Sony-Ericsson K750i mobile: VFAT fs not detected by blkid

[Gábor Pápai]

Binary package hint: gvfs

my Ubuntu 9.10 was upgraded from 9.04, earlier it was 8.10, 8.04... if it counts. It worked under 9.04.

ProblemType: Bug
Architecture: i386
Date: Wed Dec 2 18:29:35 2009
DistroRelease: Ubuntu 9.10
HotplugNewDevices: /dev/sdb1 /dev/sdb
HotplugNewMounts:

NonfreeKernelModules: nvidia
Package: gvfs 1.4.1-0ubuntu1
ProcEnviron:
 LANG=hu_HU.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-15.50-generic
SourcePackage: gvfs
Symptom: storage
Uname: Linux 2.6.31-15-generic i686

[Gábor Pápai (gaborgabor)]

I can manually mount it ("mount -t vfat /dev/sdb1 ~/temp") and use, but auto-mount still not working.

[Martin Pitt (pitti)]

Looks like the device was properly detected as a music player by devicekit-disks.

Can you please open a Terminal, do

  gvfs-mount -oi

then plug in the device, wait a couple of seconds, then press Control-C and copy&paste the entire output here? (It's likely more than a screenful, so please scroll up). Thanks!

[Gábor Pápai (gaborgabor)]

It outputs nothing! I have waited for a minute after attaching the k750i. Checked with 2 different cables on 2 computers. The dmesg shows that the k750i has been attached. (the same as the "CurrentDmesg.txt" attachment's trail)

xyz@xyz:~$ gvfs-mount -oi
Monitoring events. Press Ctrl+C to quit.
^C

[Martin Pitt (pitti)]

OK, seems there is something wrong with the volume monitor then. If you do "pidof gvfs-gdu-volume-monitor" in a Terminal, do you get a number as output or nothing?

- If you got a number, copy&paste this command into a terminal:

   strace -vvs 1024 -f -o /tmp/gdu.trace `pidof gvfs-gdu-volume-monitor`

- If you did not get a number, do

  strace -vvs 1024 -f -o /tmp/gdu.trace /usr/lib/gvfs/gvfs-gdu-volume-monitor

Then plug in the drive, wait a little for the output to settle down; press Control-C, and attach /tmp/gdu.trace here.

Thanks!

[Gábor Pápai (gaborgabor)]

xyz@xyz:~$ pidof gvfs-gdu-volume-monitor
2092
xyz@xyz:~$ strace -vvs 1024 -f -o /tmp/gdu.trace -p `pidof gvfs-gdu-volume-monitor`
Process 2092 attached - interrupt to quit
^CProcess 2092 detached

Thanks!

[Martin Pitt (pitti)]

So the gdu monitor is running, and the strace shows that it did get the DeviceKit-disks messages about the new device, and that /dev/sdb1 is not in /etc/fstab. I'm quite puzzled now.

Can you please plug in the device and copy&paste the output of

  gvfs-mount -li

?

[Martin Pitt (pitti)]

Also, once you plug it in, can you please check whether

  devkit-disks --mount /dev/sdb1

mounts it? Any output from this command?

[Gábor Pápai (gaborgabor)]

"gvfs-mount -li" 's output (it's the same whether the mobile is plugged in or not)

Drive(0): Hajlékonylemezes meghajtó
  Type: GProxyDrive (GProxyVolumeMonitorGdu)
  ids:
   unix-device: '/dev/fd0'
  themed icons: [drive-removable-media-floppy] [drive-removable-media] [drive-removable] [drive]
  is_media_removable=1
  has_media=0
  is_media_check_automatic=0
  can_poll_for_media=1
  can_eject=0
  can_start=0
  can_stop=0
  start_stop_type=unknown
Volume(0): cdrom0
  Type: GProxyVolume (GProxyVolumeMonitorGdu)
  themed icons: [drive-optical] [drive]
  can_mount=1
  can_eject=0
  should_automount=0

[Gábor Pápai (gaborgabor)]

xyz@xyz:~$ devkit-disks --mount /dev/sdb1
Mount failed: Error mounting: mount exited with exit code 32: mount: you must specify the filesystem type

xyz@xyz:~$ devkit-disks --mount /dev/sdb1 --mount-fstype vfat
Mount failed: Not a mountable file system

xyz@xyz:~$ sudo mount -t vfat /dev/sdb1 ~/temp
xyz@xyz:~$ ls ~/temp
dcim memstick.ind MemStickInfo.txt mp3 mssemc Traceability.txt

[Martin Pitt (pitti)]

Ah, sorry for missing this in the first place. /dev/sdb1 has no ID_FS_* udev properties at all. Can you please give me the output of

  sudo blkid -p /dev/sdb1

with the device being plugged in? Thanks!

[Gábor Pápai (gaborgabor)]

It outputs nothing.

xyz@xyz:~/CPT/TRUNK/Servers/CxPITS_2/bin$ sudo blkid -p /dev/sdb1
xyz@xyz:~/CPT/TRUNK/Servers/CxPITS_2/bin$

[Martin Pitt (pitti)]

Thanks. Can you please do

  sudo BLKID_DEBUG=0xffff blkid -l /dev/sdb1

and post the output here?

