5

I'm currently doing my first setup with Distributed Availability groups. This is currently a POC setup but will normally be implemented soon. I have 4 nodes, 2 nodes located together and 2 other also together but with some distance between them.

I set up the Distributed AG with the automatic seeding and everything is fine so far. The database got synchronised to the secondary Availability group without any issues.

Now comes the actual question: I have created two listener for my AG's (Primary & Secondary), but I can't create a listener for my Distributed AG (Not much information on it). This poses the problem that my application will have to be Listener aware...

Is there a way to create a listener for the Distributed AG, through which the application will always be able to connect even if we failover the distributed availability group?

Improve this question

3 Answers 3

Reset to default
5

After a call with someone from Microsoft, I was told that there are no listeners on the Distributed availability level yet. So currently if we want to use DAG, you will have to change your DNS or application connection to connect to the secondary availability group.

Improve this answer
1

You can create a CNAME that points to the primary AG that the applications use to connect. Set the time to live low. Change the CNAME to point to the DR AG on failover.

Improve this answer
0
-1

Apparently it can be done in SQL 2016/2017. See Distributed Availability Groups. Still no automatic failover but there is one listener for both AG groups.

Improve this answer
1
  • "Although each underlying availability group of the distributed availability group has a listener, a distributed availability group has no listener"
    – Pixelated
    Commented Aug 23, 2023 at 17:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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