Hi Lab Manager gurus,
I need your help in evacuating a corrupt data store that is hosting four LM Service VMs used by host spanning. I'm attaching screenshots as well.
LM version: 4.0.4 on ESXi 4.1 Enterprise Plus.
1. Even though host spanning is enabled, we don't use it or need it, so it can be disabled.
2. I have moved all the other VMs out of this datastore using SSmove, and though the service VMs have been powered off since mid-2013, SSmove still thinks they are deployed.
3. I tried to disable host spanning in LM, but got an error:
Error in undeploying the virtual machine. Contact the Administrator if the problem persists.
Virtual machine, "VMwareLM-ServiceVM-I46-H2", not found in inventory.
Virtual machine not found in inventory.
Along with:
Failed to undeploy Service Virtual Machine. Try to repair the host.
4. Repairing the host gave the error:
Failure occurred during repair of host "xxxxxxxxxxxx.com". If this is due to an temporary failure, try repair again, otherwise unprepare and then prepare the host.
Error in undeploying. Contact the Administrator if the problem persists.
Error in undeploying the virtual machine. Contact the Administrator if the problem persists.
Virtual machine, "VMwareLM-ServiceVM-I46-H1", not found in inventory.
Virtual machine not found in inventory
5. Unprepare option is greyed-out in the menu for the host.
How do I make the four LM service VMs accessible again to Lab Manager so it can go ahead and disable host spanning?
I'm assuming disabling host spanning will get rid of these service VMs. Registering a VM in VC did not make it visible to LM - I still got the "VM not found in inventory" error.
Would some DB updates resolve this?
The FsDirs for these service VMs are 1, 2, 229 and 7672.
In table DeployedFsDir I see rows for dir_ids 1,2, 229, 7672
dir_id vs_id
1 1
2 2
229 215
7672 6698
Table DeployedVirtualServer has rows for vs_ids 1,2,215, 6698
DistributedVirtualSwitch shows one DVSwitch as is_enabled = False
Any help appreciated!
Khare