Hitachi hard drive HTS541616J9SA00 requires NCQ blacklisting

what is the outcome of: hdparm -I /dev/sda

Hi! This is the log of hdparm as requested.

Thanks

Enrico

another dmesg log...

https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/78288
: HSM violation....

HSM violation...

hsm violation. no booting problems though.

Hi leoquant!

Thanks for the help...is there a solution?

Thanks again

Enrico

Hi!

I reported the bug upstream... it seems a problem of the Hitachi hard disk with the ncq so the device will be blacklisted in the next kernel release ...

This is the complete discussion:

http://groups.google.it/group/fa.linux.kernel/browse_thread/thread/e8914f11039c64d1/0ba853557f0cc704?lnk=st&q=enrico+sardi+hitachi&rnum=1&hl=it#0ba853557f0cc704

Bye

Enrico

This is fixed in gutsy.

no santa rosa here but running Gutsy with latest 2.6.22-10 kernel. i have a SAMSUNG HM160JI .

i got the following messages in dmesg along with very strange behaviour after a failed hibernation.

[ 94.664000] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 94.664000] ata1.00: cmd c8/00:10:1f:bd:82/00:00:00:00:00/e1 tag 0 cdb 0x0 data 8192 in
[ 94.664000] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 99.704000] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 104.688000] ata1: device not ready (errno=-16), forcing hardreset
[ 104.688000] ata1: soft resetting port
[ 134.904000] ata1.00: qc timeout (cmd 0xec)
[ 134.904000] ata1.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[ 134.904000] ata1.00: revalidation failed (errno=-5)
[ 134.904000] ata1: failed to recover some devices, retrying in 5 secs
[ 144.948000] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 149.932000] ata1: device not ready (errno=-16), forcing hardreset
[ 149.932000] ata1: soft resetting port
[ 153.956000] ata1.00: revalidation failed (errno=-2)
[ 153.956000] ata1: failed to recover some devices, retrying in 5 secs
[ 158.960000] ata1: soft resetting port
[ 159.348000] ata1.00: configured for UDMA/100
[ 159.348000] ata1: EH complete
[ 160.056000] sd 0:0:0:0: [sda] 312581808 512-byte hardware sectors (160042 MB)
[ 160.056000] sd 0:0:0:0: [sda] Write Protect is off
[ 160.056000] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
[ 160.056000] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA

P.S. is bug #131633 a duplicate?

i've been getting a similar error for about a year now on my laptop. The symptom is that the system will freeze for about 45 seconds, and will respond to nothing during that time. Afterwards it snaps out of it and returns to normal operations. When it started it was happening about once per hour, but with the newer gutsy releases it's been happening every 5-10 minutes, which makes using interactive programs extremely annoying. dmesg says:

[17063.276179] ata1.01: qc timeout (cmd 0xa0)
[17063.276201] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[17063.276216] ata1.01: cmd a0/00:00:00:00:20/00:00:00:00:00/b0 tag 0 cdb 0x43 data 12 in
[17063.276221] res 51/20:03:00:00:00/00:00:00:00:00/b0 Emask 0x5 (timeout)
[17068.312038] ata1: port is slow to respond, please be patient (Status 0xd1)
[17073.292096] ata1: device not ready (errno=-16), forcing hardreset
[17073.292104] ata1: soft resetting port
[17073.644198] ata1.00: configured for UDMA/100
[17073.832048] ata1.01: configured for UDMA/25
[17073.832064] ata1: EH complete

It's an Acer Aspire 9410 laptop. hdparm says:

ATA device, with non-removable media
        Model Number: WDC WD800UE-22HCT0
        Serial Number: WD-WXE406119580
        Firmware Revision: 09.07D09

i'm running the current (as of today) gutsy, and have been updating regularly for over a year (mainly in the hope that this problem would go away).

I've upgraded to Gutsy, and am still getting this problem as well (assuming it's the same issue), so maybe we should remove the "Fix Released" tag?

