Is this bad design on stretch cluster?

John JY 221 Reputation points
2021-09-17T14:15:42.233+00:00

Hi all,

one of my clients has all stretch clusters with SQL availability group role set up with
two VMware nodes(one at production site and one at DR site(FSW at DR site)(windows 2012R2 up)
Node1(1NIC ) ----- Node2 (1NIC)+FSW (storage appliance)
Production site ---- DR site

Is this a bad design?

Thank you!

SQL Server
SQL Server
A family of Microsoft relational database management and analysis systems for e-commerce, line-of-business, and data warehousing solutions.
12,653 questions
Windows Server Clustering
Windows Server Clustering
Windows Server: A family of Microsoft server operating systems that support enterprise-level management, data storage, applications, and communications.Clustering: The grouping of multiple servers in a way that allows them to appear to be a single unit to client computers on a network. Clustering is a means of increasing network capacity, providing live backup in case one of the servers fails, and improving data security.
956 questions

6 answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. Limitless Technology 39,336 Reputation points
    2021-09-17T17:38:42.067+00:00

    Hello @John JY

    I don't see any reason why it wouldn't work, but personally I would not have the FSW as part of the storate of one of the nodes. Ideally nowadays would be easy to create a cloud drive to provide resilience to the equation.

    Please check:

    https://learn.microsoft.com/en-us/windows-server/failover-clustering/file-share-witness

    Best regards,

    0 comments
  2. John JY 221 Reputation points
    2021-09-17T19:38:03.713+00:00

    Hi all,

    is there an issue with cluster network which has only one route?

    Thank you for your insights.

    0 comments
  3. YufeiShao-msft 7,051 Reputation points
    2021-09-20T06:38:06.083+00:00

    Hi @John JY ，

    for multi-site stretched-cluster, if you do not have a third site, then cloud witness is the optimal solution, more simple solution is that put up a FSW on another server in one of the data centers and make sure it is running on separate circuit.
    A witness share needs to be available for single connection, the purpose of the FSW is to have something else that can count as a vote in situations where the number of configured nodes isn’t quite enough for determining quorum.

  4. John JY 221 Reputation points
    2021-09-20T13:31:51.483+00:00

    Hi @YufeiShao-msft

    Thank you for the insights. Due to two different sites which are likely to have transient network issues, what would happen if heartbeats get lost between these two nodes (one at PR and another at DR)?

  5. John JY 221 Reputation points
    2021-09-21T14:25:42.443+00:00

    @Tom Phillips no need for separate heartbeat network?