I recently had a question around what happens to the compliance state of Virtual Machines that are using VM Storage Profiles which are built on top of storage capabilities surfaced by a VASA provider, and that VASA provider has an outage and is no longer reachable.
The current behavior is that once the provider becomes unreachable, its status is updated accordingly in the vCenter inventory, but the storage capabilities (and other inventory) data reported by the VASA provider is maintained for a period of time before being eventually flushed from the database. The point is that the data is not immediately removed from database due to the provider becoming unreachable. This gives you a window to address the issue at hand. So long as this data is present, VM compliance status is not affected.
However, the data will be flushed (removed first, and then re-populated) during a VASA provider sync operation. Since the provider is unreachable, the sync will always fail. The VASA provider sync operation is a manual process. However, if the provider is down for an extended period of time (7 days by default), the provider and data will be removed from database. This is a configurable setting in the SMS config.xml file.
The data will also be removed if the SMS service is restarted as this involves a sync operation – again, the sync will fail if the provider is not reachable and the provider and VASA data will be removed from database.
Once the provider/VASA data is removed from the database, any VM Storage Profiles that previously had a compliant state for VMs using VASA provided storage capabilities will now show non-compliant.
Get notification of these blogs postings and more VMware Storage information by following me on Twitter: @VMwareStorage
