vFRC Pool enabled --> VM gets Freeze

[YllowDnk]

Hello,

we have a 3-node vsphere 5.5 enterprise-plus licensed cluster and buyed three HCL certified 200 GB SSD's for each Host for vFRC.

On one of our Hosts we have a big VM with 12 vcpu and ~50 gb ram. During the last three months this VM got some freezes and the only possibility to get the win2k8r2 back running was to cold reboot the VM. After some troubleshooting with the vmware support we figured out that this freeze is happening when vFRC is enabled.

At the moment we don't know how to handle the issue.

This freeze happens when one of the hosts has the vFRC pool enabled - even if the VM doesnt reside on the host using the vFRC pool !

one of our test showed:

--> the affected VM was running on Host 2. Host 1 + 2 have no vFRC Pool enabled. When enabling vFRC pool on Host 3, the VM residing on Host 2 gets the freeze after some hours.

Hosts are 3 x DL 380 G7 and the HP SA P410i that handles the SSD has the latest drivers "scsi-hpsa 5.5.0.60-1OEM.550.0.0.1331820".

How is it possible that a VM that does not use vFRC + resides on a Host with no enabled vFRC pool is affected if i enable a vFRC pool on another host in the cluster. That doesn't really make sense for me & vmware support, too.

Anyone has an idea what could be the cause for this ?

Regards

Dnk

[smithar]

Could you please check the following :

1. Is the VM pingable at the time of freeze?

2. What is the CPU utilization of the VM in the performance chart at the time of freeze?

3. What is the SCSI controller that the VM uses?

[YllowDnk]

the symptoms of the freeze are that actions inside the vm and programs used by remotly connected users are gettin slower and slower until the users are dropped of the vm (but cpu utilization inside the task manager is under 30 %). After that we can see the log on screen of the vm during the vsphere client console, but logging in is not possible, just hangs at the welcome screen --> cold reboot.

1. yes the VM was always pingable but all ppl are losing the remoteconnection and reestablishing a remote connection is not possible anymore

2. i have to ask a colleague for that

3. LSI Logic SAS

the last freeze was on 04.09 between 3:00 PM and 7:00 PM. the orange color is the affected VM (never higher then 45 %)

[JPM300]

Silly question but is the drives you are using for the vFRC on VMware HCL

VMware Compatibility Guide: vfrc

[YllowDnk]

As mentioned in my first sentence - yes they are.

[JPM300]

could you post the log file of the VM that most recently froze.  See if we can find anything in there that can pinpoint the problem.

[YllowDnk]

Hello,

here are the logs of the day where we had the freeze.

What i could see was some entries similar to:

vcpu-x| I120: Guest: toolbox-dnd: Version: build-1598834

vcpu-x| I120: GuestRpc: Channel 4, conflict: guest application

vcpu-x| toolbox-dnd tried to register, but it is still registered on

vcpu-x| channel 0

There is a KB for this messages regarding VMWare Tools Update:

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=203635...

There are similar issues in this article for experiencing symptoms we had on this VM:

- You cannot map network drives.

- You are experiencing disconnected terminal sessions.

But VMWare Support told us that they are focussing on vFRC and that they are excluding this KB as root cause for the freeze.

Regards