Controller IP issue after adding wireguard interface

[camja014]

Hi,

I added a wireguard interface to one of my controller nodes in my homelab cluster so that I could eventually install Cilium and use this interface as an egress gateway for certain traffic.

I configured the interface by creating the following patch file:

machine:
  network:
    interfaces:
      - interface: wg0
        mtu: 1500
        addresses:
          - 10.2.0.2/32
        wireguard:
          privateKey: <private key>
          listenPort: 51820
          peers:
            - publicKey: <public key>
              allowedIPs:
                - 0.0.0.0/0
                - ::/0
              endpoint: <endpoint IP>:51820

However I'm experiencing some issues after adding the interface. When trying to run a debug container in the kube-system namespace in an attempt to test that the interface configuration is working, I'm getting the following error:

$ kubectl debug -n kube-system -it --image alpine node/talos-control-03
--profile=legacy is deprecated and will be removed in the future. It is recommended to explicitly specify a profile, for example "--profile=general".
Creating debugging pod node-debugger-talos-control-03-6xqkw with container debugger on node talos-control-03.
All commands and output from this session will be recorded in container logs, including credentials and sensitive information passed through the command prompt.
If you don't see a command prompt, try pressing enter.
warning: couldn't attach to pod/node-debugger-talos-control-03-6xqkw, falling back to streaming logs: Internal error occurred: error sending request: Post "https://10.2.0.2:10250/attach/kube-system/node-debugger-talos-control-03-6xqkw/debugger?input=1&output=1&tty=1": dial tcp 10.2.0.2:10250: i/o timeout
Error from server: Get "https://10.2.0.2:10250/containerLogs/kube-system/node-debugger-talos-control-03-6xqkw/debugger": dial tcp 10.2.0.2:10250: i/o timeout

Also, when running kubectl describe on any pod in the kube-system namspace running on that node, they all show as having an IP address of that interface (10.2.0.2).

Is there any configuration in Talos or one of the k8s services to disable automatic use of this interface? I wanted to manually route traffic through it using cilium when I get to that point.

Thanks!

[smira]

Yes, see documentation please.