August 30, 2013 at 8:04 am
I'm looking into using SQL 2012 AlwaysOn for DR purposes in our shop and don't find much information on the failback process. Can anyone comment on the failback process or point me to a good link? Is failback exactly the same as failover? I assume there must be some process to run to resync the servers before failing back. This is primarily what I am interested in knowing more about. For example, node 1 fails and it fails over to node 2. it takes a week to repair node 1 (or recover from a true disaster). Now we have a weeks worth of data to mirror back to node 1 before failing back. Is this built in to the failback process? How is this managed and how do you check the state of the mirroring?
I would also be interested to hear any general comment, good or bad, tips, tricks, gotchas that anyone has with AlwaysOn.
Many thanks!
August 30, 2013 at 8:24 am
Is failback exactly the same as failover?
i can comment on this statement..
Failover and Failback are two different kind of "Procedures"
They are same...
While assiging High Availability groups, You choose one as primary and the other as secondary or disaster recovery.
If the primary server fails, the Application or service failover to the seconday or the available server in order to continue its function with out disturbing the end user, They still work with same connection string, same listner, everything is same except the apllication failover.
So now the secondary server on which it was failedover becomes the primary for the application and from the server it came from becomes the secondary.....
Now failback is nothing but, if the appliation failover again to primary (either automatic or manual) says to be failback (which is same as failover).
Now the servers also become as same when it was prior to the initial failover, I.e, primary becomes primary again and secondary as secondary....
Conclusion: they are not separate kind of procedures. They are same but named different according to the servers they are located on.
August 30, 2013 at 8:34 am
I assume there must be some process to run to resync the servers before failing back.
Always On High Availability does mean they are always synchronized on all the servers in the cluster, If the cluster is 2 node (1-primary, 1-secondary)
They are always synchronized, even after failover and failback,
It is up to you to decide where to maintain the Application or services or sometimes also called as groups....
The concept of primaru and secondary is depend up on these applications, based up on their presence o server, primary and secondary are decided...
For example if you have two application (A,B) and having two node cluster(X,Y)
if A is on X and B is on Y
then for A X= primary and Y= Secondary
then for B Y= Primary and X= Secondary...
i think i gave a small clean idea,
sorry if i confused you more
August 30, 2013 at 9:03 am
OLDCHAPPY (8/30/2013)
I'm looking into using SQL 2012 AlwaysOn for DR purposes in our shop and don't find much information on the failback process. Can anyone comment on the failback process or point me to a good link? Is failback exactly the same as failover? I assume there must be some process to run to resync the servers before failing back. This is primarily what I am interested in knowing more about. For example, node 1 fails and it fails over to node 2. it takes a week to repair node 1 (or recover from a true disaster). Now we have a weeks worth of data to mirror back to node 1 before failing back. Is this built in to the failback process? How is this managed and how do you check the state of the mirroring?I would also be interested to hear any general comment, good or bad, tips, tricks, gotchas that anyone has with AlwaysOn.
Many thanks!
You're getting AO confused with FCIs. Although AO uses the WSFC functionality the cluster failback functionality is not exactly relevant.
When the Primary database in an AO group fails a replica partner will take over as the primary database. Once the original primary comes online it will become a secndary replica.
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply