we have a production HyperV cluster of 3 datacenter 2008R2 hosts, connected to a SAN via sas. we have 3-4 servers on each.
I have created a Test server on a hyperv host and wanted to test a 'live' migration - but it fails complaining a resource is offline or failed (there are none). I have noted in the event log it has Virtual network switch errors saying it cant find 'virtual network A' ...
At the beginning to try and help me track what virtual server was connected to what NIC i thought id be clever and tag them a unique name. Then it seems it failed failover validation and i worked out to migrate the server to another host it needs to see the same 'name' network on the other hosts.... so i renamed all the virtual networks back to the default - that went through ok.
So it seems i have an issue where it wont let me migrate a machine from host to host and it looks like it might still be refering to the old virtual names.
servers run 24/7 and i have not rebooted them for some time.
Is it possible to re-run the validation - which i assume will knock the resources offline - without causing a problem and would this enable the cards to be correctly named in the cluster?
I just dont want the situation whereby i run the validation again and it fails and i have a broken cluster or somthing disastrous like that...
I started as abit of a hyperv newbie and it was definitely learning as i went - albeit the servers been running fine for months i just want to pre-empt any issues and of course the whole point of the cluster is to enable us to move machines about for maintenance etc.
any help appreciated