Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations SkipVought on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Finding a simple way to have a 2008 R2 server to take some of the load of of the SBS 2011 Standard

Status
Not open for further replies.

akersrus

Technical User
May 12, 2011
3
0
0
US
We are running SBS 2011 on a server that was most likely underbuilt. It has 16 Gigs of RAM and two Xeon(R) CPU E5-2665 0 @ 2.40GHz Processors. The Memory is always peaked and the exchange is a big part of that.
Rather than taking memory away from exchange and make it availible for other processes, What about using this other server? If that is a option, would it be best to run the domain controller on the second server?
I am not much on server OSes so any help would be appreciated.
 
Exchange and SQL will gobble as much memory as you can throw at it.

The system is fairly clever and will quickly reallocate resources from the exchange memory usage and use it when its needed.

What appears to be the issue?

ACSS - SME
General Geek

 
The server Does not seeem to be releaseing the memory as designed, Memory usage will build until 15.7 or 15.8 gigs of the total 16 being used. At that point the server becomes unresponsive to any request made of the machine. At the console or over the network. We have 8 users, the server is used as a file server and holds the peachtree Database plus exchange. Seems like a small load for the server but that is what seems to happen. about every two to three days it needs rebooted. when it is responsive, I get complaints that Peachtree or Outlook takes a long time to load completely.
 
What do the underlying drives look like? It could be that the IO subsystem is just being overwhelmed. Since it's normal to have all memory in use, that's not a sure sign of a memory issue. But if all memory is in use, it will generate more writes to disk, and disk is most often the real bottleneck.

If you use Performance Monitory to track Disk Read Queue Length and Disk Write Queue Length, you'll be able to see if IO transactions are piling up and causing the lockup.

If you are doing quite a bit with it beyond the DC and Exchange roles, you may just want to add more memory instead of adding a whole new server to manage.

Dave Shackelford
ThirdTier.net
TrainSignal.com
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top