Bug 53475

Created attachment 65528 [details]
udisks --dump

Arch Linux
x86_64
i686
udisks(2) v1.94.0
---------------------

I can confirm the bug with a Western Digital Raptor 10000 RPM disk (the 'udisks' version I am using is earlier than the one reported by the original author of this bug report). I have two other WD disks (7800 RPM) in the same machine and the error does not occur when probing those.

I can manually run (without the '--no-debug' argument):

> /usr/lib/udisks2/udisksd

and the output clearly shows the housekeeping failure for the Raptor while the others succeed.

It seems to me the bug is more likely in 'libatasmart' (which 'udisks' is dependent on) as the 'Input/output error' appears in the output of:

> skdump --overall /dev/sda

As far as a potential workaround (or likely, in my case, a permanent solution), set the following variables in a proper rule in a '/etc/udev/rules.d/' rules-file for any device you don't want probed for SMART data by 'udisksd':

ENV{ID_ATA_FEATURE_SET_SMART}="0"
ENV{ID_ATA_FEATURE_SET_SMART_ENABLED}="0"

For SMART-capable drives, these will normally be set to '1'. Use:

> udevadm info --query=all –path=/sys/block/sda

to see the variables for a particular drive (change 'sda' to whichever device you desire). Piping to 'grep' or 'less' is useful as the output from that command is quite long.

Before setting those variables, I initially tried setting:

ENV{UDISKS_DISABLE_POLLING}="1"

The setting had been a successful workaround for preventing 'udisks' from waking certain devices. I erroneously believed that the setting would prevent any attempts at communication between 'udisks' and the specified device (including SMART probing), but that did not work. I also want to mention that it is likely that setting only one or the other of the SMART variables will do the trick based on a very cursory review of the 'udisks' source code, but rather than waste more time on determining exactly what the code does with its if-then logic (which tests for both values), I just set them both FALSE to be sure.

By the way, the reason I say for me it's a 'permanent solution' (rather than a workaround) is that prior to discovering the error messages in my daemons.log, I didn't even know 'udisksd' was probing my disks every 10 minutes for SMART data while I already had cron-jobs doing essentially the same thing (but at much less frequent intervals) using 'smartctl' from the 'smartmontools' package.

Finally, setting the above Udev SMART-related variables does not prevent smartmontool utilities from successfully running as they are not aware of those variables (the utilities will get the smart-capabilities settings directly from the device, which remain unchanged by the actions described here).

Edit:

The command in Comment 2:

udevadm info --query=all –path=/sys/block/sda

should be:

udevadm info --query=all -–path=/sys/block/sda

For some reason the copy / paste operation replaced '--' (two ASCII code 45 characters) with '–' (one ASCII character code 8211). My keyboard can't even produce the 8211. Shrug.

Apologies.

I have this problem with my Raptor (WD1500ADFD).

I use smartmontools. ENV{ID_ATA_FEATURE_SET_SMART_ENABLED}="0" was enough to get rid of the error message. I did not try whether the other option is enough too. RaptorBug, thanks for the tip.

Same issue on openSUSE 12.2 and a 7200 RPM WD250Gb drive.

Oct  2 11:29:26 stal-dev-lx1 udisksd[17555]: Error performing housekeeping for drive /org/freedesktop/UDisks2/drives/WDC_WD2500JD_75HBC0_WD_WCAL74728599: Error updating SMART data: sk_disk_smart_status: Input/output error (udisks-error-quark, 0)

Just a me too report
but:
The error report on my system is for a usb connected external drive, which may help in the debugging. I have attached a udisks --dump file as details.txt

"udisksd[4617]: Error performing housekeeping for drive /org/freedesktop/UDisks2/drives/WDC_WD2500BB_14RDA0_WD_WCANKF372097: Error updating SMART data: sk_disk_smart_status: Input/output error (udisks-error-quark, 0)"

Does this relate to #61998?
I saw on a downstream report that the patch for it fixed this.

Yes, that patch seems to fix this.

*** This bug has been marked as a duplicate of bug 61998 ***

I use the /usr/lib64/libatasmart.so.4 from lib64atasmart4-0.19-7.mga5 and I still get the following error message every ten minutes :
udisksd[6232]: Error performing housekeeping for drive /org/freedesktop/UDisks2/drives/WDC_WD2500JB_57GVC0_WD_WCAL77874008: Error updating SMART data: sk_disk_smart_status: Input/output error (udisks-error-quark, 0)

I am asking myself if the patch https://bugs.freedesktop.org/show_bug.cgi?id=61998 has really been applied.

+1. Running udisksd version 2.6.5 with 4.13.0-15-generic kernel I can see periodically (10 min) repeating this error:

Oct 12 11:36:56 planb udisksd[770]: Error performing housekeeping for drive /org/freedesktop/UDisks2/drives/WDC_WD7500BPVX_00JC3T0_WD_WXV1E74NAFRJ: Error updating SMART data: sk_disk_smart_read_data: Operation not supported (udisks-error-quark, 0)

SMART seems to be working fine.
> smartctl -i /dev/sdc
smartctl 6.6 2016-05-31 r4324 [i686-linux-4.13.0-15-generic] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:     Western Digital Blue Mobile
Device Model:     WDC WD7500BPVX-00JC3T0
Serial Number:    WD-WXV1E74NAFRJ
LU WWN Device Id: 5 0014ee 6051f641c
Firmware Version: 01.01A01
User Capacity:    750,156,374,016 bytes [750 GB]
Sector Sizes:     512 bytes logical, 4096 bytes physical
Rotation Rate:    5400 rpm
Device is:        In smartctl database [for details use: -P show]
ATA Version is:   ACS-2 (minor revision not indicated)
SATA Version is:  SATA 3.0, 6.0 Gb/s (current: 1.5 Gb/s)
Local Time is:    Thu Oct 12 12:01:00 2017 CEST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

The filesystem is NTFS. Consistency check passed.
> blkid /dev/sdc1
/dev/sdc1: LABEL="tv" UUID="5E363D1C363CF71D" TYPE="ntfs" PARTUUID="e66755d8-01"
> ntfsfix /dev/sdc1
Mounting volume... OK
Processing of $MFT and $MFTMirr completed successfully.
Checking the alternate boot sector... OK
NTFS volume version is 3.1.
NTFS partition /dev/sdc1 was processed successfully.
--