Mesh Tunnels

Mesh tunnels give every device in a fleet its own public address under a single tunnel. Name each device once, and it's reachable by name forever — no matter where it's running, what network it's on, or how many siblings it has.

When to use a mesh tunnel

A standard tunnel is one connection, one URL. If you're running ten Raspberry Pis, you'd need ten tunnels and ten tokens — and you'd have to invent your own naming system on top. That doesn't scale.

Mesh tunnels fix this. One tunnel, one token, and every device that connects gets:

• Its own subdomain: sensor-1.abc123.tunnel.localport.dev
• Its own TCP port (for non-HTTP services)

Setting it up

1. Create a mesh tunnel

In the Localport dashboard, go to your team and click Mesh. Create a new mesh tunnel. You'll get a single token that all devices share.

2. Connect devices

On each device, run the CLI with a unique --name:

# Device 1
localport http 8080 --token MESH_TOKEN --name sensor-1

# Device 2
localport http 8080 --token MESH_TOKEN --name sensor-2

# Device 3
localport http 8080 --token MESH_TOKEN --name camera-front

3. Access any device by name

Each device is reachable at its own URL:

https://sensor-1.abc123.tunnel.localport.dev
https://sensor-2.abc123.tunnel.localport.dev
https://camera-front.abc123.tunnel.localport.dev

For TCP devices, each also gets a dedicated port:

tcp://sensor-1.abc123.tunnel.localport.dev:25001
tcp://sensor-2.abc123.tunnel.localport.dev:25002

How device names work

The --name flag becomes part of the URL. Names are cleaned automatically for DNS compatibility:

If two devices use the same name, the second one gets a numeric suffix (-2, -3, etc.).

Reconnection behavior

When a device disconnects — network hiccup, reboot, whatever — its port and subdomain are reserved for 60 seconds. If it reconnects within that window, it gets the exact same address back. No URL changes, no downtime for monitoring systems.

Mixed protocols

Mesh tunnels support mixing HTTP, TCP, and TLS clients on the same tunnel. One device can serve a web dashboard over HTTP while another exposes SSH over TCP — both under the same mesh.

Requirements

• Available on all plans
• Each connected device counts toward your client limit

Use cases

• IoT fleets — Raspberry Pis, sensors, cameras, each individually addressable. IoT guide
• Dev environments — Every developer's machine gets its own URL within a shared tunnel
• Edge computing — Distributed compute nodes, each directly reachable
• Kiosks and displays — Manage screens in different locations through one tunnel

Next steps

• IoT Device Management — Deployment guide for device fleets
• Locked Tunnels — Add mutual TLS so only devices you've issued certificates to can join the mesh
• Shared Tunnels — Broadcast traffic to all clients instead
• Homelab Remote Access — Tunnel self-hosted services at home