interest given that the application is now stuck at the source selection
> > stage.
> 
> Its still very relevant and this new issue could be an entirely different
> bug. Could you please get a backtrace of all threads when its stuck?

Oh nm, I didn't see the screenshot. Its not stuck but rather some important UI components are for some reason not shown. I think I fixed this not so long ago in unstable branch.

--- Additional comment from Zeeshan Ali on 2014-02-17 09:21:36 EST ---

(In reply to Zeeshan Ali from comment #19)
> (In reply to Zeeshan Ali from comment #18)
> > (In reply to Christoph Frieben from comment #17)
> > > Created attachment 862609 [details]
> > > Source selection of GNOME Boxes
> > > 
> > > Even if gnome-boxes does not crash any more, this observation is of little
> > > interest given that the application is now stuck at the source selection
> > > stage.
> > 
> > Its still very relevant and this new issue could be an entirely different
> > bug. Could you please get a backtrace of all threads when its stuck?
> 
> Oh nm, I didn't see the screenshot. Its not stuck but rather some important
> UI components are for some reason not shown. I think I fixed this not so
> long ago in unstable branch.

Anyway, its a differen bug so please file it separately.

(In reply to Christoph Frieben from comment #16)
> Created attachment 862530 [details]
> Output of 'udevadm info --query=property /dev/mmcblk0p1'

The output of  /dev/sdb would be more helpful as this device is not a bootable ISO9660 media.

--- Additional comment from Igor Gnatenko on 2014-02-28 03:09:42 EST ---



--- Additional comment from Christoph Frieben on 2014-03-02 04:44:47 EST ---

Issue still present for gnome-boxes-3.11.90.1-2.fc21 when a bootable CD is inserted in the CDV drive.

--- Additional comment from Christoph Frieben on 2014-03-09 05:39:45 EDT ---

Issue still present for gnome-boxes-3.11.91-1.fc21 when a bootable CD is inserted in the CDV drive.