Skip to main content

Using the edge bridge

The edge bridge is a small service you run at your location for the cases where Cobblr can't reach a machine directly: a firewall in between, machines at a different site, or software pinned to a specific PC like LightBurn. See Digital fabrication for when you need it. This page is about setting it up and running it.

Where to run it​

Run the bridge where the machines are, on the same network as them:

  • In Docker, alongside a self-hosted Cobblr or on its own.
  • As a standalone binary on a small always-on box, like a Raspberry Pi or a NAS.
  • On a specific PC, when that's where the software lives, like the computer running LightBurn.

One bridge can run more than one machine at once, each on its own path.

Connecting it to Cobblr​

The bridge always dials outward to Cobblr, so you don't open any ports or expose anything on your network. You point it at your Cobblr instance, and Cobblr then sees the machines it manages.

Configuring machines​

The bridge runs a driver per machine. Built-in drivers cover common hardware; for anything else you write one (see Edge-bridge drivers). Each machine is an instance with its own config, and you map it to a machine in Cobblr.

Installing one​

You install a bridge from inside Cobblr, on the Edge bridges page: give it a name, mint its token (least-privilege, it can only do bridge things), and copy the generated command, as docker run or a compose block. Run that where the machines are, and the page shows the moment the bridge dials in. Nothing to configure by hand; the command carries everything.

Staying current​

A connected bridge keeps itself up to date from your own Cobblr instance: it fetches its code from the release your instance serves, so upgrading Cobblr upgrades the bridges that talk to it. No separate update chore per box.