Our exchange system is a 2 server cluster with a server hub preceeding them.
For some reason, lately more frequently than ever, this hub server goes frozen and mail flow stops, needing to restart the server powering it off and on again.
I can't see any event logs previous to its hanging, no CPU spikes or memory problems.
This is becoming quite a critical problem now and we need to correct.
We have a vmware infrastructure and all our servers are virtual. I have reviewed the events in this platform and found nothing that could give me an indication of malfunction.
The only 'new' element in our network is the gradually implementation of iPhones and iPads which connect to our webmail address (hosted in this hub server), about a quarter of our users have moved from blackberry to these apple devices.

Apreciate your help in advance.

System: Exchange 2007 SP1, OS: Ms 2008 enterprise,

asked 12/13/2011 08:21

txarli33's gravatar image

txarli33 ♦♦

6 Answers:
Do you have any errors, when the hub stops working just general failures or anything?

What version of vmware are you on, also how do your resources look on the host server?

Just a first glance thought if you have more users your resources may be to low from the vm side, may need to add some more to this vm.

Found this link for best practices:


rm-ent's gravatar image


any backups or snapshots occuring, how is the memory utilization, and cpu utilisation on the host?

does it respond to pings when hangs?

access from console?

answered 2011-12-14 at 04:32:36

hanccocka's gravatar image


Hosts seem to be fine, not events at all on any of them.
We have a vmware cluster with 5 indentical servers and HA on them so if any problem guests servers get moved from host to host.
Also, we don't have any other server having these .

answered 2011-12-14 at 04:32:57

txarli33's gravatar image


As an example, last night, it froze at around 3:40am, at this tiem there was no backups or snapshots being run and NO, it doesn't respond to pings.
At this time of the morning there are no windows warning or error events logged either neither for system nor for application, only information events, actually there is a gap in event logs from 4:17am until 8:30am this morning when i got the server up and running, the last events are all information events indicating that services entered the stopped state, all by 4:17am.

System is vSphere 4.1.0 and resources are well enough to support a host failure, so no lack of them.

Thanks for your help so far.


answered 2011-12-14 at 04:40:25

txarli33's gravatar image


Check the Virtual Machine VMware.Logs in the virtual machine folder.

answered 2011-12-14 at 06:10:35

hanccocka's gravatar image


This reply lead us to pinpoint the root of the problem, a short databasecachesize was the main culprit.

answered 2011-12-14 at 06:12:59

txarli33's gravatar image


Your answer
[hide preview]

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here



Answers and Comments



Asked: 12/13/2011 08:21

Seen: 261 times

Last updated: 12/15/2011 06:39