Converted guest issue

[rjonesin]

Issue: Converted guest works fine in a test cluster. When same guest is moved to a similar cluster, an app service in a application hangs.

Guest: W2K, converted from VMWare GSX, 4GB single CPU.

Destination: 2 separate clusters in 2 different blade enclosures with DRS/HA. Enclosure 1 blades have E7340 CPUs, Enclosure 2 blades have E7440 CPUs. Both 2.4Ghz, one has 8MB L2 cache, the other has 16MB. DRS/HA only within each enclosure, not between.

Application: Java based app that processes jobs

Guest works fine in enclosure 1, appears to work fine in enclosure 2 but this particular service will consistently freeze when it processes a certain amount of jobs.

We have converted it the same into both enclosures, cloned it from the working one into the other and still have the same issue.

The only obvious difference between the enclosures is the CPUs. They both are ESX 3.5 update 4. I would not think that any app would care about the change in CPUs. We have migrated a large number of servers and are only seeing this issue with this particular server. Any ideas?

[aldikan]

Hi Rjonesin,

- Check the W2K event viewer for any relevant events, maybe they will give a clue

- On the troubled VM it might be worth to reinstall VMware tools..

- In the VM settings "Options" make sure that correct OS version is selected.

Please rate helpful answers,

Alex

[rjonesin]

Thanks for the fast response but we have tried all of those, nothing in the event logs, not even an error when the service hangs.Our app people have to kill it.

Tools have been re-installed, and the OS is set correctly.

[AndreTheGiant]

cloned it from the working one into the other and still have the same issue.

Just to be sure you have do a cold cloning?

Andre

**if you found this or any other answer useful please consider allocating points for helpful or correct answers

[rjonesin]

Yes,the last attempt was a cold clone of the working guest.

[aldikan]

Hi,

Check the properties of the process and let us know what are the "dependencies". How much time it runs before process freezing? Is it all over the place, or is there pattern?

This might help to narrow it down somehow.

Alex

[VMmatty]

Is everything the same with the networking configuraiton between the two clusters? Could it be that it is failing to communicate with another server (database maybe) when it is on the other cluster?