How to run a node on Goerli/Prater

Who is this guide for?

This guide is meant as a starting point for users with some experience running nodes and validators. It is not meant to be a beginners guide, chances are copying and running commands from here will not work, some tweaking might be needed. We see this as a guide for people making better guides :)

The mentioned better guides, especially for beginners can be found here:

What is being merged?

The EL chain Goerli and CL beaconchain Prater are going to be merged. The merged chain will be called Goerli, the name Prater is soft-deprecated (flags using the name will work, but not supported long term).

What’s new in Goerli?

The client behaviour in Goerli is identical to Ropsten. If your staking setup works on Ropsten, it will work on Goerli as well.

How do I know my node is properly configured?

We have a basic checklist that can be found here. The checklist provides a list of things to check once your nodes are seutp as well as a common list of pitfalls to avoid.

This is too complicated, Do you have an overview?

Yes we do, this FAQ might provide you with some context and what next steps you’d need to take.

Which version/branch do I use?

Consensus layer

client name git branch docker image
Lighthouse unstable sigp/lighthouse:v2.4.0
Teku master consensys/teku:22.7.0
Lodestar master chainsafe/lodestar:v0.41.0
Nimbus unstable statusim/nimbus-eth2:multiarch-v22.7.0
Prysm develop gcr.io/prysmaticlabs/prysm/beacon-chain:v2.1.4-rc.0

Execution layer

client name git branch docker image
Geth master ethereum/client-go:v1.10.21
Nethermind master nethermind/nethermind:1.13.5
Erigon alpha thorax/erigon:v2022.07.04
Besu main hyperledger/besu:22.7.0-RC3

Preparation

Run an Execution Layer Client

Geth

We need to download and install geth

Pre-requisites: git, make, go, gcc.

(use geth branch master from github.com/ethereum/go-ethereum)

git clone -b master https://github.com/ethereum/go-ethereum.git
cd go-ethereum 
make geth
cd ..

./go-ethereum/build/bin/geth 
     --goerli \
     --datadir "geth-datadir" \
     --http \
     --http.api="eth,web3,net" \
     --http.vhosts="*" --http.corsdomain "*" \
     --authrpc.jwtsecret=/opt/jwtsecret \
     --authrpc.vhosts="*" \
     console

If you do not specify a JWT secret, then geth will automatically generate one for you. You will then have to specify this secret in the Consensus Layer client.

Note: In case the CL does not specify a virtual host when making the request, it will be rejected by geth unless --authrpc.vhosts="*" has been set. Allowing wildcard virtual host matching on the authenticated port is not a security concern.

Nethermind

Clone and build the master branch of Nethermind

git clone --recursive https://github.com/NethermindEth/nethermind.git
cd nethermind/src/Nethermind
dotnet build Nethermind.sln -c Release

In the nethermind/src/Nethermind/ folder

cd Nethermind.Runner
dotnet run -c Release -- --config goerli \
--JsonRpc.Host=0.0.0.0 \
--JsonRpc.JwtSecretFile=/opt/jwtsecret

Erigon

git clone -b alpha https://github.com/ledgerwatch/erigon
cd erigon
make erigon rpcdaemon

./build/bin/erigon --chain=goerli \
--datadir goerli-testnet \
--authrpc.jwtsecret=/opt/jwtsecret \
--http --http.api=engine,net,eth

Besu

Support for ropsten is baked into the besu main branch. If you want to use pre-built binaries, you can get the current 22.4.4 besu snapshot artifact, or use the docker image hyperledger/besu:22.4.4.

Otherwise if you want to build from source, clone the repo and build besu:

git clone --recursive  https://github.com/hyperledger/besu.git
cd besu
./gradlew installDist # installs in build/install/besu
# Alternatively, using Docker: 
# ./gradlew distDocker # build and tags a docker image

To run besu as an execution client on goerli:

build/install/besu/bin/besu     \
  --network=goerli              \
  --rpc-http-enabled=true       \
  --rpc-http-host="0.0.0.0"     \
  --rpc-http-cors-origins="*"   \
  --sync-mode="X_SNAP"          \
  --data-storage-format="BONSAI"\
  --Xmerge-support=true         \
  --rpc-ws-host="0.0.0.0"       \
  --host-allowlist="*"          \
  --engine-rpc-enabled=true     \
  --engine-host-allowlist="*"   \
  --engine-jwt-enabled=true     \
  --engine-jwt-secret=/opt/jwtsecret

Run a Consensus Layer Client

How to sync a CL beacon node?

There are two ways to sync a CL beacon node. They are:

As counter-intuitive as it may be, Genesis syncing the beacon chain is not the most secure approach. You can find more information on the topic here.

To perform a Genesis Sync, No additional flag/work is required. This is the default behaviour on all clients.

To perform Checkpoint Sync, Please follow our guide and add the Checkpoint Sync flags in addition to the flags defined below. The guide can be found here.

Lighthouse

Minimum requirement: rustc 1.59.0

Clone and build from source (use branch unstable)

git clone -b unstable https://github.com/sigp/lighthouse.git
cd lighthouse
make
cd ..

