Banshee Crashes on opening 11.04

While any crash is a strong indication of a bug, this bug report does not contain enough specific information for the underlying bug to be investigated, and it appears to have been posted for the purpose of soliciting support, rather than for the purpose of reporting a bug. Ordinarily I would convert this bug to a question, but https://answers.launchpad.net/ubuntu/+source/banshee/+question/154805 has already been posted, so I'm marking this bug as Invalid.

I honestly don't see how anyone could think a bug about Banshee could be duplicate of a *Rhythmbox bug*!!!

If there is not enough information in a bug, you ask for more information, you don't just send it into the thickets.

I'm seeing this also. I just upgraded from 10.10 to 11.04, which seems to have replaced Rhythmbox with Banshee as the default music player. However, it won't start. The main screen is displayed for a second and then disappears again.

This is on 11.04 (obviously) 64-bit, just upgraded from 10.10. I'm using the NVidia proprietary video driver.

Here is the console output from running banshee in a terminal:

[Info 13:46:13.136] Running Banshee 2.0.0: [Ubuntu Natty (development branch) (linux-gnu, x86_64) @ 2011-04-18 16:18:52 UTC]
[Info 13:46:13.997] Updating web proxy from GConf
[Info 13:46:14.065] All services are started 0.766405
** (Banshee:2626): DEBUG: SyncDaemon already running, initializing SyncdaemonDaemon object

(Banshee:2626): libsoup-WARNING **: No feature manager for feature of type 'U1RequestChrome'

** (Banshee:2626): WARNING **: Error calling get_info: Traceback (most recent call last):
  File "/usr/lib/pymodules/python2.7/dbus/service.py", line 702, in _message_cb
    retval = candidate_method(self, *args, **keywords)
  File "/usr/lib/pymodules/python2.7/ubuntuone-client/ubuntuone/platform/linux/dbus_interface.py", line 992, in get_info
    return self.syncdaemon_folders.get_info(path)
  File "/usr/lib/pymodules/python2.7/ubuntuone-client/ubuntuone/syncdaemon/interaction_interfaces.py", line 634, in get_info
    mdobj = self.fs.get_by_path(path.encode('utf-8'))
  File "/usr/lib/pymodules/python2.7/ubuntuone-client/ubuntuone/syncdaemon/filesystem_manager.py", line 773, in get_by_path
    mdid = self._idx_path[path]
KeyError: '/home/pepijn/.ubuntuone/Purchased from Ubuntu One'

** (Banshee:2626): CRITICAL **: syncdaemon_folder_info_get_subscribed: assertion `SYNCDAEMON_IS_FOLDER_INFO (finfo)' failed

** (Banshee:2626): WARNING **: Error rescanning Purchased Music: No such file or directory
** (Banshee:2626): DEBUG: Loading the real store page

** (Banshee:2626): WARNING **: Got less number of items in credentials hash table than expected!
[Info 13:46:14.833] nereid Client Started
[Info 13:46:14.885] GStreamer version 0.10.32.0, gapless: True, replaygain: False
[Info 13:46:14.971] AppleDeviceSource is ignoring unmounted volume Windows Vista

Unhandled Exception: GLib.GException: Can't recursively copy directory
  at GLib.FileAdapter.Move (File destination, FileCopyFlags flags, GLib.Cancellable cancellable, GLib.FileProgressCallback progress_callback) [0x00000] in <filename unknown>:0
  at Banshee.IO.Gio.Directory.Move (Hyena.SafeUri from, Hyena.SafeUri to) [0x00000] in <filename unknown>:0
  at Banshee.IO.Directory.Move (Hyena.SafeUri from, Hyena.SafeUri to) [0x00000] in <filename unknown>:0
  at Banshee.Podcasting.PodcastService.<DelayedInitialize>m__11 () [0x00000] in <filename unknown>:0
[Info 13:46:14.996] AppleDeviceSource is ignoring unmounted volume Ubuntu (old)

Fortunately, Rhythmbox is still installed, and works properly. I don't know why the switch to Banshee was made, but the first si...

The referenced Rhythmbox bug was actually a bug in winbind, which hooked into libc via libnss_wins.so, configured in /etc/nsswitch.conf. This affected DNS lookups in Rhythmbox, Banshee, and Liferea, among other programs. The master bug for that one is Bug #529714. That, at least, is the main cause of Banshee crashes these days.

