Volumes on cards inserted in the Realtek card reader (PCI 10ec:5208) not automounted.

upstream fd.o has discussion about the auto-mount support of scsi devices. I added the udisks/fd.o bug number.

Keng-Yu, try adding the following line to your udisks rules. I think "/lib/udev/rules.d/80-udisks.rules" would be the apropriate place.

SUBSYSTEMS=="pci", ATTRS{vendor}=="0x10ec", ATTRS{device}=="0x0461", ENV{ID_DRIVE_FLASH_MS}="1"

With the udev rule above, the SD card volume is now automounted. Attached the `udisks -moitor` result.

Ayan: should the ATTRS{device} not be 5208 there? or am I missing some other device that's affecting the automount?

Also, since this driver supports multiple flash card types, ENV{ID_DRIVE_FLASH_MS} might be better as ENV{ID_DRIVE_FLASH}. Keng-Yü: could you test that this works?

Jeremy: Yes -- my mistake. I misread the lspci output in the apport log. I also agree that ID_DRIVE_FLASH is probably more appropriate.

Okay -- I might have been wrong from the start. Kengyu, were you running gnome as root? If so, could you try testing as a non-privileged users?

my testing udev rule. automount does not work with it.

I think I mis-understood the automount behavior in comment #4. I tested with an SD card with FAT partition. The partition shows in Gnome "Places" menu, but in the terminal, `mount` does not shows it mounted in /media.

So I suppose this does not mean that "automount is working".

ubuntu@ubuntu-Latitude-2110:~$ udevadm monitor
monitor will print the received events for:
UDEV - the event which udev sends out after rule processing
KERNEL - the kernel uevent

KERNEL[1303783846.917600] change /devices/pci0000:00/0000:00:1c.2/0000:0d:00.0/host4/target4:0:0/4:0:0:0/block/sdb (block)
KERNEL[1303783846.921212] change /devices/pci0000:00/0000:00:1c.2/0000:0d:00.0/host4/target4:0:0/4:0:0:0/block/sdb (block)
KERNEL[1303783846.921357] add /devices/pci0000:00/0000:00:1c.2/0000:0d:00.0/host4/target4:0:0/4:0:0:0/block/sdb/sdb1 (block)
UDEV [1303783846.981849] change /devices/pci0000:00/0000:00:1c.2/0000:0d:00.0/host4/target4:0:0/4:0:0:0/block/sdb (block)
UDEV [1303783847.060260] change /devices/pci0000:00/0000:00:1c.2/0000:0d:00.0/host4/target4:0:0/4:0:0:0/block/sdb (block)
UDEV [1303783847.138262] add /devices/pci0000:00/0000:00:1c.2/0000:0d:00.0/host4/target4:0:0/4:0:0:0/block/sdb/sdb1 (block)

This is log of `udevadm test /devices/pci0000:00/0000:00:1c.2/0000:0d:00.0/host4/target4:0:0/4:0:0:0/block/sdb/sdb1`

From the log the udev rule we added is taking effect since "ID_DRIVE_FLASH=1" from the log.

Okay -- after reviewing the code, I'm not sure how we could get internal PCI flash drive readers to automount without changes to gvfs and adding a udev rule. I've gotten nautilus to automount with the attached patch and the following udev rule:

SUBSYSTEMS=="pci", ATTRS{vendor}=="0x10ec", ATTRS{device}=="0x5208", ENV{ID_DRIVE_FLASH}="1"

The following gvfs patch adds a check for a flash device when setting the gvfs 'should_automount' attribute.

Submitted patch upstream:

http://bugzilla-attachments.gnome.org/attachment.cgi?id=186903

Sorry -- proper bug link here:

https://bugzilla.gnome.org/show_bug.cgi?id=648980

Ayan, I committed your udev rule upstream:

  http://cgit.freedesktop.org/udisks/commit/?id=361a2ce3f2fc0b6ef02ac11d31bf77a0a9354e70

I pushed Ayan's gvfs fix upstream as well:

http://git.gnome.org/browse/gvfs/commit/?id=342ae51fe86663dce07d18a21c9827f8ee7fd131

I backported the patch to natty's gvfs, and uploaded to the natty-proposed queue, as per request from Ayan/OEM. Ayan, once Clint or another ubuntu-sru member reviews this, can you please test the proposed package, so that it can go into -updates?

Thanks!

Rejecting upload as per Ayan's request, he said he has a more generic solution for this.

