The reason we deploy SQL Always-on as an enterprise is to have multi-site resiliency for our SQL clusters for operational and business continuity reasons. The issue is that ERA will not register SQL always-on clusters if they are spread between two sites/clusters. From an “API First” solution I would expect differently! I can not introduce risk and pull back DR planning because of the lack of support from ERA. How is the NTX community solving this issue?
Already have an account? Login
Login to the community
Login with your account
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.
Are you using a single AG for the two sites or is it two AGs within a distributed AG?
Also, in my experience, Era does not have a whole lot to offer if you’re not using it to actually deploy. To get always-on with Era, I have multi cluster setup in Era, as well as a multi cluster network profile. Then I deploy an always-on db, and at the screen where it’s asking about the nodes, change the cluster they reside on, with the DR node being async. Then deploy. At that point it will deploy a single AG (which is why i was asking above) spread across the clusters/sites. One thing i also noticed is that you have to have the IP space the same on both sides or it won’t work by default. It will deploy but when you try to failover, it will not work. To work around this, add a second IP to the listener that’s for the other site. It won’t work when using the primary, but when failing over, it will work.
I’ve not tried to register an existing always-on SQL deployment, but maybe you will need to have both Nutanix clusters registered with Era in order to register an existing SQL deployment that is spread across them.