A docker-compose is used to bundle prometheus and grafana services, the *overrides files define the ports and mounts the prometheus' configuration file. If a new dashboard is designed, it can be mounted only in that *overrides file.

To run the node with metrics, the next steps should be followed:

1. Build the ethrex binary with the metrics feature enabled.
2. Enable metrics by using the --metrics flag when starting the node.
3. Set the --metrics.port cli arg of the ethrex binary to match the port defined in metrics/provisioning/prometheus/prometheus*.yaml
4. Run the docker containers, example with the L2:

docker compose -f docker-compose-metrics.yaml -f docker-compose-metrics-l2.overrides.yaml up

• For the L2 we use the following files in conjunction:
  • docker-compose-metrics.yaml
  • docker-compose-metrics-l2.overrides.yaml
  • The defaults are:
    • PORT 3702 → metrics API (used by prometheus)
    • PORT 3802 → Grafana
      • usr: admin
      • pwd: admin
    • PORT 9092 → Prometheus

An extra overrides file is available that enable alerts about the chain status. The alerts notify via Slack when the L2 is not advancing, the mempool is increasing fast (probably meaning that transactions are not getting processed), and if the chain is not advancing in L1. To enable this feature, add the docker-compose-metrics-alerts.override.yaml file to the Docker Compose command. Also, the following environment variables must be set up:

• GRAFANA_SLACK_CHANNEL: The name of the channel (or user ID) where the will be sent.
• GRAFANA_SLACK_TOKEN: A Slack token with write permissions on the desired channel's chat. This token starts with xoxb-.