We just installed Exchange 2007 on a: Duel Xeon 3.20 GHz hyperthread 6 GB ram SATA II drives Less than 35 users. No edge server Symantic Virus scanning 3 drives C: raid 0, contains log files and virtual file D: raid 1, - mailboxes E: Raid 1, not yet used The server console has always been sluggish and sometimes would not load and give an option to cancel. Over time it got worse and I brought up task manager and saw a rhythmic pulse of the CPU spiking and then going to zero. Rebooting one night cleared it and it ran normally for a day. The paging file was down to 3.2 GB for one admin who checked in the middle of the night but has always been around 5.9GB for me. Anyway, our Outlook users are getting the trying to connect to server message box. The Exchange server is easy to see on the rack it is the one with the almost solid green lights on the system drive raid. Ive read other posts from people having the same issue but the answers were a bit different. One thread had someone reducing the paging file size. Another mentioned a hotfix from Microsoft. Can anyone point me in the right direction ?

SATA is probably too slow a disk, especially in a RAID1 config. To ease the load on the disk/pagefile you'd need to add more RAM.

My guess would be Symantec (For Exchange). Check and see if it's running it's background scan, or if it has a complete store scan scheduled.Also, if you have Symantec (For Windows) client installed on the server itself, check that the proper file exclusions are in place. You could disable the client for troubleshooting.Andy

You were right about Symantec. The cause was that the Xeon hyperthreaded CPUs were seen as four processors when Symantec installed so it set the number of scanning thread accordingly. I went from 9 threads to 5 and the server started performing normally for a day but Symantec locked one service about 4:00AM local time. I came in to find the queues filling with email to be delivered. I booted the server after the service would not restart. I called Symantec and opened a case. I set the thread could for 4 and the swap file went from 6GBs to 3. It has been running well since early this afternoon but I expect it to lock again.