Project

General

Profile

Bug #2189

oi_151a: I/O errors with mr_sas driver

Added by Denny Fliegner almost 8 years ago. Updated almost 7 years ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
Start date:
2012-02-27
Due date:
% Done:

0%

Estimated time:
Difficulty:
Medium
Tags:
needs-triage

Description

On a DELL R710 with one H700 and two H800 RAID controllers we see
zpool errors when reading several TB during a zfs send/recv backup
procedure.

c4t4d0 ONLINE 0 0 0
c4t5d0 ONLINE 0 0 0
c4t6d0 FAULTED 0 0 0 too many errors
c4t7d0 FAULTED 0 0 0 too many errors
c4t8d0 REMOVED 0 0 0

fmdump reports

50%  defect.sunos.eft.unexpected_telemetry
Problem in: dev:////pci@0,0
Affects: -
FRU: -
Location: -
50%  fault.sunos.eft.unexpected_telemetry
Problem in: dev:////pci@0,0
Affects: -
FRU: -
Location: -

and a list of:

100%  fault.fs.zfs.io_failure_wait
Problem in: zfs://pool=spool
Affects: zfs://pool=spool
FRU: -
Location: -
...

The corresponding devices are gone:

scsi: [ID 107833 kern.warning] WARNING: /pci@0,0/pci8086,3410@9/pci1028,1f15@0/sd@0,0 (sd0):
drive offline
...

There is a quad-GbE card and a usually also a dual port 10GBE-card
in the machine. Removing the 10GbE-card does not help. The system
is still operational with the remaining components. After reboot the
zpool is available again. In case of a permanent error on a disk replacing /
resilvering does not work, it just runs into the same error again.

Any ideas?

History

#1

Updated by Ken Mays almost 7 years ago

  • Status changed from New to Closed

Closing ticket (not OI-dev fixable directly). The issues are being address by the Illumos devs for improved mpt_sas and mr_sas driver support. Check with them for further resolution.

Also available in: Atom PDF