Queues and Metrics for Side Project

I’ve worked with basically everything both from my company (managed by ops team) and self hosted cluster for fun and no real traffic, so kind of both ends.

What I’m wondering is the approach y’all use for metrics, alerting and queues for hobby projects with real users. Curious on the scale and what hardware you use.

I’m kind of leaning toward Postgres everything, but open to some convincing.