Prometheus Container Metrics #2
Description
Has providing a metrics endpoint for airflow components other than the webserver been considered? So on the scheduler or celery workers? In that way, issues with individual components (like a stalled task on a worker) This would also be in line with adding a /health endpoint for ease of integration with Mesos or k8s liveness and readiness probes? (https://issues.apache.org/jira/browse/AIRFLOW-1084?jql=project%20%3D%20AIRFLOW%20AND%20text%20~%20health)