Granian-native Prometheus metrics

[oduludo]

First of all, thanks for creating and maintaining Granian. It's a joy to work with so far!

I am moving from uWSGI to Granian. Things run fine, but one thing I seem to miss so far are metrics. I don't read anything about Granian-native metrics. Do those exist?

My Django app runs in a Kubernetes cluster. When running over uWSGI, I would export uWSGI's native/system metrics as Prometheus metrics (via a sidecar container) and use that data to run KEDA autoscaler queries. This allows me to scale pods up when many workers are busy. I'd like to do something similar with Granian. What are the current possibilities?

To be clear: I am not talking about Prometheus metrics in my own Django application. I am talking about specific Granian metrics that tell me things like its memory usage, amount workers and/or threads being busy, maybe the number of requests that are being handled per worker etc.

Thanks in advance. I could not find similar discussions, but please say so if this is a duplicate :)

[gi0baro]

See #610

[oduludo]

Thanks! I only checked the discussions, not the issues 🙈 my bad