(#74) collect local PCP metrics

[bastelfreak]

also includes #178

[bastelfreak]

@m0dular any chance you can test that? The telegraf agents at my PE customer start now and have the PCP configs. telegraf also doesn't log any errors. But I poked influxdb via grafana and checked for puppetlabs.pcp.connections metrics and didn't find anything. Maybe I did it wrong or maybe something else is missing.

[m0dular]

Works for me. I used your pcp branch, set puppet_operational_dashboards::telegraf::agent: local, and all the expected config was there. Testing it manually works

$ telegraf --test --debug --config /etc/telegraf/telegraf.conf --config /etc/telegraf/telegraf.d/pcp_metrics.conf

> orchestrator,host=pe-server-e5a7ce-0,mbean=default:name\=puppetlabs.pcp.connections,url=https://pe-server-e5a7ce-0.us-west1-a.c.customer-support-scratchpad.internal:8143/metrics/v2/read/default:name\=puppetlabs.pcp.connections Value=2 1686767303000000000

as does the service. I'm able to see a Value field for the default:name=puppetlabs.pcp.connections mbean under the orchestrator measurement.

[bastelfreak]

ah I had the wrong measurement. perfect. I think the sles errors aren't related? Can this be merged?

the issue was resolved