VMware - LabManager - CannotAccessNetwork error deploying VM

In a VMware LabManager environment with ~10 VMs and several networks, I encountered an issue where two would not deploy. The error seen on both was:

Unable to deploy virtual machines in resource pool "RESOURCEPOOLNAME".

  * DRS failed to find hosts to deploy the virtual machines on the resource pool "resgroup-149".
    o DRS failed to find host for virtual machine "003601-3601-VMNAME". vCenter reported: Network interface 'Network adapter 4' uses network '7a ab 37 50 df fc a8 2f-f0 b7 cd aa 34 3b a1 99', which is not accessible.
      + Unable to find host for virtual machine "CannotAccessNetwork"

The error also seemed to be repeated for each host in the cluster. It appeared that the VM would deploy on each host and then un-deploy. It seems to be a "network" error from the message. However other VMs in the environment connected to the same network, from the same datastore, and deployed to the same hosts without issue.

I google'd around and tried numerous fixes including changing the networks the VM connected to, removing and re-adding the virtual NIC, and replacing the virtual networks themselves. Another ID was to clone the VM to a LabManager template and deploy an additional copy of the VM, but that duplicate VM had the same error. VMware tech support suggested this KB article:
I started digging around on the ESXi host logs themselves. There was alot of info in there and I spotted what I believed to be the root cause error. Here is a bit from the log that I filtered by opID...

[2012-02-28 19:20:10.885 12270B90 info 'TaskManager' opID=319e0bb9-f7] Task Created : haTask--vim.dvs.HostDistributedVirtualSwitchManager.updateDVPortgroups-581655
[2012-02-28 19:20:10.886 FFE40B90 verbose 'NetworkProvider' opID=319e0bb9-f7] ActionAddDVPortgroup: dvs ea ca 37 50 f0 a5 43 ac-39 db 8a 49 9c d5 09 82 portgroup dvportgroup-14545
[2012-02-28 19:20:10.886 FFE40B90 info 'App' opID=319e0bb9-f7] AdapterServer caught exception: 0a56fe60
[2012-02-28 19:20:10.886 FFE40B90 info 'TaskManager' opID=319e0bb9-f7] Task Completed : haTask--vim.dvs.HostDistributedVirtualSwitchManager.updateDVPortgroups-581655 Status error
[2012-02-28 19:20:10.887 FFE40B90 info 'Vmomi' opID=319e0bb9-f7] Activation [N5Vmomi10ActivationE:0x470612e0] : Invoke done [updateDVPortgroups] on [vim.dvs.HostDistributedVirtualSwitchManager:ha-hostdvsmanager]
[2012-02-28 19:20:10.887 FFE40B90 verbose 'Vmomi' opID=319e0bb9-f7] Arg switchUuid:
[2012-02-28 19:20:10.887 FFE40B90 verbose 'Vmomi' opID=319e0bb9-f7] Arg configSpec:
[2012-02-28 19:20:10.897 FFE40B90 info 'Vmomi' opID=319e0bb9-f7] Throw vim.fault.AlreadyExists
[2012-02-28 19:20:10.898 FFE40B90 info 'Vmomi' opID=319e0bb9-f7] Result:
[2012-02-28 19:20:10.898 FFE40B90 error 'SOAP' opID=319e0bb9-f7] Method vim.dvs.HostDistributedVirtualSwitchManager.updateDVPortgroups threw undeclared fault of type vim.fault.AlreadyExists

Still this did not provide that much help. One thing that came back to me was even when I deleted ALL the network adapters and tried to deploy, the VM gave the error about "Network Adapter 4". I decided to actually take a look at the VMX file for the troubled VM. That is when I noticed FOUR network adapters (ethernet0,1,2,3) defined even when there were NONE listed in LabManager for the VM. Since LabManager complained about Network Adapter 4, I made a backup of the VMX file and deleted all the "ethernet3" lines in the VMX file.

PROBLEM SOLVED! The VM was then able to successfully deploy via LabManager. I performed the same fix to the other problem VM successfully. I am not sure the root cause, but this resolved my issue and everyone is happy.