And could we change the title to something more general? I'm not sure what Santa Rosa is so I probably don't have one. ;)

I've attached the output from logcheck & hdparm.

I have the same problem

I've attached the hdparm output.

tail -f /var/log/messages shows:

Nov 10 19:41:17 stan kernel: [ 764.188000] ata1.01: qc timeout (cmd 0xa0)
Nov 10 19:41:17 stan kernel: [ 764.188000] res 51/20:03:00:00:00/00:00:00:00:00/b0 Emask 0x5 (timeout)
Nov 10 19:41:22 stan kernel: [ 769.232000] ata1: port is slow to respond, please be patient (Status 0xd1)
Nov 10 19:41:27 stan kernel: [ 774.228000] ata1: device not ready (errno=-16), forcing hardreset
Nov 10 19:41:27 stan kernel: [ 774.228000] ata1: soft resetting port
Nov 10 19:41:28 stan kernel: [ 774.600000] ata1.00: configured for UDMA/100
Nov 10 19:41:28 stan kernel: [ 774.788000] ata1.01: configured for UDMA/25
Nov 10 19:41:28 stan kernel: [ 774.788000] ata1: EH complete
Nov 10 19:41:28 stan kernel: [ 774.804000] sd 0:0:0:0: [sda] 195371568 512-byte hardware sectors (100030 MB)
Nov 10 19:41:28 stan kernel: [ 774.812000] sd 0:0:0:0: [sda] Write Protect is off
Nov 10 19:41:29 stan kernel: [ 775.260000] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Nov 10 19:41:29 stan kernel: [ 775.296000] sd 0:0:0:0: [sda] 195371568 512-byte hardware sectors (100030 MB)
Nov 10 19:41:29 stan kernel: [ 775.312000] sd 0:0:0:0: [sda] Write Protect is off
Nov 10 19:41:29 stan kernel: [ 775.344000] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA

I'm changing some attributes on this so that it gets the proper attention. Someone please correct me if I did this wrong.

i found a workaround for my particular system (see above somewhere in the comments) which has worked for me for a month or two now:

/etc/rc.local:

####################################
echo "Killing the hald polling processes to try to stop CD drive from hanging..."
if /usr/bin/pkill -f 'hald-addon-storage: polling'; then
    echo "Killed."
else
    echo "Warning: hald polling process not found or not killable."
fi

echo "Disabling APM of /dev/sda ... "
/sbin/hdparm -B 255 /dev/sda
####################################

The overall problem seems to be caused by the first part - hald polling the CD drive every couple of seconds. The second part (APM) was done because some *buntu users have posted that the default settings are tough on the hard drive in terms of wear and tear.

The down side to the above solution is that DVDs/CDs won't automount. But i'm old-fashioned anyway and prefer to manually mount all external devices, so that doesn't bother me in the least bit.

i hope this helps someone else out there...

Hi!
See this page:

https://wiki.ubuntu.com/LaptopTestingTeam/AcerTravelmate6292?highlight=%286292%29

there is a fix for the problem

Enrico

It didn't worked for me...

Hi,

if it can help : this bug doesn't appear with Fedora 8 and OpenSUSE 10.3. But it appears with Mandriva 2008.

Is it because of different versions of kernel ? or something else ?

with ubuntu Gutsy alternate CD, the boot sequence start with a message like this : ata_piix : invalide map (something i could not read, too quick)

Hope it will be resolved soon.

Hi Sebastian!

Did you disable the ncq? What model of HD do you have?

Bye

Enrico

Hi Enrico,

here's the hdparm output: http://launchpadlibrarian.net/10474542/hdparm

I'm not sure if it helps, but the pertinent part of lshw is:

          *-disk
                description: SCSI Disk
                product: HTS421210H9AT00
                vendor: ATA
                physical id: 0.0.0
                bus info: scsi@0:0.0.0
                logical name: /dev/sda
                version: HABO
                serial: HKA640ARGJ54LF
                size: 93GB
                capabilities: partitioned partitioned:dos
                configuration: ansiversion=5

Sebastian

I'm noticed these errors occur briefly in my logs about the same time as I started getting errors like 'torrent paused: disk write error, seek failed: 'Invalid argument' fd: 49 offset: 2287992832 seekdir: 0' from deluge. I used smartctl to run a SMART self-test on the disk and it is fine, no errors. There was a burst of them, all in a few seconds. I have a Samsung Spinpoint drive.

Dec 2 20:44:27 navybean kernel: [114268.069489] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
Dec 2 20:44:30 navybean kernel: [114268.069501] ata1.00: cmd b0/d2:f1:00:4f:c2/00:00:00:00:00/00 tag 0 cdb 0x0 data 123392 in
Dec 2 20:44:30 navybean kernel: [114268.069503] res 50/00:f1:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM violation)
Dec 2 20:44:30 navybean kernel: [114268.378360] ata1: soft resetting port
Dec 2 20:44:30 navybean kernel: [114268.533962] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
Dec 2 20:44:30 navybean kernel: [114268.550170] ata1.00: configured for UDMA/133
Dec 2 20:44:30 navybean kernel: [114268.550189] ata1: EH complete
Dec 2 20:44:30 navybean kernel: [114268.557593] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
Dec 2 20:44:30 navybean kernel: [114268.557603] ata1.00: cmd b0/d2:f1:00:4f:c2/00:00:00:00:00/00 tag 0 cdb 0x0 data 123392 in
Dec 2 20:44:30 navybean kernel: [114268.557605] res 50/00:f1:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM violation)
Dec 2 20:44:30 navybean kernel: [114268.865094] ata1: soft resetting port
Dec 2 20:44:30 navybean kernel: [114269.020695] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
Dec 2 20:44:30 navybean kernel: [114269.036887] ata1.00: configured for UDMA/133
Dec 2 20:44:30 navybean kernel: [114269.036902] ata1: EH complete
Dec 2 20:44:30 navybean kernel: [114269.040739] sd 0:0:0:0: [sda] 488397168 512-byte hardware sectors (250059 MB)
Dec 2 20:44:30 navybean kernel: [114269.045691] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
Dec 2 20:44:30 navybean kernel: [114269.045702] ata1.00: cmd b0/d2:f1:00:4f:c2/00:00:00:00:00/00 tag 0 cdb 0x0 data 123392 in
Dec 2 20:44:30 navybean kernel: [114269.045704] res 50/00:f1:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM violation)
Dec 2 20:44:30 navybean kernel: [114269.355825] ata1: soft resetting port
Dec 2 20:44:30 navybean kernel: [114269.511415] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
Dec 2 20:44:30 navybean kernel: [114269.527618] ata1.00: configured for UDMA/133
Dec 2 20:44:30 navybean kernel: [114269.527633] ata1: EH complete
Dec 2 20:44:30 navybean kernel: [114269.532399] sd 0:0:0:0: [sda] Write Protect is off
Dec 2 20:44:30 navybean kernel: [114269.532409] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
Dec 2 20:44:30 navybean kernel: [114269.542036] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
Dec 2 20:44:30 navybean kernel: [114269.542046] ata1.00: cmd b0/d2:f1:00:4f:c2/00:00:00:00:00/00 tag 0 cdb 0x0 data 123392 in
Dec 2 20:44:30 navybean kernel: [114269.542048] res 50/00:f1:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM violation)
Dec 2 20:44:30 navybean kernel: [114269.850524] ata1: soft resett...

I'm getting similar error messages but without the 30 second lockups. This is on an old Dell GX110 desktop (PIII/700, Intel chipset, using ata_piix module, with Samsung SP2514N PATA disks). Running Feisty, fully updated, with kernel 2.6.20-16-generic, using libata for the two PATA disks. Controller is UDMA/66 while disks are UDMA/100 so ATA uses 66 from bootup.

