Use StrongMemoize for memoizing state of Prometheus metrics

Stan Hurequested to merge
sh-memoize-prometheus-metrics-enabled into master

What does this MR do and why?

If Prometheus metrics were disabled in the application settings, previously the prometheus_metrics_enabled? check would not be memoized properly. As a result, we could get into a situation where a database query is needed to check the state of the Prometheus metrics while the load balancing connections are being set up.

Normally the state of Prometheus metrics should be set up properly by the Puma master worker, but if it is set to false the database will repeatedly be checked.

To properly memoize false values, use strong_memoize_attr and clear_memoization.

References

Relates to https://gitlab.com/gitlab-com/request-for-help/-/issues/3317

MR acceptance checklist

Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.

Edited by Stan Hu

Merge request reports