-
sql database restore to a subdirectory
I am backing up using out of band on a GB switch. Fast backups/restores but restore of a SQL activity log to a subdirectory is crawling. 13 GB restored out of 25GB after running for 16 hours.
Averaging 9 mb/min.
I've checked the broadcom card settings and BACS suite configs and they look fine.
What would cause this?
Thanks,
Tom
-
Moderator
Please check the Performance Monitor for the disk the file is being restored to. Also, do a Windows Defragmenter Analyze of the volume. You may also want to see how the sqlserver process is running.
I believe it is safe to rule out the network because the same route is fast for your other backups and restores.
-
Thanks, I started restore to another drive and is at good speed. Checked Event log and Performace monitor on the original drive and needs defrag which I will do after hours.
Thanks for your help!
Tom
-
Spoke to soon. After about 5 min. the restore degrades to .15 mb/min. Smaller databases restore fine.
Wireshark reveals out of order and drops. Could this be a compression related issue on large (25gb) databases?
We moved the dpv and server to an out of network switch that had no rules and had exact same results.
-
Moderator
Tom, I believe you should open a support case so we may track the steps required to resolve this issue and attach logs. The issue should not be related to compression. If you would, please provide this forum URL to the dispatcher when opening the support case. You will need to have the appliance asset tag on hand.
-
Thanks,
I fowarded this to our developer John Papadopoulos with info you requested.
Tom
-
Added additional space to C drive partition. Was 85% full. Stopped and started the SQL VSS writer service and Shadow Copy service.
Backup went to 800GB/min.
The size of the snapshot made by the Volume Shadow Copy service is determined by the size of the volume and how much data has changed since the last snapshot was made. This could be anywhere from a few hundred Megabytes, to tens of Gigabytes on a really busy server.
-
Increased disk space on C and restore time is good. What is the amount of swap space required by Unitrends for DB restores? Assume DB is 100GB in size. Do I need 100GB of free space for swap?
Posting Permissions
- You may not post new threads
- You may not post replies
- You may not post attachments
- You may not edit your posts
-
Forum Rules