ACPI is enabled if that matters, and kernel messages indicate it's working.

The errors started on 23 Dec when I updated the kernel (aptitude log):

[UPGRADE] linux-headers-2.6.20-16 2.6.20-16.32 -> 2.6.20-16.33
[UPGRADE] linux-headers-2.6.20-16-generic 2.6.20-16.32 -> 2.6.20-16.33
[UPGRADE] linux-image-2.6.20-16-generic 2.6.20-16.32 -> 2.6.20-16.33

The errors from /var/log/messages are intermittent but always the same, e.g.:

Dec 28 12:25:48 valluga -- MARK --
Dec 28 12:45:49 valluga -- MARK --
Dec 28 12:54:07 valluga kernel: [10228.550910] res 50/00:f1:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM v
iolation)
Dec 28 12:54:07 valluga kernel: [10228.550948] ata1: soft resetting port
Dec 28 12:54:07 valluga kernel: [10228.711921] ata1.01: ata_hpa_resize 1: sectors = 488397168, hpa_sectors = 48839
7168
Dec 28 12:54:07 valluga kernel: [10228.719895] ata1.00: ata_hpa_resize 1: sectors = 488397168, hpa_sectors = 48839
7168
Dec 28 12:54:07 valluga kernel: [10228.739862] ata1.00: ata_hpa_resize 1: sectors = 488397168, hpa_sectors = 48839
7168
Dec 28 12:54:07 valluga kernel: [10228.739880] ata1.00: configured for UDMA/66
Dec 28 12:54:07 valluga kernel: [10228.767822] ata1.01: ata_hpa_resize 1: sectors = 488397168, hpa_sectors = 48839
7168
Dec 28 12:54:07 valluga kernel: [10228.767840] ata1.01: configured for UDMA/66
Dec 28 12:54:07 valluga kernel: [10228.767878] ata1: EH complete
Dec 28 12:54:07 valluga kernel: [10228.774706] res 50/00:f1:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM v
iolation)
Dec 28 12:54:07 valluga kernel: [10228.774745] ata1: soft resetting port
Dec 28 12:54:07 valluga kernel: [10228.935803] ata1.01: ata_hpa_resize 1: sectors = 488397168, hpa_sectors = 48839
7168
Dec 28 12:54:07 valluga kernel: [10228.943739] ata1.00: ata_hpa_resize 1: sectors = 488397168, hpa_sectors = 48839
7168
Dec 28 12:54:07 valluga kernel: [10228.963743] ata1.00: ata_hpa_resize 1: sectors = 488397168, hpa_sectors = 48839
7168
Dec 28 12:54:07 valluga kernel: [10228.963760] ata1.00: configured for UDMA/66
Dec 28 12:54:07 valluga kernel: [10228.983696] ata1.01: ata_hpa_resize 1: sectors = 488397168, hpa_sectors = 48839
7168
Dec 28 12:54:07 valluga kernel: [10228.983714] ata1.01: configured for UDMA/66
Dec 28 12:54:07 valluga kernel: [10228.983753] ata1: EH complete

Can anyone comment as to whether these errors are significant?

Enrico's original bug report was about his specific hard drive model requiring a black list entry in drivers/libata/libata-core.c for NCQ (serial ata native command queueing). The need for this is identifiable in the log files by " spurious completions during NCQ". His hard drive model, Hitachi HTS541616J9SA00, was added to the blacklist for the Gusty kernel so this bug really was and is Fix Released. Subsequently, I am changing the status of this bug report back to Fix Released.

It would be helpful if everyone else were to submit new bug reports and test their hardware with a Hardy Heron Live CD. You can find out more about the development release of Ubuntu at http://www.ubuntu.com/testing/ . Additionally, useful things to include in your bug report would be the full output of 'dmesg', the results of 'sudo lspci -vvnn' and 'uname -a'. Thanks in advance for your cooperation and understanding.

