August 12, 2010 at 7:20 am
Hi,
Just need to confirm other than load balancing what are the basic difference in Secondary server database restoring mode.
one thing i find that in stand by mode we get .TUF file.
but not in NoRecovery mode.
is there any performance issue also.
and what should we prefer generally.
Thanks
Rajat
August 12, 2010 at 7:36 am
STANDBY allows you to do read-only queries against the data whereas NORECOVERY is completely offline. STANDBY requires the .tuf file to manage its use. I have never seen any performance difference and I usually prefer STANDBY for log shipping as it allows me to verify the data and participate in limited DR tests.
August 18, 2010 at 10:38 am
I pretty much always use standby because of the aforementioned ability to read it in between log restores. norecovery is 100% unusable (until you need to restore it with recovery).
August 18, 2010 at 11:41 pm
Yes smith,
i am also interested in stand by mode.
but just need to figure out if there is something which norecovery mode have for log shipping but stand by mode not.
Thanks
Rajat
August 19, 2010 at 3:15 am
Hi,
I feel No recovery is mode is purely requirement based.
One of the scenarios I think of not using Standby mode is, if you are running other realtime applications on the secondary DB server and you do not want users to consume hardware resources by running queries against it:-)
- SAMJI
If you marry one they will fight with you, If you marry 2 they will fight for you 🙂
August 19, 2010 at 10:53 pm
Hi Rajat,
You shoukd go with stand by mode because it allows you to read the data from secondary server. TUF files are Transaction Undo Files. These files contain the uncomitted transactions. So that at next log backup restoration these uncomitted transactions are matched and accordingly commited/rolled back on secondary server.
August 23, 2010 at 2:58 am
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply