Skip to content
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
300 lines (194 sloc) 11.6 KB


Expose your local endpoints to the Internet

Build Status License: MIT Go Report Card Documentation


inlets combines a reverse proxy and websocket tunnels to expose your internal and development endpoints to the public Internet via an exit-node. An exit-node may be a 5-10 USD VPS or any other computer with an IPv4 IP address.

Why do we need this project? Similar tools such as ngrok or Argo Tunnel from Cloudflare are closed-source, have limits built-in, can work out expensive and have limited support for arm/arm64. Ngrok is also often banned by corporate firewall policies meaning it can be unusable. Other open-source tunnel tools are designed to only set up a static tunnel. inlets aims to dynamically bind and discover your local services to DNS entries with automated TLS certificates to a public IP address over its websocket tunnel.

When combined with SSL - inlets can be used with any corporate HTTP proxy which supports CONNECT.

Conceptual diagram for inlets

Who is behind this project?

inlets is brought to you by Alex Ellis, the founder of OpenFaaS.

The mission of OpenFaaS is to Make Serverless Functions Simple for developers. With OpenFaaS you can package any code, binary or microservice into a Serverless Function and deploy to Kubernetes or Docker Swarm without repetitive boiler-plate coding or complex YAML files. OpenFaaS has over 15.5k GitHub stars, 180 contributors and a growing end-user community.

Make a donation to support inlets via PayPal.


Initial goals:

  • automatically create endpoints on exit-node based upon client definitions
    • multiplex sites on same port through use of DNS / host entries
  • link encryption using SSL over websockets (wss://)
  • automatic reconnect
  • authentication using service account or basic auth
  • automatic TLS provisioning for endpoints using cert-magic
    • configure staging or production LetsEncrypt issuer using HTTP01 challenge

Stretch goals:

  • discover and configure endpoints for Ingress definitions from Kubernetes
  • configuration to run "exit-node" as serverless container with Azure ACI / AWS Fargate
  • automatic configuration of DNS / A records
  • configure staging or production LetsEncrypt issuer using DNS01 challenge


  • tunnelling plain (non-HTTP) traffic over TCP


Unlike HTTP 1.1 which follows a synchronous request/response model websockets use an asynchronous pub/sub model for sending and receiving messages. This presents a challenge for tunneling a synchronous protocol over an asynchronous bus. This is a working prototype that can be used for testing, development and to generate discussion, but is not production-ready.

  • There is currently no authentication on the server component The tunnel link is secured via -token flag and a shared secret
  • The default configuration uses websockets without SSL ws://, but to enable encryption you could enable SSL wss://
  • There is no timeout for when the tunnel is disconnected timeout can be configured via args on the server
  • The upstream URL has to be configured on both server and client until a discovery or service advertisement mechanism is added advertise on the client

Get started: Install the CLI

You can install the CLI with a curl utility script, brew or by downloading the binary from the releases page. Once installed you'll get the inlets command.

Utility script with curl:

# Install to local directory
curl -sLS | sh

# Install to /usr/local/bin/
curl -sLS | sudo sh

Via brew:

brew install inlets

Binaries for Linux, Darwin (MacOS) and armhf are made available via the releases page. You will also find SHA checksums available if you want to verify your download.

Test it out

  • On the server or exit-node

Start the tunnel server on a machine with a publicly-accessible IPv4 IP address such as a VPS.

./inlets server --port=80

Note: You can pass the -token argument followed by a token value to both the server and client to prevent unauthorized connections to the tunnel.

Example with token:

token=$(head -c 16 /dev/urandom | shasum | cut -d" " -f1); ./inlets server --port=8090 --token="$token"

Note down your public IPv4 IP address i.e.

  • On your machine behind the firewall start an example service that you want to expose to the Internet

You can use my hash-browns service for instance which generates hashes.

Install hash-browns or run your own HTTP server

go get -u
cd $GOPATH/src/

port=3000 go run server.go 
  • On your machine behind the firewall

Start the tunnel client

./inlets client \
 --remote= \

Replace the --remote with the address where your other machine is listening.

We now have an example service running (hash-browns), a tunnel server and a tunnel client.

