Closed shepherdjerred closed 3 years ago
I believe you've set TAILSCALE_KEY to be an Ephemeral authkey, which is what we recommend for this kind of application because they clean themselves up shortly after the Github runner completes.
Ephemeral nodes only get an IPv6 address, the fd7a:115ca1e0efe3::... addresses seen in the admin panel.
Your workflow is trying to connect to san IPv4 address 100.76.222.102, which isn't going to work.
You can find the IPv6 address of that node using tailscale ip -6 hostname
and use that in the workflow, which should work.
Ah, that makes sense! Thanks for getting back to me. I'll try setting up a AAAA record for my domain and see if I can get Ansible to connect via IPv6 instead of IPv4.
That fixed it! I used tailscale ip -6
on the machine I was trying to connect to. I created a AAAA record pointing to the output of that command. GitHub actions was able to connect when I re-ran a failed workflow.
Thanks for the help! Tailscale is so easy to use.
An error occurs when using this action.
Workflow failure: https://github.com/shepherdjerred/ansible-playbook/runs/3354929715?check_suite_focus=true
Workflow .yml file:
Action output:
Despite failing, the action succeeds and prints "Success". I'm able to see GitHub action workers in my Tailscale console.
The workflow then fails on the next step which relies on being able to connect to my Tailscale network.