For the last few days I do some reorganization of our Virtual Infrastructure. One of the steps of this reorganization is upgrade from VMFS 3 to VMFS 5 for all our storages connected to the main HA cluster.
Although it is possible in ESXi to upgrade to VMFS5 in-place, I decided to completely remove and re-create storages. The main reason was that previously all arrays were sliced in several 2TiB(-1MiB) logical drives, and I wanted to create single logical unit for each storage subsystem.
But almost each time I tried to remove an old datastore from ESXi, I've received an error:
Status: The vSphere HA agent on host '10.0.0.1' failed to quiesce file activity on datastore '/vmfs/volumes/XXXXXXXXXXXXX'. To proceed with the operation to unmount or remove a datastore, ensure that the datastore is accessible, the host is reachable and its vSphere HA agent is running.
Well, solution is quite simple. Just go to the host which generated this error (in my example it's 10.0.0.1) in "Inventory > Hosts and Clusters" view, and remove a the datastore from the Summary tab of that host.