maorojloo commented on issue #56664:
URL: https://github.com/apache/airflow/issues/56664#issuecomment-5225379089

   Hi maintainers,
   
   I would like to work on the StatsD `ServiceMonitor` part of this issue.
   
   I am currently running the official Airflow Helm chart with StatsD enabled. 
The chart exposes Prometheus metrics through the existing StatsD service and 
its `statsd-scrape` port, but Prometheus Operator users must maintain a 
separate `ServiceMonitor`.
   
   After reviewing the current Helm chart development guidelines, I understand 
that resources requiring externally managed CRDs should normally be implemented 
as Kustomize overlays rather than native chart templates.
   
   I propose the following scope:
   
   * Add an optional Prometheus Operator Kustomize overlay for the existing 
StatsD service.
   * Include a `ServiceMonitor` that selects the StatsD service using its 
existing labels and named scrape port.
   * Keep the base Helm chart and its defaults unchanged.
   * Do not install or manage Prometheus Operator CRDs.
   * Add validation/tests and usage documentation, including namespace and 
selector configuration.
   * Document the required Prometheus Operator prerequisite.
   
   Could you please confirm whether this is the preferred scope before I start 
implementing it? If native Helm chart support is preferred despite the current 
chart-routing guidelines, I can adjust the implementation accordingly.
   
   I can test this against:
   
   * Airflow Helm chart: 1.22.0
   * Kubernetes: 1.30.4
   * Prometheus Operator/kube-prometheus-stack: v0.82.2 / 72.9.0
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to