I have a 3-node hybrid VSAN cluster with 600gb HDD capacity drives that I wanted to replace with 1200gb HDDs.
The steps that I planned to take (with guidance from support):
- Put host in maintenance mode (ensure availability)
- Delete diskgroup on host
- Replace disks
- Recreate diskgroup
- Let hosts resync
Things seemed to be going good until I hit steps 5, the resync was too fast or basically non-existent. I then tried to go to the next host and I got the error:
“Failed to enter maintenance mode in the current VSAN data migration mode due to insufficient nodes or disks in the cluster”
After working with support I found that diskgroup was disk format version 3 while the other were version 2 (I’ve been meaning to upgrade but wanted more capacity first). You can see the current Disk Format Version in (cluster name) -> Manage -> Settings -> Disk Management, Disk Groups, scroll to right.
The resolution is to set the maximum disk format version on all of the hosts to v2 and then re-build the diskgroup in question.
Here is the KB
You can monitor resync from the web-client (cluster name) -> Monitor -> Virtual SAN -> Resyncing Components or from RVC (cd to cluster) vsan.resync_dashboard