October 17, 2019 at 11:11 am
Hi all, hope you can help. I'm not a DBA so have been googling as I go, so forgive my idiocy:
I've got a SQL server standard instance with a number of databases. The setup is as follows:
Servers 1 and 2 are in a SQL cluster with shared storage. Server 3 is a standalone instance for DR, with mirroring set up in high safety mode on the databases between the cluster and the DR box.
These all sit in a Windows Server 2012 failover cluster.
So - the plan was to have the instance running on S1 and S2, failing over as and when either of them fell over. If the worst happened and the cluster failed entirely, then the Windows cluster would point logins to S3, and mirroring would make the standalone instance on S3 become the principal.
What's actually happening is that when I reboot S1 to force S2 to become the primary node in the SQL cluster, the mirroring kicks off as well - the windows clustering moves the shared storage over from S1 to S2, but all the databases on S2 are now in a synchronising state, and S3 is the principal.
Any way to avoid this? The only way I can think is to disable automatic failover and have some sort of agent job running to check the condition of the cluster. That sounds messy though.... any ideas appreciated.
Thanks!
James
October 17, 2019 at 12:16 pm
have a look at Set-ClusterOwnerNode you should be able to set the order to s1, s2, s3 thus those cluster Services should attempt to Failover to s2 first. Yeah it's doable via GUI too, just can't remember where to look at.
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply