Scheduled Maintenance: We are aware of an issue with Google, AOL, and Yahoo services as email providers which are blocking new registrations. We are trying to fix the issue and we have several internal and external support tickets in process to resolve the issue. Please see: viewtopic.php?t=158230

 

 

 

Buster Android MTP

User discussion about Debian Development, Debian Project News and Announcements. Not for support questions.
Post Reply
Message
Author
appye
Posts: 33
Joined: 2007-11-21 16:26
Has thanked: 1 time
Been thanked: 6 times

Buster Android MTP

#1 Post by appye »

Hello, I am unsure what subforum is a good choice for this. I chose development because I am playing with testing.

With Buster's move to stable coming just around the corner, I have been putting together a desktop installation procedure for myself, friends and family. I start with the default Cinnamon desktop as a base and go from there. I have software choices and configs pretty well ironed out. At this point I will pretty much wait until the official announcement of Debian 10 before installing to our machines...

The only thing I have left is MTP and Android connectivity.

I get similar behavior on an LG G2, a Samsung Galaxy s7, and a Motorola Droid Turbo 2 ...

Using my S7 as an example, once I plug a device into a USB port, I get the following prompt:
Image
Unable to open folder for SAMSUNG Android
Cache invalid, retry (internally handled)

If an instance of Nemo (the default Cinnamon file manager) is NOT running, then once I enable MTP, Nemo pops up and shows me the Devices filesystems. If Nemo IS running, whether in the foreground or background, I get one of the following prompts, I have not seen any kind of pattern to it, though I mostly get the second, more specific prompt:
Image
Unable to open a folder for SAMSUNG Android
No such interface "org.gtk.vfs.Mount" on object at path /org/gtk/vfs/mount/1

Image
Count not display "mtp://SAMSUNG_SAMSUNG_Android_98545a76/"
Error: No such interface "org.gtk.vfs.Mount" on object at path /org/gtk/vfs/mount/1
Please select another viewer and try again.

... In all cases, I am able to browse and manipulate the filesystem without any problems, regardless of whether or not I acknowledge the error prompts. I don't know where to begin with this. I have looked up MTP issues all over the internet and have not found this specific issue. I see a lot of people talking about errors mounting (is MTP ever really "mounted?") and being unable to see the device filesytem at all, etc. None where it seems to just be cosmetic error prompts. I do not (and would like to avoid, if I can) installation of any specific MTP software, as it seems to be handled well enough (except for the errors) with Debian default auto installed choices. Is this specific to Cinnamon? Is it something underlying?

User avatar
Head_on_a_Stick
Posts: 14114
Joined: 2014-06-01 17:46
Location: London, England
Has thanked: 81 times
Been thanked: 132 times

Re: Buster Android MTP

#2 Post by Head_on_a_Stick »

appye wrote:Is this specific to Cinnamon?
Could be. Have you tried GNOME?

That's the default desktop so it's had more testing.

Perhaps a bug report is called for here, you can use the reportbug package to check for any that are open against gvfs-backends package.
deadbang

appye
Posts: 33
Joined: 2007-11-21 16:26
Has thanked: 1 time
Been thanked: 6 times

Re: Buster Android MTP

#3 Post by appye »

After some repartitioning, I installed default MATE and GNOME desktop versions of Buster. For all three installs (Mate, Gnome and Cinnamon), I performed an expert install using the firmware-testing-amd64-netinst.iso (timestamp 2019-04-15 08:43) so I could use the wifi hardware on this particular laptop. Aside from partitioning, the only default I changed was to not allow root login, opting for usage of sudo instead.

No errors occur in Gnome. Mate does exhibit the same MTP error, but not the cache error. Again, in all cases, I am able to browse and manipulate the device's filesystem without any problems, regardless of whether or not I acknowledge the error prompts.

When I use reportbug, it states that I should try and see if the issue is fixed by installing the newer package from experimental. This is a test machine so I went ahead and installed the experimental gvfs deb packages. I had to get some dependencies to make apt happy, and it made some rather interesting choices make it happen, borking the install in the process, no longer allowing me to run Cinnamon ...

