Public | Automated Build

Last pushed: 2 years ago
Short Description
Short description is empty for this repo.
Full Description


Software for exposing an open interface for Lighthouse on cloud providers.

Quick Setup With Boot2Docker

Docker Host

Ensure DOCKER_HOST env var is set, call echo $DOCKER_HOST to double check the address.

Start Boot2Docker

boot2docker up

Pull beacon

docker pull lighthouse/beacon:latest

Run beacon

docker run -t -i -e "DOCKER_HOST=tcp://" -p 5000:5000 \
    lighthouse/beacon:latest -token foobar -h -driver local

Try it out!

  • Check current driver curl -H "Token: foobar"
  • Check available vms curl -H "Token: foobar"

Running With Go


go get


go install




go test

Running With Docker


go get


docker build -t beacon $GOPATH/src/


docker run -d -p 5000:5000 -h


  • -driver gce Driver to use when interfacing with the vm provider.
  • -h Address to listen on when hosting the server.
  • -key server.key Path to private key used for hosting TLS connections.
  • -pem server.pem Path to Cert used for hosting TLS connections.
  • -token 123abc Authentication token used to grant access to the beacon api.


To make successful api calls to beacon from a client you will need the generated auth Token which is logged on app startup.
That token must be in the header of each request as "Token" to preform any api call. Otherwise you will be greeted with a 401 status code.



Interfaces with boot2docker and uses the address stored in $DOCKER_HOST to make requests. This also works if your running inside Docker such that you can use..

-e "DOCKER_HOST=tcp://your.docker.i.p:2375"

to manually declair the host of the Docker daemon. For example...

docker run -d -p 5000:5000 -e "DOCKER_HOST=tcp://" beacon -h -driver local


Interfaces with Compute Engine and requires that the hosting vm has "Compute" read/write privalages to the Project to detect existing vms.

Digital Ocean

Interfaces with Digital Ocean and requires that an api token be set to the hosting vm's "User Data". Read more info here.


Instead of relying on a provider api you can manually create a config file that list available ips of vms you want beacon to communicate with.
For example all you have to do is drop a config.json into the running directory of Beacon and it will take care of the rest for you.
A simple config.json can look something like this.

    {"Host": "", "Port": "2375"},
    {"Host": "", "Port": "80"}
Docker Pull Command
Source Repository