Setup: Virtual Host, running server 2012, hosting around 10 production VMs
Problem: Last night, 2 VMs stopped with no warning, event log notification, or any information. The rest of the VMs on this host were running fine. We caught this error, and since it was before business hours, decided to restart the vhost. Once the vhost came back up, all the VMs had the exact same issue that the other two had previous. Except for one. There is 1 vm out of 10 that is still running fine. The error message in its entirety is here:
[Window Title] Hyper-V Manager [Main Instruction] An error occurred while attempting to start the selected virtual machine(s). [Content] 'VMNAME' could not initialize. An attempt to read or update virtual machine configuration failed. [Expanded Information] 'VMNAME' could not initialize. (Virtual machine ID IDREMOVED) 'VMNAME' could not read or update virtual machine configuration: Unspecified error (0x80004005). (Virtual machine ID IDREMOVED) [^] Hide details [Close]
What we have checked so far:
What we have done in the meantime: Thankfully we have a disaster plan in place which allowed us to failover to our replication server which is functioning perfectly. All machines are currently working just fine there, albeit a little slow. We are running short on ideas other than to remove all the broken vms from the vhost (config files and vhds) and have the replication server replicate over back to the vhost (essentially creating a new vhd and config file on the vhost, then we can failover to the main machine again.
Anyone have any thoughts?
Check my answer here:
what you might find useful there:
a suggestion to use Process Monitor to find out exactly what is going wrong in the process
if your problem is broken file permissions on your config files, and it seems like a possibility, you can use or adapt my batch file to restore permissions (be careful with that, though!).
User contributions licensed under CC BY-SA 3.0