It looks like if I exclude those three values it makes the VM. So maybe the better question is; if I'm taking these values from the template, am I not meant to override them? The description seems like an odd one to not be able to override if that's how it works.

I'm very new to doing things with APIs and running into this error. What exactly does "is an excess property and therefore is not allowed" mean as a response? Below is the body I'm sending via Bruno client. I'm sure I'm missing something simple, the docs were a bit of a struggle though.

Endpoint is https://xo.ducknet.org/rest/v0/pools/3c6777fc-e8d8-e065-7c30-37eb2800f472/actions/create_vm/

{
  "auto_poweron": true,
  "boot": false,
  "memory": 1073741824,
  "name_description": "test_desc",
  "name_label": "test_label",
  "template": "3ceb66f7-c137-ef4a-7a06-9af524ed5110"
}

Full response

{
  "error": {
    "body": {
      "message": "\"auto_poweron,memory,name_description\" is an excess property and therefore is not allowed",
      "value": {
        "name_description": "test_desc",
        "memory": 1073741824,
        "auto_poweron": true
      }
    }
  }
}

@AtaxyaNetwork It is Debian 12. I found the problem and will post again if there's more issue. The VM stopped using its static IP, not sure how that happened, but I re-added it to /etc/network/interfaces and it's working again.

@Andrew It's the same. It should respond with a 302 to /signin as the login page. It is listening on :80 too, I just forgot it in my filter. I'll edit the correct filter in the post.

I'm using the free version of the XOA appliance (not sources/paid/etc) and recently I've not been able to connect to XOA's web interface or SSH using v6 addresses. Not sure if it stopped working after a particular update, as I've only recently fixed my DNS in my lab and was using IPv4 addresses for a while.

[22:03 04] xoa@xoa:~$ ss -nl6tp | egrep "State|22|80|443"
State  Recv-Q Send-Q Local Address:Port  Peer Address:PortProcess
LISTEN 0      511                *:80               *:*          
LISTEN 0      128             [::]:22            [::]:*          
LISTEN 0      511                *:443              *:*          

Curl attempt from my laptop, which is on the same VLAN as XOA.

bryon@blaptop:~$ curl -v -6 xo.ducknet.org
* Host xo.ducknet.org:80 was resolved.
* IPv6: 2603:xxxx:xxxx:1d02::11
* IPv4: (none)
*   Trying [2603:xxxx:xxxx:1d02::11]:80...
* connect to 2603:xxxx:xxxx:1d02::11 port 80 from 2603:xxxx:xxxx:1d02:e76a:cab:34ad:16f5 port 48918 failed: Aucun chemin d'accès pour atteindre l'hôte cible
* Failed to connect to xo.ducknet.org port 80 after 3073 ms: Could not connect to server
* closing connection #0
curl: (7) Failed to connect to xo.ducknet.org port 80 after 3073 ms: Could not connect to server

I'm not sure how to get tcpdump onto XOA in order to make sure the traffic is making it there, but I'm confident I was able to reach it via IPv6 previously. The only significant change I've made is upgrading xcp-ng to 8.3.

How can I troubleshoot this?

@olivierlambert Thanks for the quick response! I overlooked the XCP-ng setting in the Hosts > Networking tab. I don't see v6 available there, is that the same where it's just not something XOA exposes in the UI, and I'd need to SSH or console into XCP-ng?

Is there a way to edit the IPv4 of XO via the web UI? Understood that the helper script doesn't handle v6 atm.

I'm re-IPing things due to my v6 delegation having changed. Is there a better way to do this than SSH/console to the XOA VM in question and editing nmcli or /etc/network/interfaces? I was looking in the XOA UI and I don't see a way to do this. I do see the xoa network static command in the docs here, but I'm unclear how I would set both v4 and v6 there, since each would have their own gateway and I don't think the command to set a second IP makes sense to me.

Same question for the XCP-NG host too.

I feel like both of these should be possible from the web UI that XOA provides, but cannot find it.