Data Recovery 1% starting then failing

[amurray7]

We're having issues while running our backups where a couple VM's will run very slowly. I'm just looking for any suggestions to try as I've tried as much as I know. This is only happening to 2 out of 35 VM's. One is running Windows 2000 (I know, bad!) the other Windows 2003. I'm working on upgrading the 2000 server to 2008 R2 to see if that takes care of the problem. But we are backing up other 2000 and 2003 servers and they are not affected at all.

Earlier I tried to turn the Data Recovery VM and it timed out after sitting at 1% for 20 mins (while also taking down those other 2 VM's).

Checked all drives, processors, and memory... nothing is getting pushed to its limits.

Details:

30 VM's ((15) 2008 R2, (5) 2008, (8) 2003, (2) 2000)

1 DRS Cluster with 2 Physical ESX servers, esx1 and esx2

12 drive fiber chan SAN (4 LUNS w/RAID 5)

1 External backup appliance 4 TB (2GB trunk)

[Shakaal]

Would request you to check the event logs inside the VM's would also request you to provide the complete error message

check one more thing are you able to take snapshots for these 2 VM.

Regards

[amurray7]

I am able to create snapshots very quickly on both of those servers. When trying to start the VMWare Data Recovery "VM" it is immediately trying to relocate from one host (ESX2) to another (ESX3). Maybe that is where it's hanging up? Should I try manually moving it to ESX3 before starting it?

I've attached the Task and Event's logs.

Appreciate the help!

[amurray7]

So I tried to just migrate the VMware DataRecovery machine from ESX3 to ESX2 and it stayed at 1% for almost a half our and then finally gave up. During that time, those other two VMs were having a tough time. I was also having trouble with one of my linux boxes, just running Apache, with the CPU jumping all around. I move that to another physical host and it seems to be fine now.

Still trying to figure out what would cause the DataRecovery Machine to not migrate, even manually.