Question

In SQL Server Basic Availability Groups docs, I found this interesting line:

Basic Availability Groups for SQL Server 2017 on Linux support an additional configuration only replica.

  1. What is a configuration only replica?
  2. Why is it only supported on Linux?
Was it helpful?

Solution

In linux, there is no "Windows Cluster" (which is used behind the scene for AGs). In order to have automatic failover, you need to use something like Pacemaker which is like the windows Cluster but under Linux.

Pacemaker doesn't have the same integration with SQL and the "configuration only" act as a Withness for the cluster (so that you don't end up with split brain scenario). If you build an AG with 3 node, you do not require the "Configuration only".

That also explain why it is only available under Linux (as it's not required under Windows, the cluster being doing the job).

As mentionned by dhchen, this node does not contain the user databases and cannot be failed over to. It only contains the AG information and can be installed on a SQL Express even if you AG is build on top of the Enterprise Version.

P.s. This is true for all AG under Linux (not only the Basic AG)

OTHER TIPS

It is just the configuration data no user data - so probably of limited use

There is more information available here :-

https://docs.microsoft.com/en-US/sql/t-sql/statements/create-availability-group-transact-sql?view=sql-server-ver15

to quote :-

CONFIGURATION_ONLY Specifies that the primary replica synchronously commit availability group configuration metadata to the master database on this replica. The replica will not contain user data. This option:

Can be hosted on any edition of SQL Server, including Express Edition. Requires the data mirroring endpoint of the CONFIGURATION_ONLY replica to be type WITNESS. Can not be altered. Is not valid when CLUSTER_TYPE = WSFC. For more information, see Configuration only replica.

Licensed under: CC-BY-SA with attribution
Not affiliated with dba.stackexchange
scroll top