At least one snapshot has to be active to create a new snapshot for this VM

Support
#1

When trying to create a snapshot for a VM that already has one snapshot I get the folowing error: “At least one snapshot has to be active to create a new snapshot for this VM”.

A VMware snapshot cannot be active or inactive so how should I interpret this error message?

Using the VMware management console I am able to create the second snapshot for this VM without any problem. This makes it still more confusing.

Please can someone explain this error message to me.

(The message originates from app/models/vm_or_template/operations/snapshot.rb)

#2

Interesting, it looks like this only happens when there is at least one snapshot already (the first condition is !snapshots.blank?).

I’m not sure what provider requires this “current?” check but it should be moved into that provider. Can you open an issue http://github.com/ManageIQ/manageiq/issues/new ?

#3

Went through the history and this was actually there for vmware, I suspect to prevent creating new snapshots if there is a “broken” chain or something needs consolidation. Active isn’t really the correct term, it is looking for if a snapshot is a “currentSnapshot” (https://code.vmware.com/apis/358/vsphere/doc/vim.vm.SnapshotInfo.html#currentSnapshot).

Can you kick a full-refresh and see if you still don’t have any current snapshots for that VM?

#4

While ManageIQ complained about this we could create a snapshot for the same VM using the VMware vSphere client without any problem.

I believe the VMware documentation about the “currentSnapshot” property is misinterpreted when adding this check to ManageIQ. Creating a new snapshot should be possible independent of another snapshot having this “currentSnapshot” property set.

#5

I’m skeptical of the check as well, but want to confirm that the VM you were trying to create a snapshot on actually had no current snapshots in VMware inventory and it wasn’t a bug in our inventory refresh.

#6

The VM had one snapshot and our tenant wanted to create a second snapshot for this VM. We did not do a full provider refresh. Currently we cannot reproduce the issue.