0

I have Production and DR clusters as depicted on a picture below.

My question is whether there is a way to configure Availability Group from datacenter1 to datacenter2?

  1. I tried to configure regular AG from Production to DR but it seems like it will not be possible because all 4 nodes needs to be under the same WSFC.
  2. Thought about creating distributed AG, but as I understand distributed AG goes on a "top" of regular AG.

So am I correct to assume that with this structure I will no be able to implement Availability Group?

enter image description here

Error if I try to create AG using wizard:

enter image description here

2
  • DR Server 1 and DR Server 2 are part of the same WSFC. But ProdServer1 and Prod Server 2 belong to different WSFC.
    – Serdia
    Dec 10, 2021 at 17:55
  • Also, there is FCI between DR Server 1 and DR Server 2. Same with Production servers, they have FCI
    – Serdia
    Dec 10, 2021 at 18:02

1 Answer 1

4

TL;DR;

You describe that the servers are already configured in separate WSFCs, so you will either need to destroy one of the clusters (presumably the DR cluster), and reconfigure into a single WSFC, or you will need to configure the DR FCI to be a single-instance AG, so that you can then configure the Distributed AG to run atop the Prod & DR AGs.

The long answer

You can configure your Availability Group(AG) through either method you describe. Whether it is right for you to configure a single AG that spans both data centers, or to use a Distributed AG is a more complicated question--but both are possible.

  1. A single Windows Server Failover Cluster (WSFC) and a single AG, which span all 4 servers in 2 data centers.
    • ProdServer1, ProdServer2, DRServer1, DRServer2 are all part of THE SAME WSFC
    • DRServer1 & DRServer2 are part of an FCI, DR\Instance
    • ProdServer1 & ProdServer2 have standalone instances
    • An AG is configured to span ProdServer1, ProdServer2, and DR\Instance
    • The AG will be a Multi-Subnet AG, and there will be configuration details to consider on client and/or server side. I've written about some of those considerations here & here
  1. A distributed AG that spans two data centers
    • ProdServer1, ProdServer2 are part of one WSFC
    • ProdServer1 & ProdServer2 have standalone instances
    • An AG (Prod_AG) is configured to span ProdServer1 & ProdServer2
    • DRServer1, DRServer2 are all part of a different WSFC
    • DRServer1 & DRServer2 are part of an FCI (DR\Instance),
      • or reconfigured to be standalone instances, similar to the Prod Servers.
    • An AG (DR_AG) is configured with DR\Instance as the only member
      • or the reconfigured standalone instances are configured in the AG
    • A Distributed AG is configured to span Prod_AG, DR_AG
3
  • thank you for your help.
    – Serdia
    Dec 13, 2021 at 18:32
  • Just one more thing: is Availability Group is something that everyone should be using? In terms of it is the best and most reliable way to protect your data from unwanted events?
    – Serdia
    Dec 13, 2021 at 18:34
  • Availability Groups are not the only technology for High Availability & Disaster Recovery. However, how to select which technology is the right solution for your needs is more complicated than I'm able to respond to in comments.
    – AMtwo
    Dec 13, 2021 at 22:00

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.