I can confirm that in my case the problem was caused by this process:

hald-addon-storage: polling /dev/scd0 (every 2 sec)

Once I kill it i have no problems.

I have additional questions:

a) How can one see what drives are in the list?

b) How can one force ncq to happen for testing purposes even if a drive is in the list?

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.

It seems I have soled this problem on my own computer.It is a Acer Travelmate 3274.I've being geting along with this problem for almost 1 year.I found it's all because there is conflict between my cdrom and the disk.when I executed "ll /dev/ ",I saw
[code]
lrwxrwxrwx 1 root root 3 2009-10-21 13:28 cdrw -> sr0
lrwxrwxrwx 1 root root 3 2009-10-21 13:28 dvd -> sr0
lrwxrwxrwx 1 root root 3 2009-10-21 13:28 dvdrw -> sr0
lrwxrwxrwx 1 root root 3 2009-10-21 13:28 cdrom -> sr0
[/code]
so I just [code]rm /dev/sr0[/code]
And my computer work well for almost 6 hours until now.
Sorry for my poor english.Hope some one can understand what I've said.

It seems I have soled this problem on my own computer.It is a Acer Travelmate 3274.I've being geting along with this problem for almost 1 year.I found it's all because there is conflict between my cdrom and the disk.when I executed "ll /dev/ ",I saw
[code]
lrwxrwxrwx 1 root root 3 2009-10-21 13:28 cdrw -> sr0
lrwxrwxrwx 1 root root 3 2009-10-21 13:28 dvd -> sr0
lrwxrwxrwx 1 root root 3 2009-10-21 13:28 dvdrw -> sr0
lrwxrwxrwx 1 root root 3 2009-10-21 13:28 cdrom -> sr0
[/code]
so I just [code]rm /dev/sr0[/code]
And my computer work well for almost 6 hours until now.
Sorry for my poor english.Hope some one can understand what I've said.
this is my e-mail:<email address hidden>

I have reported a similar bug on
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/587859

I've just upgraded from ubuntu 8.10 to ubuntu 10.04 (8.10 -> 9.04 -> 9.10 -> 10.04)
and the bug appeared on all of them and in ubuntu upstream kernel.

Last logs:

Jun 3 19:00:07 lml kernel: [105220.860043] ata1.00: qc timeout (cmd 0xa0)
Jun 3 19:00:07 lml kernel: [105220.860054] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
Jun 3 19:00:07 lml kernel: [105220.860060] sr 0:0:0:0: CDB: Test Unit Ready: 00 00 00 00 00 00
Jun 3 19:00:07 lml kernel: [105220.860074] ata1.00: cmd a0/00:00:00:00:00/00:00:00:00:00/a0 tag 0
Jun 3 19:00:07 lml kernel: [105220.860076] res 51/20:03:00:00:00/00:00:00:00:00/a0 Emask 0x5 (timeout)
Jun 3 19:00:07 lml kernel: [105220.860080] ata1.00: status: { DRDY ERR }
Jun 3 19:00:12 lml kernel: [105225.900027] ata1: link is slow to respond, please be patient (ready=0)
Jun 3 19:00:17 lml kernel: [105230.884026] ata1: device not ready (errno=-16), forcing hardreset
Jun 3 19:00:17 lml kernel: [105230.884037] ata1: soft resetting link
Jun 3 19:00:17 lml kernel: [105231.048481] ata1: nv_mode_filter: 0x739f&0x739f->0x739f, BIOS=0x7000 (0xc0000000) ACPI=0x701f (60:900:0x11)
Jun 3 19:00:17 lml kernel: [105231.064367] ata1.00: configured for UDMA/33

You can find more logs in
http://lml.ls.fi.upm.es/pipermail/lml-adm/