I created one Windows 2022 VM in vCenter directly and then tried to add this vm to Morpheus manually.
When I added, the status is showing “available”, but after a while (2 or 3 mins), this vm will be deleted automatically.
I have disabled the firewall and started Windows Remote Management service on windows vm.
Is there any reason for this behavior? And any step I need to follow to add vm to Morpheus manually?
(actually i tested to add one CentOS vm as well but if failed with the same)
I tried to add vm to a cloud of type Morpheus. this time it is not deleted automatically. but the status is showing available and power status showing question mark all the time.
Is this expected behavior?
As my understanding, at least the power and status should show green.
This is expected because we are not actually linking to any VM, rather a SSH or WinRM connection and treating it like a bare metal machine.
Really, you should be inventorying the resource pools / clusters you desire and we’ll sync all of the VM data. Otherwise, you are not going to have true life cycle management, (if you delete or power state change we’ll never make a change).
Thank you for your explanation and understood now. But let me change my question.
Actually this is the request from one customer, they don’t want to use cloud discover capability to sync-up all the VMs at one shot, but instead they only want to add those VMs which they need to manage thru Morpheus, and meanwhile they don’t want to install Morpheus agent on them.
I have one additional question about the lifecycle management for the VM which I added manually.
Lets say I manually add one VM and convert it to managed VM with agent installation. In this case, can we have below “management” capability for the VM?
Power controls
Teardown / Deletion
Resize capabilities
Add/remove disk
Add/remove CPU
Add/remove Memory
State Sync (Up/Down)
Health Statistics or Metrics
Guidance or Resizing Recommendations
Also, could you please explain some more about how “This would possibly be circumventing the licensing and our ELA”?
Unsure, would have to test. I believe with proper RPC creds / Agent you can get power Operations
Teardown / Deletion
No.
Resize capabilities
No to all of these
State Sync (Up/Down)
Health will be based on Agent comms. State will be unknown most likely.
Health Statistics or Metrics
OS level health stats, yes with agent.
Guidance or Resizing Recommendations
Yes based on above health stats.
Also, I’m unsure of your current licensing model / agreement. I just wanted to highlight you may want to review if there are any definitions on cloud inventory.
We also have the requirement not to discovert all VMs on a vCenter in one go. We have implemented this by not globally authorising the service user used, but only on specific resources and objects. In this way, it is possible to authorise the user used in Morpheus to certain folders and thus only discover the VMs in these folders. If you want to manage a VM via Morpheus, you only have to move it in vCenter and start a sync in Morpheus.
With this solution it is also possible to reconfigure the VM, perform power controls, teardown etc. without the actual installation of the agent.