nwaku/docs/operators/docker-quickstart.md

Quickstart: running nwaku in a Docker container

This guide explains how to run a nwaku node in a Docker container.

Prerequisites

Make sure you have Docker installed. Installation instructions for different platforms can be found in the Docker docs.

For example, to use Docker's convenience script for installation:

curl -fsSL https://get.docker.com -o get-docker.sh
sudo sh get-docker.sh

Step 1: Get Docker image

Nwaku Docker images are published to the Docker Hub public registry under wakuorg/nwaku. For specific releases the published images are tagged with the release version, e.g. wakuorg/nwaku:v0.20.0. Images are also published for each commit to the master branch in the nwaku repo and tagged with the corresponding commit hash. See wakuorg/nwaku on Docker Hub for a full list of available tags.

To pull the image of your choice, use

docker pull wakuorg/nwaku:v0.20.0 # or, whichever tag you prefer in the format wakuorg/nwaku:[tag]

You can also build the Docker image locally using

git clone --recurse-submodules https://github.com/waku-org/nwaku
cd nwaku
docker build -t wakuorg/nwaku:latest .

Step 2: Run

To run nwaku in a new Docker container, use the following command:

docker run [OPTIONS] IMAGE [ARG...]

where OPTIONS are your selected Docker options, IMAGE the image and tag you pulled from the registry or built in Step 1 and ARG... the list of nwaku arguments for your chosen nwaku configuration.

For Docker options we recommend explicit port mappings (-p) at least for your exposed libp2p listening ports and any discovery ports (e.g. the Waku discv5 port) that must be reachable from outside the host.

As an example, consider the following command to run nwaku in a Docker container with the most typical configuration:

docker run -i -t -p 60000:60000 -p 9000:9000/udp wakuorg/nwaku:v0.20.0 \
  --dns-discovery:true \
  --dns-discovery-url:enrtree://ANEDLO25QVUGJOUTQFRYKWX6P4Z4GKVESBMHML7DZ6YK4LGS5FC5O@prod.wakuv2.nodes.status.im \
  --discv5-discovery \
  --nat:extip:[yourpublicip] # or, if you are behind a nat: --nat=any

This runs nwaku in a new container from the wakuorg/nwaku:v0.20.0 image, connects to wakuv2.prod as bootstrap fleet and enables Waku Discovery v5 for ambient peer discovery, while mapping the default libp2p listening port (60000) and default discv5 UDP port (9000) to the host.

Tip: The docker run command will pull the specified image from Docker Hub if it's not yet available locally, so it's possible to skip Step 1 and pull the image from your configured registry automatically when running.

If you've used the -i and -t Docker options when running the new container, the run command would have allocated an interactive terminal where you'll see the stdout logs from the running nwaku process. To detach gracefully from the running container, use Ctrl-P followed by Ctrl-Q.