Huamin Chen
Huamin Chen
@beneiltis kepler doesn't have Apple M1 energy sensor yet, it is something we haven't started. btw, I have to disable the arm64 image build because the libbpf has an architecture...
fix is in #1255
@YaSuenag thanks for the update. For 1) kepler loads ebpf program and first tries to attach `finish_task_switch`, if failed, then attaches `finish_task_switch.isra.0`. The error message`failed to attach kprobe/finish_task_switch: failed to...
I think EKS is still in cgroup v1. In this case, can you add `-enable-cgroup-id=false` in kepler? You can run `kubectl edit daemonset/kepler-exporter -n kepler` and change the args [here](https://github.com/sustainable-computing-io/kepler/blob/main/manifests/config/exporter/exporter.yaml#L72)...
The dashboard is not well tested. There is another similar case here https://github.com/sustainable-computing-io/kepler/issues/1321 Can you check if the metrics e.g. `sum(rate(kepler_container_joules_total[1m]))` on the prometheus UI? It could be somewhere in...
I suspect this is a overflow issue. Do you have this issue before or only during this period?
Maybe there are some overflow during extended tests? @vprashar2929 have you seen this in any tests?
@jochen-schuettler can you get prometheus stats `kepler_container_joules_total` over the period of spikes? This stats is used to calculate the power consumption.
@emoek can you provide more info? what is the kepler image you are using, what is the OS platform, do you have access to the kepler container logs?
@SamYuan1990 yes thats the config to enable it