October 31, 2013 at 1:53 pm
I have a SQL Server 2012 Cluster running on our network which spans between two physically separate buildings via fiber. The Cluster is running in building A and is using our SAN as the storage medium for the cluster, so all nodes and the SAN are in building A.
Building B is located a few hundred yards away and as I mentioned before is connected to building A with fiber. It just so happens that the people who use the cluster the most are in building B as well. Once in a while we will have a network outage which will "sever" the connection between the two buildings and cause our production facility to essentially go down. This only happens once in a while, but clearly is not acceptable.
My question is this - is there a way to have a failover of our cluster or some other method of caching data in building B's network on a node member or other type of device which would then sync up with the cluster when it regained connectivity? I know that I can put another node of the cluster in building B, but if the link is ever "severed" again, the node still wouldn't be able to communicate with the SAN so it wouldn't even do any good.
Does anybody have any ideas for me?
Thank you!
October 31, 2013 at 3:25 pm
Two way transactional replication might be the best option. If your SAN breaks connection it basically means that Clustering in this scenario is useless.
November 3, 2013 at 11:24 am
the first item to tackle is why the network connection is lost. Also, the implementation of an extra network may help to maintain connectivity between sites.
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply