Erigon

Installing Erigon for Archive Nodes or disk-efficient storage.

Erigon

Erigon is an implementation of Ethereum focused on storage efficiency and speed. It uses a different database architecture (MDBX) that allows it to store a full Archive Node (all historical states) in a fraction of the space required by Geth.

---

🛡️ Role: Erigon is the specialist's choice. It is primarily used by indexers, block explorers, and analytics providers who need deep historical data.

Note: Erigon syncs differently. It downloads all block headers first, then block bodies, and finally executes them. It may appear "stuck" at 100% header sync for a long time—this is normal.

⚠️ Erigon 3: This guide covers Erigon v3, which is a major rewrite. Erigon v2 is no longer supported and does not support post-Pectra Ethereum. If upgrading from v2, a full re-sync with an empty --datadir is required — v2 data is not compatible with v3.

---

1. Install Dependencies

Erigon is distributed as a single binary. We only need basic tools to download and extract it.

# Update repositories
sudo apt update

# Install utilities
sudo apt install -y curl tar

---

2. Download & Install

We will download the latest pre-built binary from the official repository.

A. Create the Application Directory

# Create a folder for the binary
sudo mkdir -p /usr/local/lib/erigon

# Define the version (Check https://github.com/erigontech/erigon/releases for the latest)
ERIGON_VERSION="3.2.2"

# Download the specific release
sudo wget https://github.com/erigontech/erigon/releases/download/v${ERIGON_VERSION}/erigon_${ERIGON_VERSION}_linux_amd64.tar.gz -O erigon.tar.gz

# Extract the archive
sudo tar -xvzf erigon.tar.gz -C /usr/local/lib/erigon

# Cleanup
rm erigon.tar.gz

B. Set Permissions

Ensure the ethereum user owns the directory and the binary is executable.

sudo chown -R ethereum:ethereum /usr/local/lib/erigon
sudo chmod +x /usr/local/lib/erigon/erigon

---

3. Configure Systemd Service

We will configure Erigon to run as a background service.

💾 Storage Mode:

• Archive Node: Retains all historical data. Add --prune.mode=archive.

• Full Node (Default in v3): Prunes old data to save space. Add --prune.mode=full.

Note: Erigon v3 defaults to Full Node mode, unlike v2 which defaulted to Archive. The prune mode cannot be changed after the first start.

💡 Built-in Consensus Layer (Caplin):

Erigon v3 ships with Caplin, an embedded consensus layer client. By default, Caplin is enabled, which means you do not need a separate Consensus Client (Step 4) when running Erigon. If you prefer to use an external Consensus Client instead, add the --externalcl flag to disable Caplin.

Create the service file:

sudo nano /etc/systemd/system/erigon.service

Paste the following configuration (using Caplin — no external CL needed):

[Unit]
Description=Erigon Execution Client
After=network.target

[Service]
User=ethereum
Group=ethereum
Type=simple
Restart=always
RestartSec=5

# Execution Command
ExecStart=/usr/local/lib/erigon/erigon \
  --chain=mainnet \
  --datadir=/home/ethereum/execution \
  --prune.mode=archive \
  --authrpc.jwtsecret=/home/ethereum/jwt/jwt.hex \
  --authrpc.addr=127.0.0.1 \
  --authrpc.port=8551 \
  --authrpc.vhosts=localhost \
  --http \
  --http.addr=0.0.0.0 \
  --http.port=8545 \
  --http.api=eth,erigon,web3,net,debug,trace,txpool \
  --http.corsdomain="*"

# Resource Limits
LimitNOFILE=65535

[Install]
WantedBy=multi-user.target

🔍 Flag Explanation:

• --prune.mode=archive: Tells Erigon to keep all historical state. (Change to full if you have <2TB storage).

• --http.api: Notice we added erigon and trace namespaces. These are powerful tools unique to Erigon for debugging transactions.

• --authrpc.jwtsecret: The standard Engine API authentication.

---

4. Start the Service

Reload the systemd daemon and start the node.

# Reload systemd configuration
sudo systemctl daemon-reload

# Enable Erigon to start on boot
sudo systemctl enable erigon

# Start the service
sudo systemctl start erigon

---

5. Verify Status

Check the logs to track the unique Erigon sync stages.

sudo journalctl -fu erigon

Expected Output (The Stages): Erigon does not sync like Geth. You will see it move through strict "Stages":

• Stage 1: Downloading headers (Fast)

• Stage 2: Downloading bodies (Fast)

• Stage 3: Executing blocks (Slow — This is where it rebuilds the state)

• Caplin: You should also see Caplin logs indicating the embedded consensus layer is running and syncing beacon chain data.