Exchange 2007 System - Services didn´t come up after migration

[joergriether]

Hey,

yesterday night during a maintenance session i migrated some e2007 servers (win2008 x64 single-cpu-vms running on esx3.5u2 servers, storage normal virtual disks on equallogic store) to another esx host (esx3.5u2, too). I DIDN´t used vmotion because that brought us some dirty 1018´s in the past. Instead i did shut down these servers using the virtual center "shutdown guest" option. After they cleanly were shutdown i migrated them to the other esx server (storage stayed at the same place). After start on the other esx machine, e2007 came up but some services remained down, the information store didn´t came up and there were some errors in the eventlog saying these services couldn´t come up because other dependant services were not up. This NEVER EVER happenend before during a normal shutdown/restart operation. If I tried to manually start these services, they started. On another E2007 were the service didn´t come up, too, i restarted the e2007 machine and voila, services came up for good.

Now my question would be: Why there were initial problems? The Host machines were BOTH DELL2950, equipped with EXACTLY the same cpu´s and memory. Also Bios-Versions ALL THE SAME.

Any idea?

best regards,

Joerg

[JoJoGabor]

Hmmm, seems strange.

Are the VT Extensions enabled on both hosts in the BIOS? This is to enable 64-bit guests, but would have thought the VM wouldnt boot at all.

Are all disks available when migrated to the new host? Ie If you have an RDM that isnt presented to that host properly that would stop Exchange starting. In Windows can you see the data and logs drives?

Is the network label the same on both host? ie can you ping the guest when its running on the second host?

[joergriether]

hi,

thanks for boarding but all the points you mentioned were self-evident double-checked.

best,

Joerg

[VMmatty]

Are you still having the issue? It doesn't really sound like a problem with your ESX config. It sounds more like something isn't right with the either the guest OS or the Exchange configuration.

[joergriether]

as i mentioned the issue was gone after one reboot. but this is not the point. the point is: this issue NEVER EVER happened before the migration. So my question is more focused on research than on problem solving.

[VMmatty]

Understood. It's one of those situations where it's fixed so it's hard to troubleshoot after the fact. When servers are down, especially email servers, your first instinct is to get them back up as quickly as possible rather than doing root cause analysis type work.

I would do some basic troubleshooting at the Exchange level. I really don't believe that it has anything to do with ESX.

1) Review the Event Logs and see if you can find anything specific

2) Run Exchange Best Practices Analyzer against the server and make sure everything is in compliance

Wish I could help more. I'm glad it's back up and running though..

[joergriether]

yup, that was on my agenda for today but i ran out of time ;( - will check it tomorrow.

best,

Joerg