[PVE-User] cluster node issue after upgrade (1.3 -> 1.5)

[Gwenn Gueguen]

> Do you have defined any Storage (NFS) that is not accessable on that node?

There is no NFS storage.

I added iscsci volumes but it was after the upgrade from 1.3 to 1.5
and the problem was already occuring.

> Does 
> # pvesm list
> run without errors on that node?

Yes, no errors and storage pools are correctly listed.

> > Can I remove the node from the cluster and add it again without
> > loosing the virtual machines hosted on it?
> 
> Yes, you can do that (Although I would make a backup before).

I removed the "faulty" node from the cluster and then another node got
the issue:

"ERROR: 500 Server closed connection without sending any data back"

I removed the two faulty nodes from the cluster and then added them
again (after removing authorized_keys, known_hosts...) but there is
always one of these two nodes having this issue.

Do I have to reinstall this node (it only contains one virtual machine
so it should be easy) ?

-- 
Gwenn