That said, I don't know if the reason Banshee is crashing for you is the same as the reason it is crashing for Patrick Wilson, who filed this bug. Therefore, I'd suggest you file a new bug (please include your symptoms, steps to reproduce, and the output of "banshee --debug" after the issue has been reproduced). I'm setting this bug to "incomplete", because the relevant information for this bug should come from Patrick.

We can merge the two bugs later on (by setting one as a duplicate of another) if we do find that it is the same, but in the event that the two bugs are different and information from both of you are stored on this bug, it can get really confusing.

That said, Patrick, please run "banshee --debug" in a terminal and post the output here.

After I marked this bug Invalid (since it functioned as a question rather than a bug, and Patrick Wilson had already filed a question **containing the exact same text as this bug report**), Patrick Wilson re-reported his problem as bug 773495, as I had suggested that he do. At that point, he also marked this bug as a duplicate of the new bug 773495. Considering Patrick's description of the crash in combination with the stack trace in bug 773495, bug 773495 seemed a lot like a duplicate of bug 529714. Just comparing bug 529714's stack trace and description (together with comments confirming that it also occurred with Banshee) to bug 773495 would have been sufficient to justify marking bug 773495 as a duplicate of bug 529714. But I decided to be more careful, just in case, so I looked at the stack traces of *all* the public duplicates of bug 529714. All were identical in relevant part (i.e., the winbind calls) to bug 773495, several were from Banshee crashes, and a couple were from Banshee crashes and had completely identical stack traces (with the absence of some symbol names not filled in). This confirmed for sure that Patrick Wilson was experiencing bug 529714. So I marked bug 773495 a duplicate of bug 529714. Since *this* bug (i.e., bug 773390) was at that time marked as a duplicate of bug 773495, marking bug 773495 a duplicate of bug 529714 resulted in this bug being marked a duplicate of bug 529714.

@Chow Loong Jin
First, thank you for helping to explain this situation to Captain Chaos and assuage his concerns. In the unlikely (given the above information) situation that any doubt were to arise about whether or not Patrick Wilson is experiencing bug 529714, discussion of that issue, and requests for information, should be posted in bug 773495. That is Patrick Wilson's actual bug report. This bug is actually a support request, and the only reason I didn't convert it to a question was that an identical question had already been posted. Therefore, I am re-marking this bug as Invalid, and re-marking it as a duplicate of bug 529714. Since bug 773495 contains a stack trace, I doubt you'll still want the output of "banshee --debug" from Patrick, but if you do, I suggest you ask for it there.

@Captain Chaos
As you can see, this bug--not the problem you're experiencing, but the one Patrick is experiencing--really is the same as a bug affecting Rhythmbox. (It's really a bug in samba in Ubuntu, which affects both Rhythmbox and Banshee.) Furthermore, when I marked this bug as Invalid, I provided a link to the question in which the issue was being discussed. If you had consulted that question instead of just assuming I had acted improperly, your initial concerns would have been immediately addressed. In the future, before objecting to changes people make to bugs, I recommend you read their explanations for those changes. For example, I am re-marking this bug as Invalid and as a duplicate of bug 529714, but before doing so I have read your post--and Chow Loong Jin's--and I am responding to both.

For my part, when I marked bug 773495 a duplicate of bug 529714, it would have been helpful if I had posted a comment in this bug explaining that its new status as a duplicate of bug 529714 was correct, since it had originally been a duplicate of bug 773495. If I had done that, it would have made things clearer--then it would not even have been necessary to examine the related question, to understand this bug's status. I am sorry that I did not do that, and I will try to be mindful to take that extra step when similar situations arise in the future.

This bug is a very bad place to work on the problem you are experiencing. Working on it here will likely result in even further confusion. Instead, as Chow Loong Jin said, you should create your own bug report. As he said, if your bug is found to be the same as bug 529714 or some other already-existing bug, then it can be marked a duplicate of such a bug. The best way for you to report your bug would be to enable Apport (https://wiki.ubuntu.com/Apport#How%20to%20enable%20apport), install the package called banshee-dbg, read https://help.ubuntu.com/community/ReportingBugs carefully (if you have not done so already), make sure your system is fully up-to-date, produce the crash, and allow Apport to send the bug data to Launchpad. You can (and should) also include the output of Banshee from the Terminal.