[Request] Performance metrics

[coetzeer]

Very cool project!

It's super quiet though - it would be nice to expose some metrics on how it's performing e.g.

• latency and callcount to the upstream peertube instance (including errors + throttling)
• when new shows are detected
• how long it takes to spin up the headless firefox
• show duration
• any relevant failures e.g. failures in starting firefox

There are probably others I'm forgetting.

If we decide to to the prometheus route, we'd probably need to add flask (or other implementation) endpoint to serve the metrics. I've not built a promethues metrics endpoint before but I could probably try my hand at writing it.

What do other people think?

[Strit]

It's an awesome and cool idea. Would be awesome to have some kind of metric to see what's going on.

[tyrsarm]

Before I put the code on GitHub, I had some simple statements on what the scripts were doing for debugging. (Such as "No New Videos", "Video is Live", and "Polling API for Live Status" in the Python console output.) I took some of them out because I thought they might clutter the code base. I can put the statements back into the scripts if you all think it will be beneficial. We can also do the Prometheus route for more in-depth metrics.