feat: custom SEL events metrics

alexandrovas commented 8 months ago

Added a new collector to collect custom metrics from SEL events via IPMI.

In config metrics are defined in this format:

modules:
  default:
    collectors:
      ...
      - sel-events
    sel_events:
      - name: correctable_memory_error
        regex: Correctable memory error.*

Metrics are shown as:

# HELP ipmi_sel_events_count_by_state Current number of log entries in the SEL by state.
# TYPE ipmi_sel_events_count_by_state gauge
ipmi_sel_events_count_by_state{state="Nominal"} 12
ipmi_sel_events_count_by_state{state="Warning"} 128
# HELP ipmi_sel_events_count_by_name Current number of custom log entries in the SEL by name.
# TYPE ipmi_sel_events_count_by_name gauge
ipmi_sel_events_count_by_name{name="correctable_memory_error"} 140
# HELP ipmi_sel_events_latest_timestamp Latest timestamp of custom log entries in the SEL by name.
# TYPE ipmi_sel_events_latest_timestamp gauge
ipmi_sel_events_latest_timestamp{name="correctable_memory_error"} 1.704979201e+09

Closes #41

alexandrovas commented 8 months ago

@bitfehler @SuperQ Hello! Can you please review my PR

bitfehler commented 7 months ago

Thanks a lot. For the record, I think we might be handing folks a decent foot-gun here, but hey :slightly_smiling_face:

prometheus-community / ipmi_exporter

feat: custom SEL events metrics #179