exchange server resetting

[westredd]

hi,

For the last week we have had 2 unexpected reboots on our 2003 exchange server.

Prior to the reboots there are a number of event 509 logs in the application log stating the following:

Information Store (5068) First Storage Group: A request to read from the file "E:\Exchsrvr\MDBDATA\stdpriv1.edb" at offset 62440579072 (0x0000000e89bfa000) for 4096 (0x00001000) bytes succeeded, but took an abnormally long time (62 seconds) to be serviced by the OS. In addition, 0 other I/O requests to this file have also taken an abnormally long time to be serviced since the last message regarding this problem was posted 315510 seconds ago. This problem is likely due to faulty hardware. Please contact your hardware vendor for further assistance diagnosing the problem.

we get about 10 of these events then the server reboots. The server is SAN attached. I think it looks like a hardware issue,, but would be intestested if anyone has had a similary problem in the past what caused it, -thanks.

 

[Zelandakh]

I'd say that is either a cable problem or a hardware problem. If it is a hardware problem I would suspect either a duff disk or a duff controller.

I have had this twice in Exchange, both times caused by corrupt RAID arrays.

 

[xmsre]

I'd say your backups are using a vss provider that relies on copy on write technolgy, and your storage was not sized properly to account for the IO Load. (A 62 second read is a really bad thing). I'd look at the system log for signs you're running out of system resources like PTEs, which isn't uncommon when you have a large storage bottleneck. Look for event ID 2019, 2020 and such as well as event id 50.

 

[Zelandakh]

If this happens during the backups then yes, I agree with xmsre.