[Gábor Pápai (gaborgabor)]

xyz@xyz:~$ sudo BLKID_DEBUG=0xffff blkid -l /dev/sdb1
libblkid: debug mask set to 0xffff.
creating blkid cache (using default cache)
reading config file: /etc/blkid.conf.
reading cache file /dev/.blkid.tab
line: <device DEVNO="0x0801" TIME="1267022879" UUID="a1d3b798-e056-4499-9211-3e33efc1525e" TYPE="ext4">/dev/sda1</device>

found device header: <device DEVNO="0x0801" TIME="1267022879" UUID="a1d3b798-e056-4499-9211-3e33efc1525e" TYPE="ext4">/dev/sda1</device>
device should be /dev/sda1</device>
found device trailer </device>
found dev /dev/sda1
    tag: DEVNO="0x0801"
    tag: TIME="1267022879"
    creating new cache tag head UUID
    tag: UUID="a1d3b798-e056-4499-9211-3e33efc1525e"
    creating new cache tag head TYPE
    tag: TYPE="ext4"
  dev: name = /dev/sda1
  dev: DEVNO="0x801"
  dev: TIME="1267022879"
  dev: PRI="0"
  dev: flags = 0x00000000
    tag: UUID="a1d3b798-e056-4499-9211-3e33efc1525e"
    tag: TYPE="ext4"

line: <device DEVNO="0x0805" TIME="1267022879" UUID="6abf783e-0536-4718-81d2-3a51cf2d7810" TYPE="swap">/dev/sda5</device>

found device header: <device DEVNO="0x0805" TIME="1267022879" UUID="6abf783e-0536-4718-81d2-3a51cf2d7810" TYPE="swap">/dev/sda5</device>
device should be /dev/sda5</device>
found device trailer </device>
found dev /dev/sda5
    tag: DEVNO="0x0805"
    tag: TIME="1267022879"
    found cache tag head UUID
    tag: UUID="6abf783e-0536-4718-81d2-3a51cf2d7810"
    found cache tag head TYPE
    tag: TYPE="swap"
  dev: name = /dev/sda5
  dev: DEVNO="0x805"
  dev: TIME="1267022879"
  dev: PRI="0"
  dev: flags = 0x00000000
    tag: UUID="6abf783e-0536-4718-81d2-3a51cf2d7810"
    tag: TYPE="swap"

The lookup option requires a search type specified using -t

[Martin Pitt (pitti)]

Argh, sorry, I meant

  sudo BLKID_DEBUG=0xffff blkid -p /dev/sdb1

i. e. "-p", not "-l".

[Gábor Pápai (gaborgabor)]

xyz@xyz:~$ sudo BLKID_DEBUG=0xffff blkid -p /dev/sdb1
libblkid: debug mask set to 0xffff.
reseting blkid_probe
ready for low-probing, offset=0, size=0
--> starting probing loop [idx=-1]
linux_raid_member: call probefunc()
ddf_raid_member: call probefunc()
isw_raid_member: call probefunc()
lsi_mega_raid_member: call probefunc()
via_raid_member: call probefunc()
silicon_medley_raid_member: call probefunc()
nvidia_raid_member: call probefunc()
promise_fasttrack_raid_member: call probefunc()
highpoint_raid_member: call probefunc()
adaptec_raid_member: call probefunc()
jmicron_raid_member: call probefunc()
vfat: magic sboff=0, kboff=0
vfat: call probefunc()
ufs: call probefunc()
sysv: call probefunc()
<-- leaving probing loop (failed) [idx=49]

[Martin Pitt (pitti)]

OK, so this seems to be a confirmed bug in blkid. It doesn't recognize the VFAT partition, although mount -t vfat seems to work just fine.

[Gábor Pápai (gaborgabor)]

further investigation:
the mobile reads its memory card (memory stick pro duo) when attached to the computer, so I tried this memory card in a card reader and the result is the same: it's not automounted. (but eg. my gps's 2gb sd card with fat32 filesystem is automounted). It means to me that somehow the filesystem is "faulty" or ubuntu can no more auto-detect its filesystem.

# file -s /dev/sdf
/dev/sdf: x86 boot sector; partition 1: ID=0x6, active, starthead 2, startsector 39, 124377 sectors, code offset 0x0

attachment: the first 40*512 bytes of the memory card, containing the file system signature: "FAT ". I hope it helps somewhat.

[Gábor Pápai (gaborgabor)]

tadam!

analysis:
the "FAT " identification string isn't supported, but if I changed it on the memory card to "FAT16 " then it's automounted.

suggested solution:
add the "FAT " to identify as vfat (into probe.c's blkid_magic type_array ?)

[Jeroen Oortwijn (idefix)]

Thank you for your investigation. I've submitted your suggested solution to upstream (see http://article.gmane.org/gmane.linux.utilities.util-linux-ng/3639 ).

[Jeroen Oortwijn (idefix)]

Upstream committed the patch (see: http://git.kernel.org/?p=utils/util-linux-ng/util-linux-ng.git;a=commitdiff;h=24d1e474a2d7db57968d406265460a97f6143854 ). If/when Ubuntu will incorporate this fix, I don't know.

[Jeroen Oortwijn (idefix)]

Ubuntu 11.10 should have incorporated this fix.