/lighthouse-4844

setup for running lighthouse on 4844 testnets

Primary LanguageShell

lighthouse-4844

A docker-compose setup for running lighthouse with different execution clients on public 4844 testnets.

Features

  • All public 4844 testnets are supported
  • Major execution clients supported: geth, nethermind, besu, reth (erigon coming soon)
  • All data is stored on host disk (no annoying docker volumes to deal with)
  • Local directories are bind-mounted with the same permissions as the host user by default (no annoying permissions issues)
  • Optional proxy between beacon and execution node for debugging
  • Optional metrics gathering via prometheus and viewing with grafana
  • Lots of easy configuration options set via environment variables

Quick Start Guide

Clone this repository along with submodules for latest testnets:

git clone --recurse-submodules https://github.com/ethDreamer/lighthouse-4844.git

For best results, update to the latest version of 4844-testnet:

cd lighthouse-4844 && git submodule update --remote

Set Essential Variables in globals.sh

The essential variables that must be set correctly before starting the containers are:

  • ETH_TESTNET # the name of the testnet to run (see directories in shared/4844-testnet/network-configs)
  • EXECUTION_NODE # pick between geth, nethermind, besu, or reth
  • CONSENSUS_DISC # the discovery port (TCP/UDP) for lighthouse (should be accessible from internet)
  • EXECUTION_DISC # the discovery port (TCP/UDP) for execution node (should be accessible from internet)

After editing these be sure to run:

source ./globals.sh

Also be sure to generate a jwt token:

./gen-jwt.sh

Start the nodes

To enable only the necessary services (consensus/execution nodes), simply run:

docker compose -f beacon.yml -f execution.yml up -d

The yaml files are split up so that services can be swapped in as needed. There are two optional additional services:

  • metrics.yml # Enables prometheus metrics collection & grafana server
  • proxy.yml # A proxy between the beacon and execution node for debugging

These can be enabled by just adding them into the docker compose command. For example the proxy service can be enabled with:

docker compose -f beacon.yml -f execution.yml -f proxy.yml up -d

Stopping All Services

Services can be stopped in a similar fashion:

docker compose -f beacon.yml -f execution.yml [-f OPTIONAL SERVICES] down

Viewing Node Logs

To view the logs for each service on a continuous basis, simply run docker logs -f [CONTAINER NAME].

Node Software Container Name
lighthouse ${CONTAINER_PREFIX}_beacon
geth ${CONTAINER_PREFIX}_execution
nethermind ${CONTAINER_PREFIX}_execution
besu ${CONTAINER_PREFIX}_execution
reth ${CONTAINER_PREFIX}_execution
proxy ${CONTAINER_PREFIX}_proxy
prometheus ${CONTAINER_PREFIX}_prometheus
grafana ${CONTAINER_PREFIX}_grafana

Modifying Node CLI Arguments & Wiping the Data Directories

This setup was designed around making this as easy as possible for rapid development. Each service has a corresponding run directory on the local disk which contains an execute.sh script controlling how the node is launched. All data is stored inside run/datadir for easy removal. The run directory is bind mounted in the docker container (by default with the same permissions as the local user) so there are no docker volumes to wipe and no images that need to be rebuilt.

Node Software Data Directory Run Script
lighthouse lighthouse/beacon/run/datadir lighthouse/beacon/run/execute.sh
geth geth/run/datadir geth/run/execute.sh
nethermind nethermind/run/datadir nethermind/run/execute.sh
besu besu/run/datadir besu/run/execute.sh
reth reth/run/datadir reth/run/execute.sh

For example, if the services are already running, and the user wishes to change the CLI options for lighthouse and restart with a fresh data directory, they would modify lighthouse/beacon/run/execute.sh accordingly and then run:

docker stop ${CONTAINER_PREFIX}_beacon && rm -rf lighthouse/beacon/run/datadir && docker start ${CONTAINER_PREFIX}_beacon

Advanced Configuration

There are additional environment variables in globals.sh which can be modified as needed. Changing any of the variables listed in this section requires the docker images to be rebuilt. There is a convenience script to remove all existing docker images (forcing a rebuild):

./remove-images.sh

Users can change the base docker image for any of the containers by setting the appropriate environment variables in globals.sh:

  • LIGHTHOUSE_IMAGE
  • GETH_IMAGE
  • BESU_IMAGE
  • NETHERMIND_IMAGE
  • RETH_IMAGE

By default, the processes will be run with the same credentials as the user that sources globals.sh. These credentials can be controlled individually by setting the following variables in globals.sh:

  • BEACON_UID
  • BEACON_GID
  • EXECUTION_UID
  • EXECUTION_GID
  • METRICS_UID
  • METRICS_GID

The names of the images & containers generated by docker-compose can also be changed using the variables:

  • IMAGE_PREFIX
  • CONTAINER_PREFIX

This is useful when running multiple instances of this compose setup on the same machine.

Again be sure to run source ./globals.sh after modifying any of these variables.