Code: Select all

appyecinnamon@BusterCinnamon:~/Downloads$ ls -al
total 1220
drwxr-xr-x 2 appyecinnamon appyecinnamon 4096 Apr 21 22:20 .
drwxr-xr-x 16 appyecinnamon appyecinnamon 4096 Apr 21 22:14 ..
-rw-r--r-- 1 appyecinnamon appyecinnamon 129556 Apr 21 22:15 gvfs_1.40.1-1_amd64.deb
-rw-r--r-- 1 appyecinnamon appyecinnamon 355236 Apr 21 22:17 gvfs-backends_1.40.1-1_amd64.deb
-rw-r--r-- 1 appyecinnamon appyecinnamon 503296 Apr 21 22:19 gvfs-common_1.40.1-1_all.deb
-rw-r--r-- 1 appyecinnamon appyecinnamon 133996 Apr 21 22:18 gvfs-daemons_1.40.1-1_amd64.deb
-rw-r--r-- 1 appyecinnamon appyecinnamon 112704 Apr 21 22:20 gvfs-libs_1.40.1-1_amd64.deb
appyecinnamon@BusterCinnamon:~/Downloads$ sudo apt install ./gvfs*.deb
[sudo] password for appyecinnamon:
Reading package lists... Done
Building dependency tree
Reading state information... Done
Note, selecting 'gvfs' instead of './gvfs_1.40.1-1_amd64.deb'
Note, selecting 'gvfs-backends' instead of './gvfs-backends_1.40.1-1_amd64.deb'
Note, selecting 'gvfs-common' instead of './gvfs-common_1.40.1-1_all.deb'
Note, selecting 'gvfs-daemons' instead of './gvfs-daemons_1.40.1-1_amd64.deb'
Note, selecting 'gvfs-libs' instead of './gvfs-libs_1.40.1-1_amd64.deb'
The following packages were automatically installed and are no longer required:
cjs gir1.2-appindicator3-0.1 gir1.2-caribou-1.0 gir1.2-clutter-1.0 gir1.2-cmenu-3.0 gir1.2-cogl-1.0
gir1.2-coglpango-1.0 gir1.2-gdesktopenums-3.0 gir1.2-gnomedesktop-3.0 gir1.2-gtkclutter-1.0
gir1.2-json-1.0 gir1.2-keybinder-3.0 gir1.2-meta-muffin-0.0 gir1.2-nm-1.0 gir1.2-nma-1.0
gir1.2-upowerglib-1.0 gnome-backgrounds libappindicator3-1 libcaribou-common libcaribou0 libcjs0
libkeybinder-3.0-0 libmozjs-52-0 metacity-common python3-pampy python3-pyinotify sgml-base
Use 'sudo apt autoremove' to remove them.
Suggested packages:
samba-common
The following packages will be REMOVED:
blueman cinnamon cinnamon-common cinnamon-core cinnamon-desktop-environment gvfs-bin gvfs-fuse
task-cinnamon-desktop
The following packages will be upgraded:
gvfs gvfs-backends gvfs-common gvfs-daemons gvfs-libs
5 upgraded, 0 newly installed, 8 to remove and 0 not upgraded.
Need to get 0 B/1,235 kB of archives.
After this operation, 13.1 MB disk space will be freed.
Do you want to continue? [Y/n]
I ended up restoring a Timeshift snapshot ... I suppose I could try with dpkg, but I am guessing that will be less than optimal ...

So I guess before I report a bug, I should ask, does this still point to gvfs-backends?

appye
Posts: 33
Joined: 2007-11-21 16:26
Has thanked: 1 time
Been thanked: 6 times

Re: Buster Android MTP

#4 Post by appye »

Well, I went ahead and posted bug report #928058 ... I hope this is will be fixed. I guess I can attempt a SID install to see if this bug is squashed upstream and report back.

User avatar
4D696B65
Site admin
Site admin
Posts: 2696
Joined: 2009-06-28 06:09
Been thanked: 85 times

Re: Buster Android MTP

#5 Post by 4D696B65 »

s9+ on gnome sid, mtp works

appye
Posts: 33
Joined: 2007-11-21 16:26
Has thanked: 1 time
Been thanked: 6 times

Re: Buster Android MTP

#6 Post by appye »

4D696B65 wrote:s9+ on gnome sid, mtp works
Yep. It works in Buster with no errors as well. The issue comes up in Mate and Cinnamon...

appye
Posts: 33
Joined: 2007-11-21 16:26
Has thanked: 1 time
Been thanked: 6 times

Re: Buster Android MTP

#7 Post by appye »

Okay, identical install on SID and the problem still exists. I am not sure how to upgrade to the experimental packages but I will figure it out.

appye
Posts: 33
Joined: 2007-11-21 16:26
Has thanked: 1 time
Been thanked: 6 times

Re: Buster Android MTP

#8 Post by appye »

Okay, I can confirm that the same error messages pop up on a SID install, even after installing (force, dpkg --force-all) newer packages from experimental:

gvfs_1.40.1-1_amd64.deb
gvfs-backends_1.40.1-1_amd64.deb
gvfs-common_1.40.1-1_all.deb
gvfs-daemons_1.40.1-1_amd64.deb
gvfs-libs_1.40.1-1_amd64.deb

I will add this to the bug report. Are we still sure that the issue is related to gvfs-backends?

User avatar
pylkko
Posts: 1802
Joined: 2014-11-06 19:02

Re: Buster Android MTP

#9 Post by pylkko »

you can also consider not using a cable at all:

http://forums.debian.net/viewtopic.php? ... =+amaze%27

appye
Posts: 33
Joined: 2007-11-21 16:26
Has thanked: 1 time
Been thanked: 6 times

Re: Buster Android MTP

#10 Post by appye »

Yeah I know, I would like to avoid the error messages, and be able to use the cable as well. Also, I have other people I manage linux computers for and I would like to avoid the questions ...

Post Reply