So send a request to the public IP address or hostname:

./inlets client --remote= --upstream  ""
curl -d "hash this" -H "Host:"
# or
curl -d "hash this"
# or
curl -d "hash this"

You will see the traffic pass between the exit node / server and your development machine. You'll see the hash message appear in the logs as below:

~/go/src/$ port=3000 go run server.go 
2018/12/23 20:15:00 Listening on port: 3000
"hash this"

Now check the metrics endpoint which is built-into the hash-browns example service:

curl | grep hash


For development you will need Golang 1.10 or 1.11 on both the exit-node or server and the client.

You can get the code like this:

go get -u
cd $GOPATH/src/

Contributions are welcome. All commits must be signed-off with git commit -s to accept the Developer Certificate of Origin.

Take things further

You can expose an OpenFaaS or OpenFaaS Cloud deployment with inlets - just change --upstream= to --upstream= or --upstream= You can even point at an IP address inside or outside your network for instance: --upstream=

You can build a basic supervisor script for inlets in case of a crash, it will re-connect within 5 seconds:

In this example the Host/Client is acting as a relay for OpenFaaS running on port 8080 on the IP within the internal network.


while [ true ] ; do sleep 5 && ./inlets client --upstream=  ; done


while [ true ] ; do sleep 5 && ./inlets server --upstream= ; done

Run as a deployment on Kubernetes

You can even run inlets within your Kubernetes in Docker (kind) cluster to get ingress (incoming network) for your services such as the OpenFaaS gateway:

apiVersion: apps/v1beta1 # for versions before 1.6.0 use extensions/v1beta1
kind: Deployment
  name: inlets
  replicas: 1
        app: inlets
      - name: inlets
        image: alexellis2/inlets:2.0.3
        imagePullPolicy: Always
        command: ["./inlets"]
        - "client"
        - "--upstream=http://gateway.openfaas:8080,http://endpoint.openfaas:9090"
        - "--remote=your-public-ip"

Replace the line: - "--remote=your-public-ip" with the public IP belonging to your VPS.

Run on a VPS

Provisioning on a VPS will see inlets running as a systemd service. All the usual service commands should be used with inlets as the service name.

Inlets uses a token to prevent unauthorized access to the server component. A known token can be configured by amending prior to provisioning

# Enables randomly generated authentication token by default.
# Change the value here if you desire a specific token value.
export INLETSTOKEN=$(head -c 16 /dev/urandom | shasum | cut -d" " -f1)

If the token value is randomly generated then you will need to access the VPS in order to obtain the token value.

cat /etc/default/inlets 

How do I enable TLS / HTTPS?

  • Create a DNS A record for your exit-node IP and the DNS entry (replace as necessary).

  • Download Caddy from the Releases page.

  • Enter this text into a Caddyfile replacing with your subdomain.

proxy / {

proxy /tunnel {
  • Run inlets server --port 8000

  • Run caddy

Caddy will now ask you for your email address and after that will obtain a TLS certificate for you.

  • On the client run the following, adding any other parameters you need for --upstream
inlets client --remote wss://

Note: wss indicates to use port 443 for TLS.

You now have a secure TLS link between your client(s) and your server on the exit node and for your site to serve traffic over.

Where can I get a cheap / free domain-name?

You can get a free domain-name with a .tk / .ml or .ga TLD from - make sure the domain has at least 4 letters to get it for free. You can also get various other domains starting as cheap as 1-2USD from

Namecheap provides wildcard TLS out of the box, but freenom only provides root/naked domain and a list of sub-domains. Domains from both providers can be moved to alternative nameservers for use with AWS Route 53 or Google Cloud DNS - this then enables wildcard DNS and the ability to get a wildcard TLS certificate from LetsEncrypt.

Digital Ocean

If you are a Digital Ocean user and use doctl then you can provision a host with ./hack/ Please ensure you have configured droplet.create.ssh-keys within your ~/.config/doctl/config.yaml.


Scaleway offer probably the cheapest option at 1.99 EUR / month using the "1-XS" from the "Start" tier.

If you have the Scaleway CLI installed you can provision a host with ./hack/

Running over an SSH tunnel

You can tunnel over SSH if you are not using a reverse proxy that enables SSL. This encrypts the traffic over the tunnel.

On your client, create a tunnel to the exit-node:

ssh -L 8000: exit-node-ip

Now for the --remote address use --remote ws://

You can’t perform that action at this time.