Hard Reset Guide

Reset your node to a clean state using the latest chain snapshot for fast recovery and resync.

A hard reset wipes all local node data and restores the most recent network snapshot. This is the fastest and most reliable way to recover a degraded, stuck, or outdated node.

After a hard reset:

1. A fresh snapshot is downloaded.

2. The TrieDB database is rebuilt (1–5 minutes on mainnet).

3. The node catches up using statesync and blocksync (2–5 minutes for a recent snapshot).

1. Prerequisites

Ensure aria2 is installed:

apt install -y aria2

---

2. Stop Services & Reset Workspace

SSH into the server as root and run:

bash /opt/monad/scripts/reset-workspace.sh

This clears:

• Ledger data

• Forkpoint history

• Validator sets

• TrieDB state

• Runtime artifacts

---

3. Download & Import the Latest TrieDB Snapshot

Choose one provider (Monad Foundation or Category Labs).

Monad Foundation snapshot:

MF_BUCKET=https://bucket.monadinfra.com
curl -sSL $MF_BUCKET/scripts/mainnet/restore-from-snapshot.sh | bash

Category Labs snapshot:

CL_BUCKET=https://pub-b0d0d7272c994851b4c8af22a766f571.r2.dev
curl -sSL $CL_BUCKET/scripts/mainnet/restore_from_snapshot.sh | bash

---

4. Fetch Latest Forkpoint & Validators Files

(Skip if your node already uses automatic remote config fetching via REMOTE_VALIDATORS_URL and REMOTE_FORKPOINT_URL.)

Manual fetch:

MF_BUCKET=https://bucket.monadinfra.com
VALIDATORS_FILE=/home/monad/monad-bft/config/validators/validators.toml

curl -sSL $MF_BUCKET/scripts/mainnet/download-forkpoint.sh | bash
curl $MF_BUCKET/validators/mainnet/validators.toml -o $VALIDATORS_FILE
chown monad:monad $VALIDATORS_FILE

---

5. Start All Monad Services

systemctl start monad-bft monad-execution monad-rpc

Check logs to confirm syncing:

journalctl -u monad-bft -f
journalctl -u monad-execution -f

When statesync completes, the node will automatically switch to blocksync and reach chain tip.