This bug was fixed in the package udisks - 1.0.3-1

---------------
udisks (1.0.3-1) unstable; urgency=low

  * New upstream release.
    - Recognize Realtek PCI card reader. (LP: #759963)
    - Stop polling CD drives with udev >= 172 (soon to be released), which
      will enable in-kernel polling. With older udevs, this bug is "wontfix".
      (Closes: #576954)
  * Drop 00git-sd-card-id.patch, 00git-udf-permissions.patch,
    11-remove-unused-probers.patch: Applied upstream.
  * Drop 10-ide-cd-support.patch. ide-cd is dead; squeeze already uses libata.
  * debian/copyright: Add missing "Format:" header.
  * debian/control: Drop libpolkit-backend-1-dev build dependency, not
    required by this upstream version any more.
  * debian/udisks.install: Drop polkit extension, not built by this version
    any more.
  * debian/control: Recommend "eject". (Closes: #626592)
  * debian/control: Bump Standards-Version to 3.9.2 (no changes necessary).
 -- Martin Pitt <email address hidden> Wed, 29 Jun 2011 16:56:49 +0100

Testing the new gvfs, gdu, and udisks this morning to verify that the fix survived on its way downstream.

I backported the udisks patch and uploaded to natty-proposed. Needs ubuntu-sru review now.

Hello Ayan, or anyone else affected,

Accepted udisks into natty-proposed, the package will build now and be available in a few hours. Please test and give feedback here. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you in advance!

Attached is a debdiff for the gvfs package.

Sponsored gvfs into natty-proposed, thanks!

Unsubscribing sponsors, as there are no other patches to sponsor right now.

And finally, I need a rule added to udisks 1.0.2 for natty. This rule comes from the latest git head. I've attached a debdiff.

Ayan, the udisks part is already in natty-proposed, awaiting testing feedback.

Do we also need a gnome-disk-utility fix here? It's not even marked as fixed in oneiric.

Hi there, thank you for working on this bug. A fix for Ubuntu 10.04 (Lucid) would be appreciated, too.
I use Ubuntu 10.04 on a Dell Latitude 2110.
I have added the line suggested in post #13 to /lib/udev/rules.d/80-udisks.rules, but I don't see the SD card in nautilus nor in disk utility.

Is the gvfs component of this bug fixed in Oneiric? It seems like it should be, but there's been enough activity on this bug that it's not clear to me whether it has been fixed or not.

Once that's been cleared up I'm happy for gvfs to go into -proposed.

Yes, the patch is in oneiric's 1.9.3 (it was backported from upstream).

Ayan, do we actually need to change anything in g-d-u? I don't think so, but please confirm.

Martin,

No, we don't need g-d-u but we do need gvfs in Natty.

Hello Ayan, or anyone else affected,

Accepted gvfs into natty-proposed, the package will build now and be available in a few hours. Please test and give feedback here. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you in advance!

Hi Martin,

I tested this on two affected Dell Latitude 2110 systems with the Realtek card reader.

Prior to applying any updates, I'd insert an SD card which would be detected as sdb (visible on dmesg) but didn't automount. I'd have to manually go into Nautilus to mount the recognized filesystem.

After applying updates today (I did a dist-upgrade which updated gvfs among other things) I rebooted and tested, and now when inserting an SD card it gets detected and auto-mounted, an icon even appears in Unity and the device's Nautilus window opens automatically.

gvfs was updated to version 1.8.0-0ubuntu3.

Let me know if I can help by testing anything else.

Thanks for testing!

Verified fixed in -proposed.

This bug was fixed in the package udisks - 1.0.2-4ubuntu2

---------------
udisks (1.0.2-4ubuntu2) natty-proposed; urgency=low

  * Add 00git-rts_pstor-flash.patch: Mark all rts_pstor devices as card
    readers, to get proper automounting. Backported from 1.0.3.
    (LP: #759963)
 -- Martin Pitt <email address hidden> Tue, 09 Aug 2011 09:47:28 +0200

This bug was fixed in the package gvfs - 1.8.0-0ubuntu3

---------------
gvfs (1.8.0-0ubuntu3) natty-proposed; urgency=low

  * 10_flash_drive_autmount.patch: SRU of patch that allows gvfs to
    recognize ID_FLASH udisks attributes (lp: #759963)
 -- Ayan George <email address hidden> Tue, 16 Aug 2011 16:07:47 -0400

Marked as resolved in the upstream bug freedesktop-bugs #30381.