Hi all, i'm in trouble with a 4 node 2012 hyper-v cluster. After some weeks of normal operations i cannot connect to failover cluster manager who tells me that "the object does not exists" even if i could connect in rdp to the virtual ip server, resolve the address with nslookup and so on. Meanwhile even Hyperv management console goes unresponsive and does not let me connect and manage the server (localhost or not is the same) the error in this case is "check the virtual machine management service is running and that you are authorized to connect the server". Vms hosted by these servers still are responsive and in normal operations but i cannot manage them and/or view any state. I then check the service trying to restart that but it hangs and or crash. The only way i can solve this situation is shutdown the host to migrate the vms with several (minimal, modearate) downtimes. looking around in services and dependecies i noticed the RPC object seems to not exists and there is no way to restart the service.
Now i've restarted the servers and everything works fine, even if i think this issue should come back in some days/weeks. what should i check? maybe the failures comes after a live migration but i cannot tell this were an unluky situation or that should start the problem: everyday migrations don't give us any problems. this is going to be very frustrating.