On 09/ 7/16 08:39 AM, Robert Mustacchi wrote:
On 9/5/16 20:30 , Ian Collins wrote:
I'm seeing lots of these messages on the console of a system with a Dell
PERC H730 Mini controller:
2016-09-06T03:15:07.044518+00:00 headnode mr_sas: [ID 887724
kern.warning] WARNING: mr_sas0: mrsas_tbolt_reset_ppc:resetadapter bit
is set already check retry count 101
2016-09-06T03:15:08.044506+00:00 headnode mr_sas: [ID 270009
kern.warning] WARNING: mr_sas0: io_timeout_checker: FW Fault, calling
reset adapter
2016-09-06T03:15:08.044529+00:00 headnode mr_sas: [ID 643100
kern.notice] mr_sas0: io_timeout_checker: fw_outstanding 0x32
max_fw_cmds 0x39F
The messages appear when one KVM is under load and that machine becomes
unresponsive. I can still access (but not ssh onto) the GZ and other
KVMs on the same system.
A pool scrub ran through without error.
Once the messages start to appear, zpool status hangs.
fmdump -e gives a few of these:
Sep 06 01:40:59.1282 ereport.io.scsi.cmd.disk.dev.uderr
Power cycling the box calms things down until the KVM gets busy again.
Any clues?
Hi Ian,
I have heard another report of something similar, but I'll need to dig a
bit. Are you using the disks in HBA mode or using logical volumes?
They should be in HBA mode (non-RAID disks in the the controller setup).
I should add that I had been testing the box for over a week before the
issues started.
--
Ian.
-------------------------------------------
smartos-discuss
Archives: https://www.listbox.com/member/archive/184463/=now
RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00
Modify Your Subscription:
https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb
Powered by Listbox: http://www.listbox.com
