3

I have two SQL Server 2016 installations configured in a mirror with a separate server as the witness. We've been running some failover tests and every test has succeeded except the following two scenarios. Does anyone know why these scenarios wouldn't result in a failover?

  • Dismounting the storage that the database files reside on the primary server will not fail the server over. I thought the witness would notice that the files do not exist anymore and fail over?
  • Throttling the network down to 1kbps on the primary server disconnects the mirror on both the primary and secondary. I would think that the witness would lose connectivity to the primary because the network is so slow and fail over but instead both servers go to disconnected status.

Has anyone ran into any of these issues?

4

•Dismounting the storage that the database files reside on the primary server will not fail the server over. I thought the witness would notice that the files do not exist anymore and fail over?

Database mirroring does not look at the individual database, even though that's the unity of failover. It's the same as an availability group in 2012/2014 or 2016 without the database failover option.

This is expected.

•Throttling the network down to 1kbps on the primary server disconnects the mirror on both the primary and secondary. I would think that the witness would lose connectivity to the primary because the network is so slow and fail over but instead both servers go to disconnected status.

If all are disconnected because the network is exceeding sluggish or gone, then it won't be able to talk between the servers for a failover - the databases need to be synchronized, the primary is online and the session timeout for the mirror elapsed so it looks like a mirror disconnect rather than a primary issue. Even if something were to happen, after the disconnect the databases would not be in a state for automatically fail over.

Reference of the required conditions for automatic failover for mirroring.

Also, with Mirroring completely gone at this point, I'd suggest migrating away from using it... though you probably already know this :)

  • 1
    so what HA option do I have available if the storage goes offline? – user133514 Aug 28 '17 at 19:22
  • 1
    @user133514 You're using 2016, so use availability groups with the database_failover option. That'll get you most of the way - the actual failing of the instance/server will put you on the other instance(s)... so hopefully they aren't on the same storage. – Sean says Remove Sara Chipps Aug 29 '17 at 12:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.