ISCSI IO issue

Hello,

Sorry for my poor english, I try to explain what happen with my DSS system as best as I can :

I got some IO issue on a DSS V6 lite since this WE and I don't know why.
This server has been installed since July 2011, and the only trouble I had was to change the supermicro motherboard after a huge crash in September 2011.
This WE, all the VMs of my customer have frozen, the 3 ESXi servers didn't respond anymore but DSS was up and seem to be fonctionnal, no error has been reported in Admin GUI.
I restarted manualy all ESXi cluster, and VMs one by one, there were some corrupt datas on 2 VM, then I restored some backups and everything was fine.
Except I got lower performance than before this crash.
I open a ticket support to VMWare and provide the logs, They reply it's a storage issue, please, contact your provider.

To setup my DSS I follow the howto by Open-E.
I have :
3 ESXi servers with 6 Gigabits NICs on each system, 4 NICs used like ISCSI initiators for vstorage and vmotion in differents sub LANs.
2 aggregated Cisco SMB switchs
and 1 DSS Raid6 1.6To with 10 VMs inside, 4 Intel Gigabits NICs in ISCSI mode.

Thank you for your help.

Best Regards.

download the logs from the DSS V6 Lite at STATUS-->Hardware, and see if there are issues.
This will help:

http://www.open-e.com/service-and-su...ts-and-videos/

2010.10.26

http://www.open-e.com/site_media/the...011/ico_EN.png

How to view and understand the log files from DSS V6

Hello Gr-R,

Thanks for your reply.
I'll read the log I just download.
VMWare verified and validate my conf. The support ingeneer gives me some KB for the latency issue of the DSS storage :

http://kb.vmware.com/selfservice/mic...rnalId=1007688
Look for the DAVG/cmd field (device latency). This gives you an idea of how long the ESX host is waiting (in milliseconds) for SCSI commands submitted to the storage to come back with a response.

http://kb.vmware.com/selfservice/mic...rnalId=1008205
This is the average response time in milliseconds per command being sent to the device

http://kb.vmware.com/selfservice/mic...rnalId=1010213
The device latency average may vary depending of the device and the type of connection. The values are in milliseconds. Check with your array vendor for the appropriate value. If you see a spike or consistently high values (for example, values around 100), investigate with your storage vendor. you should see as well abort from the VMkernel log

I didn't restart the DSS yet, What do you think about this ?

Thanks again for any help.

Well, I try to fin some issues in the log files, I read some of them : critical, error, syslog, etc... without success.
I don't know what I'm looking for.

For this server, I use DSS V6.0up75.8401.5377 64bit Lite release.
2 Intel Corporation 82571EB Gigabit Ethernet Controller (rev 06) and 2 Intel Corporation 82574L Gigabit Network Connection as ISCSI Target in different sub LANs.
Jumbo frame is activate in all chain.
1 3Ware controler with HDD SAS 15k in Raid6 mode for one array of 1.6 To, 500 Go in used for 10 VM.

Let me know if I need further information.

I reboot the DSS system and everything runs fine now.
I also change the iscsi path mode to "recently used'
I don't understand what happened.

Thanks for your explanation.

Glad to hear that, well we can't tell for sure what happened without have a look to the system logs first, but it could be an issue related to the initiator connection, so it's better to check the initiator settings, also try to update to the latest build as that will give you the latest drivers for your hardware which will fix and update your old drivers.