Start the client

lighthouse \
          --spec mainnet \
          --network goerli \
          --debug-level info \
          beacon_node \
          --datadir ./testnet-lh1 \
          --eth1 \
          --http \
          --http-allow-sync-stalled \
          --metrics \
          --merge \
          --execution-endpoints http://127.0.0.1:8551 \
          --enr-udp-port=9000 \
          --enr-tcp-port=9000 \
          --discovery-port=9000 \
          --jwt-secrets="/opt/jwtsecret" \
          --suggested-fee-recipient=<enter-eth-address-here> \

Lodestar

Clone and build the client

git clone https://github.com/chainsafe/lodestar.git
cd lodestar
yarn install --ignore-optional
yarn run build
cd ..

Start the client

cd lodestar
./lodestar beacon --rootDir="../lodestar-beacondata" \
--network=goerli \
--eth1.enabled=true \
--execution.urls="http://127.0.0.1:8551" \
--network.connectToDiscv5Bootnodes \
--network.discv5.enabled=true \
--chain.defaultFeeRecipient=<Enter-eth-address-here> \
--jwt-secret="/opt/jwtsecret"

Teku

Ensure Java 11 or above is installed (Ubuntu: sudo apt install default-jre).

git clone https://github.com/ConsenSys/teku.git
cd teku
./gradlew installDist
cd ..

The teku executable is now available in ./teku/build/install/teku/bin/teku. You can also use the pre-built distribution or the consensys/teku:develop docker image.

Start the client

./teku/build/install/teku/bin/teku \
  --data-path "datadir-teku" \
  --network goerli \
  --ee-endpoint http://localhost:8551 \
  --ee-jwt-secret-file "/opt/jwtsecret" \
  --log-destination console \
  --validators-proposer-default-fee-recipient=<Enter-eth-address-here> \

Prysm

Please ensure you have the latest version of the configs! (Specified in the preparation step)

git clone -b develop https://github.com/prysmaticlabs/prysm.git
cd prysm
bazel build //beacon-chain:beacon-chain
cd ..

Start the client

cd prysm
bazel run //beacon-chain -- \
--goerli \
--datadir $db_path  \
--suggested-fee-recipient=<Enter-eth-address-here> \
--http-web3provider=http://localhost:8551  \
--jwt-secret=/opt/jwtsecret

Nimbus

Pre-requisites: git, make, gcc.

Clone and build from source (use branch unstable)

git clone --branch=unstable https://github.com/status-im/nimbus-eth2.git
cd nimbus-eth2
make update OVERRIDE=1
make nimbus_beacon_node
cd ..

Start the client

nimbus-eth2/build/nimbus_beacon_node \
    --network=goerli \
    --web3-url=http://127.0.0.1:8551 \
    --rest \
    --metrics \
    --jwt-secret="/opt/jwtsecret" \
    --suggested-fee-recipient=<Enter-eth-address-here>
Last changed by  
 
launchpad
95133

Read more

Withdrawals

In this document I will include all the methods to make full withdrawals with the different clients and also show some examples about how to do a BLS to Execution change with ethdo tool. General assumptions Instead of using a long public key in all the queries, I just refer to it as $validator. export validator="0xb37101f32ee0d54ca6958764d21bd416b397c0a199d5ff447b8cd41bba89d78da9d12622134cc8a3cafe30daad85cd46" We assume for full withdrawal the withdrawal credential of type 0x01. If you have an address that is 0x00, then proceed to the bottom of the doc where it discusses how to convert a BLS (0x00) address into an execution address (0x01) All of our beacon nodes use port 4000 for API communication.

Feb 5, 2024
How to run a node on the Holesky testnet?

Who is this guide for? This guide is meant to be a starting point and provides a mid-level summary of the required steps. It is not intended to be a beginner's guide, so copying and running commands from here may not work without some tweaking. The testnet is meant to mimic a chain that is post-merge. Therefore, the genesis state will be in a merged state. The chain will have to go through a the shapella fork at epoch 256. The JSON-RPC endpoint, faucet, and other tooling URLs can be found on https://holesky.ethpandaops.io/ Which version/branch do I use? Consensus layer

Sep 28, 2023
ETH Withdrawals FAQ

Who is this guide for? If you run Ethereum validators, you can start testing withdrawals and preparing for withdrawals to go live on mainnet. Look out for Goerli and Sepolia, as these will fork before the mainnet. Before that, there will be devnets and mainnet shadowforks. DO NOT ATTEMPT TO CHANGE YOUR BLS KEYS ON MAINNET until the client software is ready to accept your BLS key changes. Each client team will announce once they are ready. We will provide a testnet environment where the BLS change can be tested out without any risk! [toc]

Jul 26, 2023
How to run a node on the Zhejiang testnet?

Who is this guide for? This guide is meant to be a starting point and provides a mid-level summary of the required steps. It is not intended to be a beginner's guide, so copying and running commands from here may not work without some tweaking. The testnet is meant to mimic a chain that is post-merge. Therefore, the genesis state will be in a merged state. The chain will have to go through a fork on both EL and CL at the same time. This fork is called Shanghai on the EL side and Capella on the CL side. These two events will be triggered simultaneously. This is achieved by EL forking at a specific time rather than at a block number, and CL forking at a specific epoch. What is in the Zhejiang testnet? The Zhejiang testnet aims to test the following features:

Feb 6, 2023
Read more from launchpad
Published on HackMD