Hi guys. I built out a pair of identical Windows server 2019 VMs yesterday to serve as an SQL cluster. Followed the instructions I found to the last detail. The servers are running on 2 separate esxi 6.7 HOSTs and backed by a shared, all-flash, Fibre channel connected datastore.
When I start the Failover Cluster validation the disk Failover test step fails and what I notice in the disk management console of the servers is that the validation wizard takes the shared disks offline and does not bring them online again.
The shared disk for the SQL and quorum are thick, eager zeroed.
Does anyone know if this is normal behavior because I'm dreading putting this into a production environment if it's not going to play nice when I really need the cluster to Failover.
Thanks in advance.
It should work ok from scratch if all the config is working ok. We have a legacy env with many of those. But now moving to a smb share to remove limitations and be able to remove the dependency from the infra.
Just in case, did you have the chance to create everything again from scratch? Maybe something is missing in the original config. (Also could be something related to OS)
I typically validate against the Windows cluster right after I set up the first node with the core resources and re-validate as I change things (like adding additional nodes or installing SQL).
Have you already added SQL are are you still working on the initial cluster build?
Does the validator give you any indication as to why the disk failed validation?