kernel-packages team mailing list archive

[Armin Schindler <1516269@xxxxxxxxxxxxxxxxxx>]

After checking with kernel 4.2.0-16-generic again, I got the error too.
I have another server (same hardware) still running with 4.2.0-16-generic (not updated yet) and the error doesn't show up, so I thought the update of the kernel brought the problem. The servers hardware was completely exchanged, but problem stays.

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1516269

Title:
  ata NCQ error with md, drbd and high load

Status in linux package in Ubuntu:
  Confirmed

Bug description:
  With Ubuntu wily kernel 4.2.0-18-generic a SyncTarget to a DRBD on LVM
  on MD Raid1 produces errors:

  Nov 14 17:46:20 kobol kernel: [  198.351892] ata1.00: exception Emask 0x60 SAct 0x38 SErr 0x800 action 0x6 frozen
  Nov 14 17:46:20 kobol kernel: [  198.351922] ata1.00: irq_stat 0x20000000, host bus error
  Nov 14 17:46:20 kobol kernel: [  198.351935] ata1: SError: { HostInt }
  Nov 14 17:46:20 kobol kernel: [  198.351944] ata1.00: failed command: WRITE FPDMA QUEUED
  Nov 14 17:46:20 kobol kernel: [  198.351957] ata1.00: cmd 61/48:18:00:d8:95/05:00:07:00:00/40 tag 3 ncq 692224 out
  Nov 14 17:46:20 kobol kernel: [  198.351957]          res 40/00:20:40:e1:95/00:00:07:00:00/40 Emask 0x60 (host bus error)
  Nov 14 17:46:20 kobol kernel: [  198.351988] ata1.00: status: { DRDY }
  Nov 14 17:46:20 kobol kernel: [  198.351997] ata1.00: failed command: WRITE FPDMA QUEUED
  Nov 14 17:46:20 kobol kernel: [  198.352009] ata1.00: cmd 61/40:20:40:e1:95/00:00:07:00:00/40 tag 4 ncq 32768 out
  Nov 14 17:46:20 kobol kernel: [  198.352009]          res 40/00:20:40:e1:95/00:00:07:00:00/40 Emask 0x60 (host bus error)
  Nov 14 17:46:20 kobol kernel: [  198.352039] ata1.00: status: { DRDY }
  Nov 14 17:46:20 kobol kernel: [  198.352047] ata1.00: failed command: WRITE FPDMA QUEUED
  Nov 14 17:46:20 kobol kernel: [  198.352059] ata1.00: cmd 61/80:28:80:e1:95/00:00:07:00:00/40 tag 5 ncq 65536 out
  Nov 14 17:46:20 kobol kernel: [  198.352059]          res 40/00:20:40:e1:95/00:00:07:00:00/40 Emask 0x60 (host bus error)
  Nov 14 17:46:20 kobol kernel: [  198.352090] ata1.00: status: { DRDY }
  Nov 14 17:46:20 kobol kernel: [  198.352099] ata1: hard resetting link
  Nov 14 17:46:20 kobol kernel: [  198.352106] ata2.00: exception Emask 0x60 SAct 0x70000 SErr 0x800 action 0x6 frozen
  Nov 14 17:46:20 kobol kernel: [  198.352133] ata2.00: irq_stat 0x20000000, host bus error
  Nov 14 17:46:20 kobol kernel: [  198.352145] ata2: SError: { HostInt }
  Nov 14 17:46:20 kobol kernel: [  198.352154] ata2.00: failed command: WRITE FPDMA QUEUED
  Nov 14 17:46:20 kobol kernel: [  198.352166] ata2.00: cmd 61/48:80:00:d8:95/05:00:07:00:00/40 tag 16 ncq 692224 out
  Nov 14 17:46:20 kobol kernel: [  198.352166]          res 40/00:88:40:e1:95/00:00:07:00:00/40 Emask 0x60 (host bus error)
  Nov 14 17:46:20 kobol kernel: [  198.352196] ata2.00: status: { DRDY }
  Nov 14 17:46:20 kobol kernel: [  198.352204] ata2.00: failed command: WRITE FPDMA QUEUED
  Nov 14 17:46:20 kobol kernel: [  198.352217] ata2.00: cmd 61/40:88:40:e1:95/00:00:07:00:00/40 tag 17 ncq 32768 out
  Nov 14 17:46:20 kobol kernel: [  198.352217]          res 40/00:88:40:e1:95/00:00:07:00:00/40 Emask 0x60 (host bus error)
  Nov 14 17:46:20 kobol kernel: [  198.352246] ata2.00: status: { DRDY }
  Nov 14 17:46:20 kobol kernel: [  198.352254] ata2.00: failed command: WRITE FPDMA QUEUED
  Nov 14 17:46:20 kobol kernel: [  198.352266] ata2.00: cmd 61/80:90:80:e1:95/00:00:07:00:00/40 tag 18 ncq 65536 out
  Nov 14 17:46:20 kobol kernel: [  198.352266]          res 40/00:88:40:e1:95/00:00:07:00:00/40 Emask 0x60 (host bus error)
  Nov 14 17:46:20 kobol kernel: [  198.352295] ata2.00: status: { DRDY }
  Nov 14 17:46:20 kobol kernel: [  198.352304] ata2: hard resetting link
  Nov 14 17:46:21 kobol kernel: [  198.671678] ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
  Nov 14 17:46:21 kobol kernel: [  198.673618] ata2.00: configured for UDMA/133
  Nov 14 17:46:21 kobol kernel: [  198.673628] ata2: EH complete
  Nov 14 17:46:21 kobol kernel: [  198.727651] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
  Nov 14 17:46:21 kobol kernel: [  198.735336] ata1.00: configured for UDMA/133
  Nov 14 17:46:21 kobol kernel: [  198.735343] ata1: EH complete

  The error didn't show up before the update from kernel 4.2.0-16-generic.
  Disabling NCQ with 
   echo 1 > /sys/block/sda/device/queue_depth
  is a workaround.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1516269/+subscriptions