Post reply

Distributed AG

  • June 7, 2022 at 3:18 pm

    #4045684

    Hi,

    I setup distributed AG between two clusters and did failover testing. During failover I found an issue. Please help me to resolve the issue or provide me links showing if it is limitation.

    Problem:

    Configuration:

    WSFC 1: Node1 and Node2

    WSFC2: Node3 and Node4

    Availability Groups: AG1(WSFC1). AG2(WSFC2)

    AG1 Primary node: Node1

    AG2 Primary node(Forwarder): Node3

    Distributed AG: DISTRIAG

    Problem:

    1. When Primary node1 is changed to secondary, node2 becomes Primary on WSFC1 cluster sync is not happening to node3 and node4. After changing Node1 as primary sync restarted and everything sync.
    2. When Primary node1 is down on WSFC1, same issue as mentioned above.
    3. Same behavior happening after failover to Node3.  If all nodes are up, node1, node3 are primary nodes everything works as expected. If node1 and node3 down, sync is not happening. Please help to provide the details to resolve the issue.
    4. Thanks for everyone.

     

  • June 8, 2022 at 4:10 pm

    #4046322

    Thanks for posting your issue and hopefully someone will answer soon.

    This is an automated bump to increase visibility of your question.

  • June 14, 2022 at 5:28 pm

    #4049421

    I am guessing you have installed SQL server as a cluster. When you configure Always On, there are some things that only need doing once, and some that need doing for every cluster node.  I will go through the list of what I do which should cover what is needed.

    All of these processes are included in SQL Finebuild. If you download it and look in module FineBuild4InstallXtras.vbs in section 4AL and its subroutines you can find the syntax of all commands used.

    1) Enable Always On on each node of the cluster, and restart SQL for each node

    2) Create the mirror endpoint, first node only

    3) Create the AG and listener, first node only

    4) Configure the AG with GRANT CREATE ANY DATABASE and allow it to start up the AlwaysOn_health event session, first node only

    5) Add some databases to the AG on the intended primary node of the dAG

    If you are configuring a Distributed Availability Group (dAG), do the above on both clusters before setting up the dAG.

    For the dAG:

    a) Create the dAG on the active node of each SQL cluster

    b) Verify that all DBs in the primary AG are getting created and synchronised on the secondary AG via the dAG

    c) Create a DNS alias for the dAG. Point this to the listener for the current primary node of the dAG

    d) Create a linked server entry for all listeners in the dAG, both SQL clusters

    e) If you connect your applications to SQL via DNS aliases (you really should be doing this), point these aliases to the alias for the dAG

    Test Failover for each cluster and for the dAG. When you fail over the dAG also manually update the dAG alias to point to the new listener name. Your applications should then connect ok with no changes needed to connection strings. Verify that DB updates are flowing from the new primary cluster back to the old (now secondary) cluster.

    If in the future you ever want to re-host to a new set of servers, remember that the primary cluster in a dAG can link to multiple dAGs. This will allow you to replicate your DBs to multiple target servers, and ease your migration process.

    Original author: https://github.com/SQL-FineBuild/Common/wiki/ 1-click install and best practice configuration of SQL Server 2019, 2017 2016, 2014, 2012, 2008 R2, 2008 and 2005.

    When I give food to the poor they call me a saint. When I ask why they are poor they call me a communist - Archbishop Hélder Câmara

  • July 25, 2022 at 3:06 pm

    #4067925

    Problem:

    When Primary node1 is down, AG1 is enabled but AG2 disabled, AG2 enables after 5hrs. Is there any setting where I can change the time to reduce.

    Thanks for everyone.

Viewing 4 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic. Login to reply