Metrics

By default Prometheus metrics are enabled for all clients. The metrics are exposed to the Docker network and accessible only from other containers - refer to Exposing ports page for details on exposing them to the host or to the network.

The metrics can be disabled by overriding the following variables and setting them to false:

execution_client_metrics_enabled: true
consensus_client_metrics_enabled: true
validator_client_metrics_enabled: true

The ports the metrics are exposed on can be modified using the following variables:

execution_client_metrics_port: 6060
consensus_client_metrics_port: 8008
validator_client_metrics_port: 8009

Geth expensive metrics

Be default Geth's expensive metrics are enabled. They can be disabled by overriding the following variable and setting it to false:

geth_metrics_expensive_enabled: true

slingnode.ethereum_observability

SlingNode has developed an Ansible role that can deploy an Observability stack that seamlessly integrates with nodes deployed using slingnode.ethereum role. The role deploys Prometheus that will automatically start scraping the metrics.