[FEATURE] Prometheus metrics

[lukasmalkmus]

I'm keeping this short: I'm pretty sure the project and its users can benefit from exporting prometheus metrics. Metrics are important in modern application deployments, especially if there are many moving parts.

I would suggest to export some basic metrics about the application itself, queues, jobs and the web server (latency, http status codes, etc.). They are also interesting for scaling Kubernetes deplyoments: See #19.

If this is desired, we can discuss it in detail and I would love to contribute a basic implementation.

[mperham]

The INFO command provides basic metrics already but you'd have to pull them. I'm still in the process of deciding how to support various metrics standards like Prometheus and Statsd.

[lukasmalkmus]

Today I hacked together a very basic exporter. It isn't very powerful, just supports a few key metrics.

[mperham]

Faktory Pro now has Statsd support. I will add first-class Prometheus support based on customer demand.

[alondahari]

@mperham Was there any work towards supporting exports to Prometheus? We would like to utilise those for alerting for our enterprise account.

[mperham]

Not yet, you're the first to request Prometheus support.

[MrDHat]

Adding my vote to support prometheus metrics!

[MRigal]

Thanks @lukasmalkmus for bringing this up and providing a first exporter. As I had problem to use your package from source as well as in a docker container, I've written a similar Python implementation: https://github.com/prospect-server/faktory-prometheus-exporter