Uploaded image for project: 'Network Observability'
  1. Network Observability
  2. NETOBSERV-2195

BPF resource usage metrics

XMLWordPrintable

    • Icon: Story Story
    • Resolution: Unresolved
    • Icon: Undefined Undefined
    • None
    • None
    • Prometheus
    • None
    • None
    • False
    • Hide

      None

      Show
      None
    • False
    • None
    • None
    • None
    • None

      Bpfman should at some point expose more BPF metrics

      When it's implemented, we should capture them in our Health dashboard.

       

      Previously:

      Expose CPU & memory metrics per BPF programs/maps, as Prometheus metrics

      See also: https://github.com/Netflix/bpftop

      https://elixir.bootlin.com/linux/v6.6.16/source/include/uapi/linux/bpf.h#L792

      Since it impacts performances, this would be an opt-in feature

              Unassigned Unassigned
              jtakvori Joel Takvorian
              None
              None
              Amogh Rameshappa Devapura Amogh Rameshappa Devapura
              None
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

                Created:
                Updated: