I've got a cluster of several hosts. Network is set to use a distributed switch.
I have the default VM restart priority set to disabled as I don't have enough capacity at the moment to power on all vms if a host fails.
A few days ago I had a host crash and now a problem I've seen before it back:
When I go to power on a vm (or one of my users with limited rights), it usually works but the network is disconnected.
If I try to click the connected checkbox, I get an error saying: invalid configuration for device '0'
There's only two options to fix the problem at this point:
- Change the portgroup to something else then change back to the original one and then I can click the connected box.
- Move the vm back to the host it was on originally before powering it on.
Once the vm powers on, I can then power it off, vmotion to a different host and then power it on again without any trouble. Even if the vm is powered, i can vmotion back to the original host and then reconnect the network.
This is not the first time I've has this problem. Makes me wonder if I had the vms set to restart, would everything be ok or would I still have this issue and end up with a bunch of powered on vms and no network.
Anyone ever seen this behavior before with HA?
Thanks
If it helps anyone else I opened a case with support and they pointed me to KB2014469
Restarting the management agents seems to help although I'm still not sure if this only happens because I don't ask to restart the vms (ie. if I ask to restart vms the network will come back correctly)
see below link.
Virtual machines lose network connectivity after HA fail over. (ESXi 5.5) | CloudXC
If you found my answers useful please consider marking them as Correct OR Helpful