August 2, 2011 at 5:59 pm
Nightly, database from server1 is backed up, the dump file copied to server2, and then restored on server2.
The copy takes about 45 min, and the restore used to be about an hour. A week ago, for no reason that we can think of, the restore just decided to start taking 4 to 5 hours. We loaded on the pssdiag tool and ran it all night. Checked for blocking contention through the night, there is none. The perfmon output is so huge I'm not sure what to look for. Nothing changed on the machine, nothing installed. Disk array shows no bad disks, all error logs are clean. Can anybody shed some light on what to look for, before we end up calling MS support?
August 2, 2011 at 7:08 pm
Network?
A bit early to waste a call with PSS (unless you got them for free).
SOMETHING changed, you just need to figure out what.
Antivirus?
Slight change in san config (assume they lie) ;-).
New fileserver backup running at the same time?
August 31, 2012 at 8:13 am
I am having the exact same issue. There was no change in disk consumption on either server (prod and backup). Additionally, the log files show that the data and log pages created are relatively on par with what they have been.
August 31, 2012 at 12:56 pm
Assuming you're using a SAN here - are you working with a dedicated or shared array?
Ninja's_RGR'us (8/2/2011)
Slight change in san config (assume they lie) ;-).
"Nothing's changed. <hours later> We patched the controller but that wouldn't have affected the db."
_____________________________________________________________________
- Nate
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply