Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.novacula.io/llms.txt

Use this file to discover all available pages before exploring further.

This guide takes you from zero to a running blockchain full node serving RPC. By the end you will have an active executor and your first node syncing, ready to expose :8545 (or the per-chain equivalent) to your own reverse proxy.

​
Prerequisites

  • An email address you can receive messages on.
  • One of:
    • A Linux host (Ubuntu 22.04+ or Debian 12+) with sudo access — for the Agent deployment.
    • A Kubernetes cluster (1.27+) with cluster-admin access — for the Operator deployment.
You do not need to expose any inbound ports. The executor connects outbound to the control plane.

​
Steps

1

Create an account

Sign up at the Novacula console. A personal organization is created for you on completion of sign-up, and you become its Owner.
2

Generate an executor API key

From the organization sidebar, open Executors → Connect new executor. Pick the backend that matches the host you’ll install on:
  • agent — for bare-metal or VM targets. The Agent runs as a systemd service on a Linux host.
  • operator — for Kubernetes targets. The Operator runs as a controller inside the cluster and reconciles nodes via CRDs.
Give the executor a name and copy the displayed API key.
The API key is shown only once. Store it somewhere you can reference during install — a password manager, a secret store, or your provisioning system.
3

Install the executor

The console shows a copy-pasteable install command for the backend you chose:
  • Agent — a one-line installer that downloads the binary, writes a systemd unit, and starts the service.
  • Operator — a helm install command that deploys the controller and applies the CRD.
On first start, the executor authenticates with the API key, registers itself, and declares its capabilities — the chains, clients, and versions it can run. See Executor capabilities.For backend-specific setup details, see Provision on Kubernetes or Provision on bare-metal.
4

Deploy a node

Open Nodes → Deploy node. Fill out the wizard:
  • Target — pick the executor you just connected, then a chain, network, client, and version.
  • Resources — accept the per-network defaults, or override CPU / RAM / disk.
  • Custom config — leave empty for the first run; you can add overrides later.
  • Review — the Advanced section lets you pick how RPC ports are published. On the Operator backend, Service type (ClusterIP / NodePort / LoadBalancer) controls whether the endpoint is reachable only inside the cluster or also externally. See Deploy a node — Review for the full breakdown.
Confirm the review screen. The executor picks up the spec on its next poll (typically within seconds).
5

Watch it sync

The node detail page shows live status from the executor. Logs stream in as the underlying process boots, and sync progress appears once the client starts processing blocks.

​
Next steps