July 28, 2010 at 12:33 am
We have a cluster sql 2005 and it is on a storage unit called Shark and they want to move to EMC. The Guy said everything should be Transparent and we just need to shutdown sqlserver. Is that true? Is there any extra considerations I need to be aware of?
July 28, 2010 at 3:56 am
Yes. It should be smooth sail. Please ensure to have good backups for the instance. During the migration all you have to do it Stop and Start the SQL Services.
Pradeep Adiga
Blog: sqldbadiaries.com
Twitter: @pradeepadiga
July 29, 2010 at 2:02 am
How they really do that? Is it something like Norton Ghost they copy all the drives in Physical Level?
July 29, 2010 at 8:31 am
It depends on what they're actually doing. Are they swapping old disks for new disks? Or just the software used to control the disks?
If the later, they'll backup the data on the disks to make sure they don't lose anything and then install the software.
If the former, then there are all sorts of software options for copying over everything including "Ye Olde Copy-N-Paste Via Windows Explorer" method (which is a crude method of doing the job, but it works in a pinch for small stuff). [EDIT] C-N-P is not a recommended way of doing things, I'm just including it here for completeness.
Regardless, just make sure you have database backups of everything, including SystemDBs.
If you're worried, ask them for details. Tell them you're curious how everything works and ask them to explain it to them. I'm sure if you're polite about it, they'll be happy to tell you what they're doing.
July 30, 2010 at 12:06 pm
This is nothing new as far as technology goes even Windows Server has something similiar to replicate volumes to new volumes! I can't think of what it is called off the top of my head.
Like previous comments always have good and complete backups prior to doing this move!
July 30, 2010 at 1:03 pm
I don't know storage. I just want to share my bad experience happened two years ago. it may not relate to your case.
At that time, we had a cluster SQL 2000. somehow, the Disk array didn't show the correct disk information, like disk speed. so we called Technical support from Dell. they decided to change one raid controler and promised every thing would be fine. nothing needed to worry about.
But after they replaced new controler, Cluster did not work and SQL server could not start. we called Microsoft support. Microsoft had no way to help us. We had to build a new SQL box and restored every thing from backup. that took us a day (we have more than 100 database there). got complaint from clients.
The good thing is since that our company management team realized we really need a standby server.
If you only backup to tape, I suggest you also backup databases to other disk as well.
Viewing 6 posts - 1 through 5 (of 5 total)
You must be logged in to reply to this topic. Login to reply