Pair a runner
The runner works on the machine that holds your repositories, which can be the server itself or another machine. It connects out to the server over a WebSocket, so the runner machine needs no open port.
1. Create a pairing code
Section titled “1. Create a pairing code”In Nautelle, open Admin › Runners and choose Add runner. The dialog
shows a code like ABCD-EF23. It is valid for 10 minutes and works once.
2. Install and pair
Section titled “2. Install and pair”On the runner machine:
brew install andreybegma/tap/nautelle-runner # or: npm i -g nautelle-runnernautelle-runner pair --server <NAUTELLE_URL>/api --code XXXX-XXXX --allow-root ~/dev--serverisNAUTELLE_URLfollowed by/api, for examplehttps://dock.example.com/apiorhttp://localhost:8000/api.--allow-rootnames the folders the runner may work in, usually the folder that holds your clones. Repeat it for more than one. A runner works in no folder you have not allowed, and the server cannot change that list (see Runner security).
A runner refuses plain HTTP to any host but localhost, because the pairing
token would travel in clear text. Use a tunnel or a domain with HTTPS; on a
trusted network only, pair --insecure accepts an http:// server.
3. Keep it online
Section titled “3. Keep it online”nautelle-runner install-service # with Homebrew: brew services start nautelle-runnerinstall-service writes a systemd user unit on Linux or a launchd agent on
macOS and prints the command that starts it. The runner then appears online
under Admin › Runners.
Check it
Section titled “Check it”nautelle-runner statusshows the configuration, the event spool and whether the server answers. To allow another folder later:
nautelle-runner policy allow-root ~/workIts configuration and token live in ~/.config/nautelle/runner.json
(mode 600). To retire a runner, revoke it under Admin › Runners: its token
stops working at once.
Next: connect a project.