2

I am attempting to fail over a SQL cluster with 2 nodes using Zerto replication. I am running Server 2016 Standard and SQL 2017 standard.

My network is not stretched, so I need to failover using a different vlan. I am only attempting to bring up one node of the cluster (the active node that holds all the rolls).

I am obviously running into all kinds of issues getting the cluster to start since there is a new IP range. I have looked all over and was surprised to not be able to find anything on this.

When I attempt to connect to the cluster I get "The operation has failed. An error occurred while opening the clustered role".

Has anyone attempted to do a full failover of a virtual machine in a Microsoft failover cluster to a different VLAN at a DR location?

Marcello Miorelli
  • 215
  • 1
  • 12
oke
  • 21
  • 2

2 Answers2

1

If you are open to redoing the architecture, SQL AlwaysOn works well with multiple subnets. I recommend using the same version of SQL for all replicas though.

https://www.mssqltips.com/sqlservertip/4597/configure-sql-server-alwayson-availability-group-on-a-multisubnet-cluster/

Systemic
  • 31
  • 2
0

Our problem was actually not network related. When replicating the RDMs, Zerto did not replicate using Thick Eager Zero, which was needed to get the cluster disks to be recognized. We replicated the disks and then made the disks thick eager zeroed via command line and we were able to get the node up.

oke
  • 21
  • 2