Clustering Question

  • Hope I can be pointed in the right direction and clustering isn't my strongest point.

    Had a scenario given to me whereby we have a Active Passive cluster at one site (production) with client apps pointing at the SQL instance (SQL08 R2 - Windows 2012 server), the question arose however asking what would happen if the entire site went down.

    We have a pre-production cluster at a DR site different subnet, exactly the same set up in SQL and Windows terms and basically it was how to failover to this other site if necessary. We discussed a very simple backup/restore mechanism, mirroring also and modifying the DNS on the pre-prod side so the clients did not require re-configuring - and that seems straight forward enough but to me not very elegant.

    The prod environment sits on very fast storage, but apparently we dont have disk replication available (I need to look into this further though) - is it possible to add a cluster IP as another passive node that would support automatic failover from the first cluster? Sure I read 2008 R2 does not support different subnet cluster nodes?

    As you can tell, not my strongest area at all so please excuse any daft questions or points!

    'Only he who wanders finds new paths'

  • I believe this can be done with disk replication, but no other way in 2008. In 2012 you have Availability Groups, which do not depend on some type of shared storage.

  • Thanks for the reply Steve and yes we did go through the 'if only we had 2012' process!

    'Only he who wanders finds new paths'

Viewing 3 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic. Login to reply