VM Exchange services dropping every night... HELP!!

Having a very peculiar issue with my Exchange VM server. Every night the services begin to  error then W3SVC stops and Blackberry Dispatcher stops. My clients lose RPC over HTTP access and of course Blackberries stop receiving email. I don't know what is causing the spike in CPU and service stoppage overnight.

VM Server Details
OS: Windows Server 2003 SP2
CPU: 3.00 GHz
RAM: 3 GB


Details
Blackberry Controller began having issues (the service never stopped, but threw out warnings for the next few hours)
Event 20000 – Source: Blackberry Controller

Browser warnings and then immediately following was the SMTPSVC & W3SVC service errors
Event 8021 & 8032 – Source: Browser
Event 2012 – Source: SMTPSVC
Event 1120 & 1121 – Source: W3SVC

CPU running at 100% for a couple hours
Processes causing the bottleneck
Svchost (I couldn’t figure out which one it was, as there are about 7 of them) – forgot to add the PID column in task manager
Sqlservr PID 1612 [SQL Blackberry]

Memory usage up over 3 GB
Processes memory hogs:
Store climbed up to nearly 1 GB (after a reboot, store is at about 100 MB)
BAS-AS sits at 600 MB
Inetinfo also jumped up to a couple hundred MB (don’t remember how much)
IAmDHAsked:
Who is Participating?
 
AmitIT ArchitectCommented:
It seems, this server is used 24/7, first you check if DC is working fine, with the question details, i can see that server is overloaded, as you are running Exchange, BES and SQL also in same server, the best method is to have them in separate servers, atleast exchange should be installed alone.

Run Process Explorer
http://technet.microsoft.com/en-us/sysinternals/bb896653

This will help to determine which process is causing spike
0
 
Elmar KoschkaIT System EngineerCommented:
You have a Backup Job for this time ?!
0
 
IAmDHAuthor Commented:
ntbackup has a scheduled backup job that supposed to run at 6:00pm, but it hasn't been running right now. My problem doesn't even start until about midnight.
0
Ultimate Tool Kit for Technology Solution Provider

Broken down into practical pointers and step-by-step instructions, the IT Service Excellence Tool Kit delivers expert advice for technology solution providers. Get your free copy now.

 
thomasdavisCommented:
What is the first error in the event viewer at that time for system?
0
 
AmitIT ArchitectCommented:
It seems online defrag might be running during that time, it might be conflicting with backup time too. That could be the reason for spike, change the online defrag time, so it should not conflict with backup time. Also make sure you have AV exclusion set properly. You can put the /3gb switch also and increase ram to 4 GB if possible.
0
 
IAmDHAuthor Commented:
@ThomasDavis Event 8021 & 8032 – Source: Browser is the first, immediately followed by the others that I listed above. These are in the system event log

Also, I did have the Blackberry Controller warning that I listed, that preempted all of the above. This was in the application log.
0
 
IAmDHAuthor Commented:
@amitkulshrestha Thanks for that. I will give it a try. 3 GB switch is already in place.

Question: does the Online defrag have to be done after hours? The reason I ask, is that there isn't much of a window for this client as they are a hotel and hours are much longer than a regular business day. If the online defrag causes performance issues then my window for having it performed is still pretty small.
0
 
thomasdavisCommented:
The online defrag can be ran at anytime its best to do this during off peak hours, you can change the maintenance schedule for the database in properties of the storage group. Mine take a few hours to run.. You'll see this in event view under application "source" ESC
0
 
thomasdavisCommented:
I would suggest to upgrade the ram to at least 4 if possible.
0
 
IAmDHAuthor Commented:
I've created a new VM and have moved BES/SQL & Cloudmark to it.

I'm going to see what happens tonight with my Exchange server.

Will update on my findings...
0
 
IAmDHAuthor Commented:
After moving the BES and its SQL the server has settled down. Thanks for your help everyone...
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.