August 2, 2019 at 10:58 pm
I have three cluster nodes configured for Always on Availability group, two nodes are in same location(node1 primary and node2 secondary with synchronous mode) and the third node(node 3 another secondary replica in DR with Asynchronous) would be in DR location.
Also, I have configured for automatic failover and trying to understand and ensure that why Quorum matters and cluster is determined by the number of voting elements that must be part of active cluster membership for that cluster to start properly or continue running. This means that the number
Of voting members in a WSFC determine whether or not the cluster stays online together with the applications running on top of it.
Every node in the cluster, by default, will have a vote. In order to keep the cluster up and running, the total number of votes have to reach a majority.
Majority = ( Total number of votes from voting members/2) + 1
The calculation for majority of votes is as follows. But, first, let’s start with the common 2-node cluster and how quorum behaves on different versions of Windows Server.
In a 2-node cluster, the majority of votes will be one (1.) And because both nodes in a cluster will have a vote, by default, we don’t really have a majority since 50% is sitting right in the middle. This is the reason why an additional vote in the form of a disk or file share witness is introduced. We need a “tie-breaker” in order to have majority of votes. Now that we have three (3) votes – two from the cluster nodes and one from the witness – the majority of votes will be two (2) – three divided by two is 1.5, plus one is 2.5 . We need to round down the results of the calculation of votes since we cannot really have half-a-vote. 66.67% is definitely higher than 50%, making it a majority.
Since DR is asynchronous if ever this becomes primary with main data center being down, how would you handle this? Please advise?
What is listener- It has dedicated IP address and What it does, It is somewhat comparable to witness, in that be looking for server1 and server2 and anytime if there is a failover between those two replicas it will automatically direct the clients to one that is primary database which is live. Is that right?
Thanks in advance!
August 3, 2019 at 11:10 pm
Thanks for posting your issue and hopefully someone will answer soon.
This is an automated bump to increase visibility of your question.
August 8, 2019 at 12:59 pm
The listener is just a name/address, has nothing to do with voting. Otherwise I think you're on the right track.
I'm not sure if this is best practice, but when we DR to another datacenter, we change who gets a vote. We have 2 nodes and a witness in both datacenters so this is simple to do. Seems to work. Especially in Win 2016/SQL 2016 (or newer?), where we really have not had any problems. We use the max cross-subnet clustering thresholds--I think this is important.
The reason we do this is that the connectivity between datacenters is not always 100%, and we don't want connectivity issues experienced only by the DR nodes to impact the active and perfectly fine primary node(s).
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply