> 1. what happened to the PCP approach we talked about in the past?

Thats going in parallel, PCP is just a monitoring framework you can
integrate with instrumentation data just like any other.

> 2. how would you integrate this to sosreport/foreman-debug? I'm thinking of
> storing the statsd data locally, collecting them with foreman-debug, and
> then, being able to import them later to the prometheus and other tools. Is
> this how this could work? Any other options?

This is my ultimate goal to have working PCP deployment including
telemetry data and archives could be collected by foreman-debug, they
are pretty small (few MBs per day).

> 3. does every host/runtime needs it's own statsd service, or there would be
> one shared process? Asking bith for multi-host and containers use-case

It is up to you if you want one statsd service per guest/container,
host or subnet. Prometheus endpoint will not require any external
daemon once sharing metrics is merged into upstream. For this reason,
statsd will server as a temporary solution and alternative for the
future.

> The proposal of the telemetry api itself seems reasonable, let's discuss
> that on an actual PR

Thanks, I hope to finish it this year.

-- 
Later,
  Lukas @lzap Zapletal

-- 
You received this message because you are subscribed to the Google Groups 
"foreman-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to