Exchange computer becomes unresponsive....

Hi.  By, new location, do you mean 10 feet, or do you mean another building? city?

The reason is that a Global Catalog server might not be in the same site as the Exchange server anymore, or that the connection between the Exchange server and the GC is too slow.  Other reasons could be network changes or failures.  Is the patch cord new?  Is the speed of the switch different?  Is a different DNS server at the new location?  Is the server on another VLAN?  Does it have a hosts file?

I'd start with the easy things first . . . patch cord, event log, connectivity between the Exchange box and the GC, nslookup.   Then, I'd go for the hard things: http://www.microsoft.com/exchange/downloads/2003/default.mspx

go down about 2/3 and use the MTA check.

Hopefully, it's just something simple!

Could I have more details regarding the problem, like did you install any new s/w etc. and the other stuff. I can first relate this to the patch cords. or the n/w point itself. check the event logs and do revert back.

The patch cord is new, but the only other thing to change was the location (new building).

I'll be going in today to check if there is anything strange in the event viewer etc.

okay, the only things I can find in the event viewer for the same day that the connectivity problem happend were in the application log.

Event ID:1400     MSExchangeSRS

The Microsoft Exchange Site Replication Service could not initialize its Exchange database (EDB) and returned error 1.  The Site Replication Service will wait in a semi-running state so the database can be restored from backup and the SRS can mount it.

---------
Event ID:2219     MSExchangeMTA

The MTA is running recovery on the internal message database because the MTA was not shut down cleanly. This operation may take some time.  Status updates will be written to the Windows 2000 Event Log. [DB Server MAIN BASE 1 0] (14)

---------
Event ID:5008     MSExchangeSA

The message tracking log file C:\Program Files\Exchsrvr\JAWS.log\20050702.log was deleted.

---------

All of these were just warnings though. I can't find anything critical in the event logs that seems to say something about why the connectivity would fail.
There were also another couple of warnings that said that the recovery of the MTA went successfully, but I didn't bother putting those in there.

I just didn't think it would be the patch cord because this is a very once in a while thing. It might be something like once every two weeks, so I thought the failure of a patch cord would be a little more frequent.

Hi.  What about the Global Catalog location.  Is there one at the new site?  Is the new site caching?

there are two GC's here. They are both on our domain controllers....

I switched power supplies to test and it hasn't done it for quite some time.

I think this may have been a hardware issue.

This was actually the amount of power that was going to the machine from the wall. I had a 6 devices plugged into 4 power outlets via a 6-outlet power bar. When we had more power outlets installed and every device had it's own power outlet on a seperate line, everything worked great.

Glad to hear it's resolved.  Just ensure no one else is on the same circuit, or when winter arrives and the space heaters are fired up, your problems will return.

This problem disappeared for a long time but came back again. However, after running basic tests it turns out that one of the sticks of RAM was bad. Just thought it might help someone if they had the same problem and ended up chasing the power.

That blasted RAM!