Chainflip Docker

Chainflip node in Docker Compose

Getting started

./flipd install can be used to install docker-ce and Docker Compose, if they aren't already installed

cp default.env .env and adjust variables such as the version tags, HOST_IP and ETH_RPC as well as ETH_WS URLs, as well as the ARB_RPC and ARB_WS variables. For SOL_RPC, use an RPC provider.

Generate keys

Read this alongside the official Chainflip docs

Make sure the keys directory is owned by uid 1000. This will already be the default if you have just one user in Debian/Ubuntu: sudo chown 1000:1000 ./keys

Generate the keys with ./flipd cmd run --rm cli generate-keys --path /etc/chainflip/keys.

Make sure to back up your Seed Phrase and make a note of the public keys and account ID. You will need the Seed Phrase if you ever need to restore your node or recover your funds if you lose access to the node. DO NOT LOSE THIS.

Back up the ./keys directory, as another failsafe alongside the mnemonic.

Take special note of the Validator Account ID beginning with cF. This is the ID that you will need to add funds and track your node.

NEVER REVEAL YOUR PRIVATE KEYS TO ANYONE.

Sync node

Start the Chainflip node with ./flipd up. Check logs of the node with ./flipd logs -f node. The engine service will print "wait" messages until node is synced.

Funding and bidding

Read this alongside the official Chainflip docs.

Make sure node is synced before proceeding.

Stake tFLIP to the node exactly as described in the official docs.

Restart the engine: ./flipd restart engine

Register your account for the validator role: ./flipd cmd run --rm cli register-account-role Validator

As long as you saw a successful tx, next rotate: ./flipd cmd run --rm cli rotate

Check logs of node and engine, make sure you're synced and everything is working well, check status is green in the web app, then start bidding: ./flipd cmd run --rm cli start-bidding

Optionally, set a vanity name: ./flipd cmd run --rm cli vanity-name <my-discord-username>

Get metrics

You can either add :grafana-cloud.yml to COMPOSE_FILE in .env, and send metrics to Grafana Cloud or, indeed, your own Mimir or Thanos. Adjust prometheus/custom-prom.yml with the desired remote write.

Or you can run central-proxy-docker in a directory such as traefik, configure it for remote write with its prometheus.yml, and also promtail.yml as desired, and have the chainflip metrics be scraped centrally. Use ext-network.yml to connect the chainflip stack to the central traefik/prometheus/promtail stack.

Updating Chainflip

If you are using specific version tags for node and engine, instead of latest, nano .env and set the desired new version tags.

./flipd update to pull in the new versions.

./flipd up to restart node and engine on the new versions.

Restoring from backup

After you restore the keys directory from backup and sync the node and engine again, run ./flipd cmd run --rm cli rotate. This is necessary after restore.

This is Chainflip Docker v4.0.0