Why 3 node cluster for Automatic Failover?:

Hi All

Some one has posted this topic for 10 months ago but there is no answer yet. I need explain to my client because they think 2 nodes is enough for auto failed over.

Pleas help. Thanks.

The answer is simple: if you do a auto-failover against 2 nodes, you don’t have a replica anymore. The cluster does not want to put you into a bad state without you manually performing that operation. Note that you can do manual failover on two nodes.

Other than that, note that we strongly recommend to run at least 3 nodes in production, for the exact same reason as above. If you have 3 nodes and one fails, you can still failover & rebalance and be back at one replica.

In addition to what @daschl said, the autofailover is based on all but one node being able to communicate. In a two node scenario, the cluster can’t guarantee that. It can with three. In effect, what you need for autofailover to take effect for a given cluster size of N is N-1>=2 in the failure case.

See the documentation for more information.

I defined 3 node cluster without auto fail.
When i shut down one server and tried to make query in the other server on the cluster, i saw that the query was now working, till i click on fail over on the server that was down.

As i see it, this is not suppose to be like that, in production when users do a lot of actions in the site and suddenly 1 server stop working, the data for the db is stopping working.

am i right?