After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 328126 - Eject fails and gives a blank error dialog
Eject fails and gives a blank error dialog
Status: RESOLVED FIXED
Product: gnome-vfs
Classification: Deprecated
Component: File operations
cvs (head)
Other Linux
: Normal normal
: ---
Assigned To: gnome-vfs maintainers
gnome-vfs maintainers
Depends on:
Blocks:
 
 
Reported: 2006-01-22 03:39 UTC by Ryan P Skadberg
Modified: 2006-01-24 12:49 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
Initial work on fix (1.89 KB, patch)
2006-01-23 15:22 UTC, Alexander Larsson
none Details | Review
proposed patch (8.33 KB, patch)
2006-01-24 06:26 UTC, David Zeuthen (not reading bugmail)
committed Details | Review

Description Ryan P Skadberg 2006-01-22 03:39:03 UTC
Tried to eject a CD, got this on the command line:

(rhythmbox:1121): GLib-CRITICAL **: g_str_has_prefix: assertion `str != NULL' failed
closing

And then the CD did not eject.  And I got a blank error dialog, just the stop sign .
Comment 1 William Jon McCann 2006-01-22 17:49:44 UTC
Backtrace was generated from '/home/gnome/bin/rhythmbox'

Using host libthread_db library "/lib/libthread_db.so.1".
`shared object read from target memory' has disappeared; keeping its symbols.
[Thread debugging using libthread_db enabled]
[New Thread -1208346944 (LWP 13663)]
[New Thread 48348080 (LWP 4824)]
[New Thread 132328368 (LWP 26466)]
[New Thread 36330416 (LWP 13669)]
[New Thread 57801648 (LWP 13665)]
[New Thread 82652080 (LWP 13664)]
0x00e83402 in __kernel_vsyscall ()

Thread 1 (Thread -1208346944 (LWP 13663))

  • #0 __kernel_vsyscall
  • #1 __waitpid_nocancel
    from /lib/libpthread.so.0
  • #2 libgnomeui_segv_handle
    at gnome-ui-init.c line 786

Comment 2 James "Doc" Livingston 2006-01-23 04:19:55 UTC
I'd say that this is a gnome-vfs issue, since all RB does is call gnome_vfs_volume_eject with the GnomeVFSVolume of the audio cd. The error dialog is shown if the "suceeded" value of the callback is false, and it contains whatever gnome-vfs gives us for the error message.
Comment 3 Alexander Larsson 2006-01-23 15:22:52 UTC
Created attachment 57946 [details] [review]
Initial work on fix

This patch is a step on the way. It makes the "strange" hal volumes such as burn:/// and cdda:/// not be GNOME_VFS_VOLUME_TYPE_MOUNTPOINT.

It requires more work to make sure blank cds (burn:) eject still works. Also, it needs testing.
Comment 4 David Zeuthen (not reading bugmail) 2006-01-24 06:26:43 UTC
Created attachment 57989 [details] [review]
proposed patch

Here's a patch to do this. Tested with blank discs, should also work with audio discs but I don't have any right here. Please review.

Oh, if you're using gnome-mount you'll need a fix to hal to actually eject - get

 http://freedesktop.org/~david/20-storage-methods.fdi

and put it in /usr/share/hal/fdi/policy/10osvendor/20-storage-methods.fdi - no need to restart hald, just reinsert the disc.
Comment 5 Alexander Larsson 2006-01-24 08:14:41 UTC
I'd prefer a more general way for backends to allow them to implement eject and unmount, like a function pointer in the volume object. But this will do for now. Please commit.
Comment 6 David Zeuthen (not reading bugmail) 2006-01-24 12:49:58 UTC
Right. I've committed this patch. Thanks.