October 18, 2014 at 12:14 am
Hi Expert,
I have following infrastructure at par. where i have 3 server in WFC(Windows Failover cluster) Win2012 and out of 3 servers two servers are active/passive cluster with SQL Instance hosted on it and i have created AG group for 3 users database and setup Sync Replica on 3rd server within my WFC cluster.
I have SQL VIP(Virtual Server Name) for cluster and also i have AG Listener resources also on the cluster. Now this cluster is attached to SAN single.
3rd server where we kept sync replica its storage is not coming from SAN it has local storage attached to it.
Now we know we have single point failure if SAN goes down My SQL Server 2node active passive cluster instance will also come down.
Now in this kind of event will my AG Listener can be online on WFC even if SAN is down so that it can redirect connection to the 3rd server where it has sync replica copy.
Also if in case Listener will not be able to be online in event of SAN failure what would alternative path we consider here
your valuable suggestion and time are most welcome.
October 19, 2014 at 1:38 pm
Nirav_Joshi (10/18/2014)
Hi Expert,I have following infrastructure at par. where i have 3 server in WFC(Windows Failover cluster) Win2012 and out of 3 servers two servers are active/passive cluster with SQL Instance hosted on it and i have created AG group for 3 users database and setup Sync Replica on 3rd server within my WFC cluster.
So, you have 3 node cluster, 2 nodes have a failover cluster instance and the 3rd node has a standalone instance installed, is this correct?
Nirav_Joshi (10/18/2014)
I have SQL VIP(Virtual Server Name) for cluster and also i have AG Listener resources also on the cluster.
To be clear, VIP is the acronym for virtual IP address, VNN is the acronym for Virtual Network Name 😉
Nirav_Joshi (10/18/2014)
3rd server where we kept sync replica its storage is not coming from SAN it has local storage attached to it.
That's perfectly valid
Nirav_Joshi (10/18/2014)
Now we know we have single point failure if SAN goes down My SQL Server 2node active passive cluster instance will also come down.
Correct, if you lose the shared storage totally, the clustered instance will be unavailable on either node.
Nirav_Joshi (10/18/2014)
Now in this kind of event will my AG Listener can be online on WFC even if SAN is down so that it can redirect connection to the 3rd server where it has sync replica copy.
Providing just the shared storage is lost and not the actual nodes themselves, quorum will be maintained at least and the Windows cluster will stay online. However, FCI replicas in an AlwaysOn group cannot be configured for automatic failover so some manual work will be required by the DBA.
Nirav_Joshi (10/18/2014)
Also if in case Listener will not be able to be online in event of SAN failure what would alternative path we consider hereyour valuable suggestion and time are most welcome.
The listener will come online when the group is manually brought online by the DBA.
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply