November 5, 2012 at 4:27 am
Gurus, I have a question regarding the 2012 Availability Groups. please consider this scenario and answer at your earliest convenience.
We have APP1, APP2. When I create Windows Server Failover Cluster(WSFC) with 3 nodes
node 1 - Primary Replica - APP1 is connected with cluster name
node 2 - Secondary Replica (Synchronous)
node 3 - Secondary Replica (Asynch)
Now, the question is.. Since my cluster name is already used for APP1 to connect to node1, if I want my App2 to connect to secondary replica(synch) how should I do that.?
Also, In the event of failover, node 1 will failover the availability group to node 2. This will allow my App1 to connect to node 2 via cluster name. what would be the status of my app2? what would be the status of my other availability groups in node 2. I dont want to put all the load on one single node at any given point of time. how can I automate to failover my other Availability groups to node 3? Again how will my app2 connects to the new node?
Hope I am clear with my question!!
Thanks
Jagan Kondapalli
Thanks
Jagan K
November 5, 2012 at 5:36 am
jvkondapalli (11/5/2012)
Gurus, I have a question regarding the 2012 Availability Groups. please consider this scenario and answer at your earliest convenience.We have APP1, APP2. When I create Windows Server Failover Cluster(WSFC) with 3 nodes
node 1 - Primary Replica - APP1 is connected with cluster name
node 2 - Secondary Replica (Synchronous)
node 3 - Secondary Replica (Asynch)
Now, the question is.. Since my cluster name is already used for APP1 to connect to node1, if I want my App2 to connect to secondary replica(synch) how should I do that.?
Also, In the event of failover, node 1 will failover the availability group to node 2. This will allow my App1 to connect to node 2 via cluster name. what would be the status of my app2? what would be the status of my other availability groups in node 2. I dont want to put all the load on one single node at any given point of time. how can I automate to failover my other Availability groups to node 3? Again how will my app2 connects to the new node?
Hope I am clear with my question!!
Thanks
Jagan Kondapalli
With Availability Groups, each of the applications can be a separate Availability Group in the WSFC and would have separate Availability Group Listeners that your applications would connect to, not the WSFC cluster network name. At application failover, the Listener moves between nodes and redirects the application to the new node. If you want to prevent multiple nodes from owning the same instance you have to write your own code to check for that and execute failover commands for the AG's to move them where you want. The better way to do this in a 3 node setup would be to have Node 1 and Node 2 as the HA sync replicas for App1, and then Node 3 and Node 2 as the HA sync replicas for App2.
Without your full layout and implementation details, it is impossible to give an exact answer, and there are a lot of considerations that we cover when consulting on these types of setups that I won't go into over the forums because of the time involved. The above should give you enough ideas to be able to implement what you asked about.
Jonathan Kehayias | Principal Consultant | MCM: SQL Server 2008
My Blog | Twitter | MVP Profile
Training | Consulting | Become a SQLskills Insider
Troubleshooting SQL Server: A Guide for Accidental DBAs[/url]
November 5, 2012 at 1:40 pm
jvkondapalli (11/5/2012)
We have APP1, APP2. When I create Windows Server Failover Cluster(WSFC) with 3 nodesnode 1 - Primary Replica - APP1 is connected with cluster name
The WSFC name is immaterial to the availability group, separate this in your mind immediately. For each availability group you create you will see a cluster application, this is where the clustered listener resource sits if you create it.
jvkondapalli (11/5/2012)
Now, the question is.. Since my cluster name is already used for APP1 to connect to node1, if I want my App2 to connect to secondary replica(synch) how should I do that.?
Create a listener for each AO group
jvkondapalli (11/5/2012)
Also, In the event of failover, node 1 will failover the availability group to node 2. This will allow my App1 to connect to node 2 via cluster name.
No, listener name, if you create it.
jvkondapalli (11/5/2012)
what would be the status of my app2?
did you create a new AO group for the database(s) for App2?
How did you configure the group(s) for backup priority?
jvkondapalli (11/5/2012)
I dont want to put all the load on one single node at any given point of time.
That's pretty much what you will be doing unless you employ read only routing to spread the read load across replicas in the AO group
See these links for an overview on AlwaysOn and also how to combine them with Failover Cluster Instances of SQL Server.
AlwaysOn availability groups in SQL Server 2012[/url]
Combining AlwaysOn Groups With Failover Cluster Instances[/url]
-----------------------------------------------------------------------------------------------------------
"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