Hi all, We encountered an issue during the upgrade process which results in RWX volumes not mounting in the harvester cluster which was caused because a longhorn setting was not applied. Has anyone done this without requiring a downtime such as a node by node rollout. This requires all volumes to be detached for these 2 longhorn settings to be applied. As you can see this has not been applied. After detaching all volumes it applies the setting. This is needed in the upgrade as the importer-prime pod in the upgrade creates an RWX that does not get mounted. kubectl get settings.longhorn.io storage-network storage-network-for-rwx-volume-enabled -n longhorn-system

It is not possible to enable storage network without stopping all the Virtual Machines and/or not having any volume attached on Longhorn

Do u know of any uprade path that does not require an RWX volume on the harvester cluster

RWX is required for Live Migrations

@quiet-smartphone-38704 what’s the harvester version? could you generate a support bundle for us to inspect?