Explore the possibility of realtime monitoring

[massa-bot]

In GitLab by @damip

Explore the possibility to expose some real-time metrics from the node.
Possibilities:

• prometheus
• influxdb
• victoriametrics https://github.com/VictoriaMetrics/VictoriaMetrics
• etc...

Some ideas about metrics:

• CPU/RAM/internet/disk usage
• block/op/endorsement/header creation/processing/transmission events
• finality / stale rates for blocks/transactions etc...
• pool size
• connected node count
• known node count
• clique count
• api calls
• etc...

[AureliaDolo]

How much is this related to #1266 and massalabs/massa-sc-runtime#3 ?

[qdrn]

I don't think it's related at all to metering.

[damip]

it's not related indeed

[qdrn]

Bumping this discussion as people are struggling to identify what is going wrong with their node. As the community grow this will take more and more energy to respond to people.

Related to #2009

[damip]

Added to the refactoring roadmap #2162

[AureliaDolo]

It's also related to #1312