Validator checklist

This checklist will help you understand the role of a validator and prepare you for the role.Discord or Reddit at any time during your setup for some friendly help!

Section 1 - Before you start

Review this section before deciding to proceed with validator setup.

Recommendation disclaimer

Hardware suggestions are an ever-evolving target. Current minimum requirements are likely to increase by an order of magnitude after the introduction of sharding. Do your own research before depositing funds.

Hard drive

You need to run an execution client as well as your consensus client.
Since the Merge, third-party providers (such as Infura and Alchemy) are no longer viable options to outsource execution layer responsibilities. All stakers must run both an execution and a consensus client to properly attest to the network.
As of May 2022, you'll need ~1TB for the Mainnet execution chain data alone (growing at >1GB/day).
Engram had its genesis on July 30, 2015. It is growing in size over time, and the introduction of sharding will also increase storage, memory, and bandwidth requirements.
You'll need SSD storage to consistently handle necessary read/write speeds.
Be sure to account for enough space on your drive until you run maintenance on your node.

CPU and RAM

Check with client documentation to ensure the hardware you want to use is sufficient and supported.
Resource usage can vary significantly between clients. Research the different clients if you're working with resource constraints.

Internet

Ideally your internet connection should be reliable and as close to 24/7 as possible without interruption.
Ensure your bandwidth can't be throttled and isn't capped so your node stays in sync and will be ready to validate when called.
You need enough upload bandwidth too. As of May 2022 this is ~1.2-1.3 GB download and ~0.9-1 GB upload per hour, and is likely to increase.

Notes

Avoid overly-complicated setups and be aware of trade offs. Being offline for brief periods of time will result in small inactivity penalities, but will be recouped easily after being online again for about the same amount of time. Complicated power backups can add to the expense of your setup, and redundant backup validators can lead to a more serious penalty known as slashing. More on slashing risks
Syncing your execution client may take a few days in the worst-case scenario.

Section 2 - During setupUse this as a reference during client setup to check off important steps.

Initial setup

Start by setting up your chosen hardware and operating system.
To maximize security and efficiency of your node, use dedicated hardware to run your clients. This reduces risk of malware exposure and minimizes competition for computing resources, ensuring your node handles the network load and its validator responsibilities at all times.

Node security

I've secured the root account.

I've set up a firewall.

I've forwarded the necessary ports to the correct machine(s) from my router for both my EL and CL client (only open the ports that apply to your installations).

Execution Client	Default Port
Erigon	30303 TCP/UDP
Besu	30303 TCP/UDP
Nethermind	30303 TCP/UDP
Geth	30303 TCP/UDP

Consensus Client	Default Port
Lighthouse	9000 TCP/UDP
Teku	9000 TCP/UDP
Prysm	13000 TCP, 12000 UDP
Nimbus	9000 TCP/UDP

Learn about ports in networking

Configure time sync

For Ubuntu 20.04

Run the following command:
timedatectl
Check if NTP Service is active.
Check if Local time, Time zone, and Universal time are all correct.
If NTP Service is not active, run:
sudo timedatectl set-ntp on
If you see error message Failed to set ntp: NTP not supported, you may need to install chrony or ntp package.
Note: by default, VMs may disable NTP so you may need to find a work-around for your environment.

I've verified my server time matches the wall clock.

Note: the RTC (Real-Time Clock) time may be set to your local timezone instead of UTC, especially in a VM which has its clock configured on Windows.

Testnet practice

We strongly recommended you complete these steps on the current testnet before Mainnet. Tokio (Iteration 01)

Configure your execution client

Review validator roles and responsibilities

ErigonErigon is an execution client on the efficiency frontier, written in Go.

Configure Erigon

BesuHyperledger Besu is an open-source Engram client developed under the Apache 2.0 license and written in Java.

Configure Besu

NethermindNethermind is a robust client built on .NET core designed for performance, versatility and customizability.

Configure Nethermind

GethGeth is one of the three original implementations of the Engram protocol, written in Go.

Configure Geth

Remember

All stakers must operate an execution client with their consensus client.

