[OmniOS-discuss] Ang: Re: Windows crashes my ZFS box

[Johan Kragsterman]

Hi!

-----"OmniOS-discuss" <omnios-discuss-bounces at lists.omniti.com> skrev: -----
Till: Rune Tipsmark <rt at steait.net>
Från: "Schweiss, Chip" 
Sänt av: "OmniOS-discuss" 
Datum: 2015-02-02 15:52
Kopia: "omnios-discuss at lists.omniti.com" <omnios-discuss at lists.omniti.com>
Ärende: Re: [OmniOS-discuss] Windows crashes my ZFS box

On Sun, Feb 1, 2015 at 6:21 PM, Rune Tipsmark <rt at steait.net> wrote:
I got some major problems... when using Windows and Fibre Channel I am able to kill my ZFS box totally for at least 15 minutes... it simply drops all connections to all hosts connected via FC. This happens under load, for example doing backups writing to the ZFS, running IO Meter against my ZFS...

... 

Latest FW on all items, HBA, Switch etc. Monitoring shows a distributed load on the ports as expected using Round Robin and MPIO.



This might be a shot in the dark, but the latest firmware on LSI HBAs is known to have serious problems.  It has more to do with data corrupting, so I'm not sure this is your cause.  Use P18 or P19, but not P20.

-Chip




Can't be that, cos' it works fine with other host operating systems.

No, when I think about it, the only thing I can imagine is the windows handling of sparse volumes. Are your LU's based on sparse volumes? I had a problem a couple of yrs ago with oVirt nodes, that is the oVirt virtualization management system hypervisor nodes. When I wanted to install these to the SAN, I got problems due to nested qcow volumes. I didn't really dig into it back then, but I know that the problems was the implementation of LVM/qcow they used, on top of ZFS/COMSTAR sparse volumes/LU's.

So I thought it MIGHT be something similar on window$, like the software volume handling or so. I am NO expert on win, but it might be worth trying to configure a non sparse volume for a LU to send to window$?

Johan






 

One thing that irritates me is that I don't get any more than ~80-120 MB/sec (sync=always) throughput when writing to this LUN in Windows, where I get 6-700 MB/sec (sync=always) when writing from a VM on ESXi... The abysmal performance is a pain, but the fact that I can downright crash or hang my ZFS box just by running IOMeter is disturbing...

 

Any ideas why this might happen? Seems to me like a queue problem but I can't really get any closer than that... maybe Windows is just crappy at handling Fibre Channel... however no problems against HP EVA Storage.... same machine, same tests....

 

br,

Rune

 

 


_______________________________________________
OmniOS-discuss mailing list
OmniOS-discuss at lists.omniti.com
http://lists.omniti.com/mailman/listinfo/omnios-discuss


_______________________________________________
OmniOS-discuss mailing list
OmniOS-discuss at lists.omniti.com
http://lists.omniti.com/mailman/listinfo/omnios-discuss