July 10, 2014 at 10:15 am
We are about to move our infrastructure to a new data center, as part of the migration we are hoping to move from 2008 R2 to 2012. We are currently running 2008 R2 w/mirroring, in the new environment we have a SAN backed 2012 failover cluster setup. Looking for any advice, guidance, references to white papers, gotchas to watch out for, etc. on the best process for migrating our databases to the new environment with minimal to no downtime (if that is possible). Any help, references or suggestions would be greatly appreciated!
July 10, 2014 at 1:40 pm
I don't have a lot of experience in this area, but some things to start looking at include checking for use of DEPRECATED SQL. I'm not sure what goes away between SQL 2008 and SQL 2012, but a Google search would be quite likely to find the Microsoft documentation. Also, there may well be similar kinds of changes to SSIS and SSRS, and perhaps even SSAS, so there's a fair amount of potential work just making sure that the existing code won't break once it starts running under SQL 2012. Also, sometimes, things just work differently, so testing in advance can often expose this kind of problem.
One possible scenario that might help highlight potential issues would be to build a test box with a 2012 instance, then backup the databases from the 2008 instance and restore those backups on the 2012 box. This might not be dissimilar to a typical DR scenario. Once restored, just change the compatibility level upward to 2012 and connect some apps and test things out. Folks don't always have this luxury, however.
Hope that helps you get started...
Steve (aka sgmunson) 🙂 🙂 🙂
Rent Servers for Income (picks and shovels strategy)
July 10, 2014 at 9:41 pm
Thanks sgmunson, I sincerely appreciate you taking the time to reply.
We will certainly be doing static testing on a 2012 instance before we migrate. The crux of my initial question is more around the best operational process for attempting to do a live rolling migration from a 2008 R2 instance w/Mirroring to a 2012 failover cluster. We have been reading through this white paper http://msdn.microsoft.com/library/hh923056.aspx but were looking for any additional real world advice/experience that anyone might be willing to provide on this process. Unfortunately we don't have a DBA on staff at the moment so we are hoping to fill the gaps by learning what we don't know due to lack of knowledge and expertise in this area. We have a team of competent seasoned software developers and architects but we don't have anyone with true database administration background on the team right now. We are trying to assess whether or not this is something we can figure out ourselves or if we are in over our heads and need to bring in someone with more DBA experience to assist with this migration.
July 10, 2014 at 10:40 pm
Whether or not you're in over your heads on the upgrade is a question I can't answer, but... that said, doing it without a DBA just doesn't sound like a very good idea. Push comes to shove there are doubtless a number of companies these days that for all practical purposes are the equivalent of "rent-a-DBA", and while I don't have any experience with that either, I do know they exist. In the long run, you may be better served by getting the DBA role filled first, but I'm just looking at the risks. There's a lot that can go wrong in this kind of scenario, and not having a DBA there with experience to handle those kinds of scenarios could be a disaster waiting to happen. Sorry that I can't be all that much assistance. I've not gotten into being a DBA, as I prefer the development side of things.
Steve (aka sgmunson) 🙂 🙂 🙂
Rent Servers for Income (picks and shovels strategy)
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply