Cluster Disk not accessible on one node

[edmacfly]

I have a 2-node hyperconverged setup between two Windows Server 2019 machines using VSAN Free.

Windows updates were installed on Node 2 last night and it restarted.

I can now no longer move roles to that node using Failover Cluster Manager. I can see the ClusterStorage folder mounted on the root of Node 1 (along with a secondary ClusterStorage.000 folder) and have access to all the Hyper-V files within, however Node 2 is set as the Owner Node of both the CSV and Witness disk as shown by the Failover Cluster Manager and the Witness is mounted on there too.

The StarWind console and Powershell scripts seem to suggest that everything is synced, so I am a bit stumped as to what to do next (whilst limiting distruption). The only other thing I can see is that the date modified on the CSV1 and Witness disk images are different - the last modified time on Node 2 seems to match with the reboot time of that server whereas on Node 1 they are a few hours later.

Any help, most appreciated - thanks.

[yaroslav (staff)]

Hi,

Can you please check if the iSCSI sessions were reconnected? go to the iSCSI initiator and make sure the CSVs are connected from both sides.
I think this case is related more to Windows than to StarWind if HA devices are synchronized and accessible.
Also, do you have any antivirus installed on your hosts?

[edmacfly]

All CSVs are connected. I'm wondering if restarting the Cluster Service on the affected node might be the way forward. Looking at the logs, seems like there was some resyncing following the restart.

[yaroslav (staff)]

Resynchronization is a normal event.
Can you move the CSV to the affected node or will it fail? Any antivirus installed (e.g., Sophos, Acronis CyberProtect, etc)?