Ansible Role: Ethereum

slingnode.ethereum is an Ansible role that facilitates deployment of Ethereum clients. Its objective is to enable a consistent way of deploying and upgrading the chosen client mix. The role can be used to deploy:

• a single node running execution, consensus and validator layers
• hundreds of nodes running all three layers
• a distributed set up with each layer running on a separate server on a single or hundreds of servers
• change client mix (as seamlessly as this can be done)
• upgrade clients

Supported Ethereum clients

Execution clients:

• Geth
• Nethermind
• Besu
• Erigon

Consensus clients:

• Lighthouse
• Prysm
• Teku
• Nimbus (checkpoint sync currently not supported)

Validator clients:

• Lighthouse
• Prysm
• Teku
• Nimbus

Supported Operating Systems

The role has been tested on the following operating systems

RedHat based:

• AlmaLinux 9.1
• AlmaLinux 8.7
• RockyLinux 9.1
• RockyLinux 8.7
• AmazonLinux 2

Debian based:

• Ubuntu 22.04
• Ubuntu 20.04

Documentation

The README file provides a basic overview only. Full documentation describing the role in details is available at https://docs.slingnode.com/slingnode.ethereum/.

Requirements

Ansible Docker module is required on the Ansible controller. It can be installed using the below command:

ansible-galaxy collection install community.docker:==3.8.0

Dependencies

The role requires a running docker daemon and docker compose plugin installed on the target server. The role has been tested with "geerlingguy.docker" role. Refer to examples repository for sample Playbooks.

Role Variables

Role variables are defined in 'defaults'. This means they have the lowest precedence and can be easily overridden. See Ansible documentation for details on the precedence.

All client specific variables are defined in their corresponding variable file. All client specific variables have unique names (prefixed with clientname_) so there's no risk of a clash.

Variable location	Purpose
defaults/main/main.yml	generic variables for the role; common variables for all clients
defaults/main/{{client}}.yml	variables specific to the client
vars/{{os_family}}.yml	variables specific to the OS

Important variables

This section outlines variables that you will most likely want to modify.

clients

"clients" variable defines what clients will be deployed. It defaults to geth and lighthouse. Possible values are:

• execution: geth / nethermind / erigon / besu
• consensus and validator: lighthouse, prysm, teku, nimbus

clients:
  execution: geth
  consensus: lighthouse
  validator: lighthouse

deploy_execution / deploy_consensus / deploy_validator

These three variables define which layer of the stack will be deployed. By default all three layers are deployed. Check out examples to see how they can be used.

deploy_execution: true
deploy_consensus: true
deploy_validator: true

enable_firewall

By default set to true, the role will make sure the firewall package is installed and will configure rules for the client that is being deployed.

enable_firewall: true

This is meant to be a safe default for anyone who uses the role to deploy on a server that is directly exposed to the internet (for example an OVH bare metal server). It is expected that for more enterprise type of deployments this would be set to false. Refer to [Host firewall]https://docs.slingnode.com/slingnode.ethereum/architecture/security#host-firewall) section for details and notes on SSH port.

network

Specifies Ethereum network to connect to.

network: goerli

consensus_checkpoint_sync_enabled

Enable/disable consensus client checkpoint sync feature. Defaults to "true" (sync from checkpoint).

consensus_checkpoint_sync_enabled: true

Refer to Checkpoint sync section for details.

consensus_checkpoint_sync_url

Beacon endpoint to sync from. This needs to match the network you are deploying on. Defaults to Goerli.

consensus_checkpoint_sync_url: https://sync-goerli.beaconcha.in

Refer to Checkpoint sync section for details.

suggested_fee_recipient

A priority fee recipient address on your validator client instance and consensus node.

suggested_fee_recipient: "0xa10214731A6D9eC03d36d1437796D1cEe6a061f7"

graffiti

graffiti: SlingNode.com

Example playbook

The best place to start is to check the examples project. There are multiple example playbooks ranging from a simple single node deployment to a distributed deployment of a large number of nodes. The examples project includes sample playbooks, inventories and group vars that you can adapt to your needs. The examples project is available here https://github.com/SlingNode/slingnode-ethereum-examples

Sample Playbook:

---
- name: Deploy ethereum stack - whole stack per node - default clients (geth & lighthouse)
  hosts: all
  become: true

  vars:
    suggested_fee_recipient: "0xa10214731A6D9eC03d36d1437796D1cEe6a061f7"
    graffiti: SlingNode.com


  roles:
    - role: geerlingguy.docker
      vars:
        docker_package_state: present
        docker_install_compose_plugin: true

    - role: slingnode.ethereum

Contact

If you have any questions join our Discord server or log a GitHub issue.

License

MIT

Author Information

This role was created in 2023 by pgjavier and karolpivo.