Restart the toolstack on each server.

Interesting read that mentions xcp-ng!

https://www.theregister.com/2022/09/30/citrix_xen/

Yes, it's technically possible, as long as the underlying hardware is compatible. For example, you can't put a server using AMD processors into a pool with Intel processors.

The server being moved into the existing pool will lose its local storage and all VMs stored there, so you will need to migrate those VMs to the main pool before the server is merged. Afterwards, you can move them back if desired.

This sounds like the same issue discussed here. If you are using XOA (the prebuilt appliance), then you should either contact support for the fix or wait until the next release (tomorrow?) and then retest using the latest release channel.

@DCT said in Cannot Delete a VM Template through CLI or XE:

xe template-param-clear uuid=39bf95ba-1f92-4a54-1c14-d9d800ceca30 param-name=is-a-template

Try this instead --

xe template-param-set is-a-template=false uuid=39bf95ba-1f92-4a54-1c14-d9d800ceca30

Then try to access the VM in XO and disable the Protect from accidental deletion option. Hopefully you will be able to delete it once that's done.

@julien-f Technically, it is possible with post processing as I mentioned above. Here's an example using jq and awk --

 xo-cli --list-objects type=VM-snapshot | jq '.[]."$snapshot_of"' | sort | uniq -c | awk -F " " '{print "{\"uuid\":" $2 ",\"count\":" $1"}"}'| jq ' select( .count > 1)'

Output looks like this --

{
  "uuid": "07604606-e937-5147-2dac-c1b4280df234",
  "count": 2
}
{
  "uuid": "0f11c701-c5e5-b09b-4c5f-00cf38153783",
  "count": 2
}
{
  "uuid": "83139da5-48d5-33f6-79f9-79ea91a7163c",
  "count": 2
}
{
  "uuid": "fb4703e4-bfd6-47b8-3fd1-f1f52f9ed234",
  "count": 2
}
{
  "uuid": "ff8b29e3-4445-5b22-e475-450270f2a7f3",
  "count": 2
}

Note: I am not proficient with these linux utilities so there may be better ways to accomplish this.

There was a recent change that resulted in the launched module being renamed. You need to redo your service so that it loads ./dist/cli.mjs instead of ./bin/xo-server

You're running out of memory. Allocate more memory to the VM. I believe the current build requirement is 3GB.

You need a space between the URL and the user name.

@stormi Updated and rebooted. I'll report back if I encounter any issues.

Maybe this is your issue -- https://xcp-ng.org/forum/topic/3933/host-status-in-xcp-ng-center

@rk9268vc You want it off, which is why I wanted you to verify the setting. Is there anything unusual about the original PC's configuration?

Have you tried the Recovery Start option on the VM's Advanced tab as a way to boot the VM?

@rk9268vc Switch back to UEFI and then check to see if Secure Boot is on or off.

Your screen shot shows bios for boot firmware, but you previously stated that options didn't work.

I think that you are on the right track. Have you tried changing the Boot Firmware setting under the VM's Advanced tab in XO?

@manilx It would likely help if you shared more details on your environment, such as --

• Client OS
• Browser
• MRD version
• Any proxies?
• Etc

Interesting. FWIW, I'm on the latest commit and just ran a metadata / xo config backup that completed successfully.

@alpenliebeanj This setting can be viewed / changed from the VM's Advanced tab in XO.

Could this be a BIOS vs UEFI issue.

@florent Thanks for the fix.

P.S. @Gheppy was the original reporter.