> ## Documentation Index
> Fetch the complete documentation index at: https://docs.morpheum.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Deploy a validator

> Set up a Morpheum validator: keys, private peering, running the node, and keeping consensus infrastructure safe

## Who this is for

Teams that **operate consensus** on the network: producing and voting on blocks and executing the canonical state transition. Validators are high-trust; compromise of validator keys is a network-level incident.

***

## Before you start

* **Hardware:** Plan for sustained CPU, fast storage, and reliable networking. Validators are not bursty batch jobs—they run continuously.
* **Network:** Prefer **private peering** only. Validators should not serve as public RPC endpoints.
* **Topology:** In production, place **sentry** nodes in front of validators so user and peer traffic hits sentries first. See [Deploy a sentry](/node/deploy-sentry).
* **Keys:** You will create a **validator identity** during initialization. Back it up securely (offline copy, hardware security module, or your organization’s key policy). Never commit keys to source control or share them in chat.

***

## Initialize the node

1. Install the Morpheum CLI for your platform.
2. Run **init** with role **validator**, choosing a data directory on durable storage, a **moniker**, and the **chain ID** for the network you join.
3. Complete any prompts for key generation unless you intentionally skip keys to import them separately (advanced).

Confirm that validator keys exist only on this host (or in your approved secret store) and that filesystem permissions follow your security baseline.

***

## Configuration

After init, review the generated configuration for:

* **Consensus** — Enabled for validators.
* **Peering** — Listen addresses and allowed peers aligned with **private** operation; avoid advertising a public RPC URL for this machine.
* **Shards** — Match the network you operate on; use defaults unless your operator docs say otherwise.

Edit configuration with care; invalid settings can prevent startup or cause missed participation.

***

## Run the node

Start the node with **run**, pointing at the same data directory and enabling **validator** mode so voting and consensus match your role.

Run the process under a **process manager** or **orchestrator** so it restarts on failure and rotates logs. Monitor health, disk space, and peer counts continuously.

***

## Security and operations

* **Rotate and audit** access to the host; validators are high-value targets.
* **Do not** expose validator ports to the open internet beyond what strict peering requires.
* **Backup** keys and critical configuration; test restore procedures.
* **Upgrade** in a controlled window; read release notes and coordinate with other validators on testnets first when possible.

***

## Related

* [Node CLI](/node/cli) — `init` and `run` options.
* [Deploy a sentry](/node/deploy-sentry) — Public-facing layer in front of validators.