Re: [PROBLEM] reproduceable storage errors on high IO load

From: Lars TÃuber
Date: Fri Jul 01 2011 - 09:13:19 EST


Hi.

And now I retried with 2.6.39.2. Same result. Complete dmesg available.

I tried to create a RAID with fewer drives and with more drives but still the
same problem.

Then I tried to use RAID5 without any luck.

Then I tried to format a single disk.
# mke2fs -jF /dev/sdc

And: The problem still occurs!!!!
I could send the complete dmesg if needed.

Then I tried with Ânosmp kernel option. And bad luck again.
I could send the complete dmesg if needed.

[ 760.138201] ata7.00: failed command: WRITE FPDMA QUEUED
[ 760.138305] ata7.00: cmd 61/00:00:10:02:70/02:00:28:00:00/40 tag 30 ncq
262144 out [ 760.138308] res 01/04:00:10:0c:7c/00:02:28:00:00/40
Emask 0x3 (HSM violation) [ 760.138563] ata7.00: status: { ERR }
[ 760.138652] ata7.00: error: { ABRT }
[ 760.138745] ata7: hard resetting link
[ 770.160095] sas: smp task timed out or aborted
[ 770.160105] drivers/scsi/mvsas/mv_sas.c 1718:mvs_abort_task:rc= 5
[ 770.160110] sas: SMP task aborted and not done
[ 772.170093] drivers/scsi/mvsas/mv_sas.c 1586:mvs_I_T_nexus_reset for device
[1]:rc= -70 [ 772.170108] sas: sas_ata_hard_reset: Unable to reset I T nexus?
[ 772.170112] sas: sas_ata_hard_reset: Found ATA device.
[ 772.170119] ata7: follow-up softreset required but no softreset available
[ 772.170252] ata7: hard resetting link

This is so simple it can't be I'm the only one with this problem!?

Regards
Lars
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/