Symmpi and Disk errors filling system log

[Daniyalis]

All -

In the past months I have issue with one particular Guest machine which is our Fileserver with RDMs for their File Share. ESX 3 and VC 2.0.2.

Only this vm give the problem in entire Datacenter.

The VM guests exhibit Event ID 11 and Event ID 15.

Event ID 11: The driver detected a controller error on \Device\Harddisk0. and then followed with the below error message

Event ID 15: The device, \Device\Scsi\symmpi1, is not ready for access yet.

Essentially what happened in both cases is the VM Guests were unresponsive so I had to kill the vmware processes running them. When they came back online I checked the event log and noticed that it was filled with these events.

Anyone know what I need to do to resolve this? I don't want any more VM Guests to hang. Please help. This is giving me lot of headaches.

[SimaW]

I am getting the same error on my guest print servers, did you get that problem resolved.

The device, \Device\Scsi\symmpi1, is not ready for access yet.

The driver detected a controller error on \Device\Harddisk0.

Can anyone else help with this, much appreciated.

S.

[zanmk]

Hello to everyone!

A week ago we did upgrade from 3.5U2 to U4. Since that, once a day, on a single Windows 2003 EE Server we receive about 200 error entries in system log.

All of them appear within 2 seconds and contain following data:

1: Source : symmpi, Event id: 15 "The device \Device\Scsi\symmpi1, is not ready for access yet"

2. Source: Disk, Event id : 11 "The driver detected controler error on \Device\Harddisk1"

System has SP2 applied and "LSI Logic PCI-X Ultra-320 SCSI Host Adapter" 5.2.3790.3959 (manuf. LSI Logic Inc.) configured for storage adapter.

ESX itself uses EMC Clariion storage array.

So far we can's see any influence on performance or system behavior but system administrators start complaining about it.

Has anyone faced this problem ? Are there recommended storage drivers for guest systems (Windows 2003) ?

best regards

Martin

[Rockapot]

I have seen this before at a previous client site. This issue can occur if you are doing MSCS clustering with the storage for the VM's incorrectly setup.

The issue is usually caused by a scsi reset occuring which locks the LUN causing other VM's not to be able to access their disks for a period of time. (ttp://communities.vmware.com/thread/75459)

**As a side note Windows Server 2008 no longer uses SCSI Bus Resets which can be disruptive on a SAN

This link gives a brief explanation of the issue. The fix mentioned here can help improve the situation by allowing the VM to wait longer for access to its storage however this is not really a true fix, more of a workaround

(http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1014)

Another posting on a similair issue can be found here. (http://communities.vmware.com/thread/9753)

Can you provide some additional information on how your LUN's are layed out. Are you using any shared storage in vmdk format for virtual clusters where the vmdk for that storage is on the same LUN as that of VM's affected by the Symmpi errors? or do all your VM's use RDM's for shared storage etc.., hope that makes sense

Carl

[amanvcp]

I saw these errors on one of my client site, they occured as a trigger of network outage as we have a netapp and our storage is iSCSI.

bottom line - Shared storage halt may cause this error to log in VMs.