add operational_metrics support

[eranra]

No description provided.

[eranra]

@memodi please have a look at https://github.com/netobserv/flowlogs-pipeline/blob/e4c38d75c32ecc49b639fce89516a83e0bd272fc/docs/operational-metrics.md those are the new operational metrics that should help us to understand if and how FLP is working. If needed we can add more, please advise :-)

@jotak FYI ^^^

[memodi]

thanks @eranra these will be certainly helpful. Added couple of comments on naming but probably have more feedback once I actually test it.

Couple of questions:

1. are these enabled by default?
2. will these be pushed to prometheus?
3. I believe once this task https://issues.redhat.com/browse/NETOBSERV-247 is finished, we'd be also able to grab these metrics from the FLP endpoint, correct?

[eranra]

thanks @eranra these will be certainly helpful. Added couple of comments on naming but probably have more feedback once I actually test it.

Couple of questions:

1. are these enabled by default?

Yes, as long as you enable the Prometheus encode stage. Currently, this is NOT the configuration in NOO so we will need to change that for those metrics to be available for you :-)

2. will these be pushed to Prometheus?

YES

3. I believe once this task https://issues.redhat.com/browse/NETOBSERV-247 is finished, we'd be also able to grab these metrics from the FLP endpoint, correct?

@jotak please help with (3) --- I think that the answer is yes, but just want to be on the safe side here.

[ronensc]

I like the idea of auto-generated documentation of the operational metrics.