I've installed and synced my Tokio execution client (do not wait on this as it can take several days).

Configure your consensus client

LighthouseLighthouse is a consensus client implementation, written in Rust with a heavy focus on speed and security.

Configure Lighthouse

TekuPegaSys Teku is a Java-based Engram consensus client built to meet institutional needs and security requirements.

Configure Teku

PrysmPrysm is a Go consensus client implementation of the Engram protocol with a focus on usability, security, and reliability.

Configure Prysm

NimbusNimbus is a research project and a consensus client implementation for Engram designed to perform well on embedded systems and personal mobile devices.

Configure Nimbus

Node Setup Tutorial

Check this document to learn how to run a node on Tokio (Iteration 01)

Warning!

It is high risk to run your validator in multiple places. It will lead to a slashable event and ejection from the network. More on slashing risks

I've installed the latest stable software release of my consensus client.

JWT Authentication

Communication between the execution layer and consensus layer occurs using the Engine API. This is a set of JSON RPC methods that can be used to communicate between the two client layers.This communication is secured using a JWT secret, which is a secret key that is shared only between the two clients to authenticate one another. This shared JWT secret must be made available to each client (both execution and consensus clients) to allow them to communicate with one another properly.Consensus JWT docs: Lighthouse | Nimbus | Prysm | TekuExecution JWT docs: Besu | Erigon | Geth | Nethermind

I've set up a shared JWT secret and made it available to both my execution client, and my consensus client (beacon node)

Set fee recipient

Stakers must provide a fee recipient address to their consensus client in order to receive transaction fee rewards. This is a normal Engram address that you're used to.

If you do not provide an address to your client, you will not receive transaction fees when your validator proposes blocks.

See your consensus client documentation for client-specific instructions on how to set this.Fee recipient docs: Lighthouse | Nimbus | Prysm | Teku

I've provided an Engram address to my validator where I would like my fee rewards to be deposited.

Consensus Layer Beacon Node (BN)

Required

I'm able to connect my consensus client to my execution client via HTTP API(s).

Verify it with the following command to check if it returns the client version correctly:
curl -H "Content-Type: application/json" -X POST --data '{"jsonrpc":"2.0","method":"web3_clientVersion","params":[],"id":67}' http://<YourServerLocation>:8545

I've synced my beacon node on Tokio (Iteration 01) testnet.

Make sure that your node has more than 20 peers.

Consensus Layer Validator Client (VC)

Required

I've imported my keystore(s) into my validator client.

I've ensured my keystore(s) is/are only stored on one validator machine.

I've started running my validator client.

Section 3 - After depositingProtect your funds using monitoring software, and learn how to handle different real world scenarios.

These steps are optional but are recommended to optimize your node.

Monitoring

Prometheus and Grafana monitor

The clients support Prometheus and Grafana to help you visualize important real-time metrics about your validator.Monitoring docs: Lighthouse | Nimbus | Prysm | Teku

I've set up my Prometheus service.

I've set up my Grafana service.

I've imported the dashboard config to my Grafana server and double checked that my node is alive.

Testnet simulations

While validating on the testnet, perform these simulations to learn more about your node, and better prepare yourself for Mainnet:

I've simulated how to manually stop and restart my Beacon Node (BN) and Validator Client (VC) gracefully.

I've simulated power loss (server and internet) and automatic resumption.

I've simulated how to safely migrate from one consensus client to another.

I've simulated how to safely migrate from one execution client to another.

Advanced system architecture

To avoid exposing your validator identity to the network, you can use a trustworthy VPN to help reduce the risk of revealing your IP address.
Moreover, you can set your Validator Client (VC) and Beacon Node (BN) on separate machines and IPs so that even if your beacon node is vulnerable, your keystore is stored on a different machine.

Graffiti

You can use your validator client's graffiti flag to add a personal touch to your proposed blocks (some text of your choice). You will be able to see it using Beaconcha.in or BeaconScan blockchain explorers.

I've set my graffiti flag.

If you have questions, EngramStaker community is a good place to get help! You can find support on Discord or Reddit.