June 2, 2005 at 4:17 pm
We are in the process of determining whether to stay with a 2 node (active/passive) cluster or move to a multinode cluster (3 active/1passive) however we don't have a clear understanding of how the failover happens in a multinode cluster. Meaning when 1 active node fails to the only passive node what happens when another active node fails; does it fail to another online active node or to the existing passive node that just became active? Do we assign this failover mapping with our cluster resources by telling what order and where to fail to? thanks for you time!!
Thanks,
Todd
June 3, 2005 at 6:48 am
Failover will follow the Preferred Node order list for each Resource Group. Each SQL Instance should list the preferred node first, the 'empty' node second, and any remaining nodes last. If you have a node offline, you will need to make provisions to host any SQL instances among the remaining nodes. This will likely include altering memory settings and failover order.
Geoff N. Hiten
Microsoft SQL Server MVP
June 8, 2005 at 11:41 am
Thanks Geoff, I wanted to confirm my thoughts.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply