Missing metrics for pods/containers in Kepler v0.10.2 (older version - v0.9.0 works fine)

[jaworski-piotr]

Hi,
I’m new to Kepler and I’m having trouble getting metrics for pods and containers starting with version v0.10.2. For comparison, I also tested an older version, which works fine.

Environment:

• Kubernetes: v1.27.16 (Client & Server)
• OS: Ubuntu 20.04.6 LTS
• Kernel: 5.15.0-139-generic
• Container runtime: Docker 26.1.3
• CPU: Intel(R) Xeon(R) Gold 6330N @ 2.20GHz
• Kepler installation: Helm chart

Older version (working):
Installed following instructions: https://sustainable-computing.io/archive/installation/kepler-helm/
Metrics for pods and containers were available.
I’m attaching a Grafana screenshot, Kepler logs, Prometheus metrics (in text format), and example metrics exported to a CSV file.

kepler_metrics_old.txt
kepler_log_old.log
csv_metrics_old.zip

New version (problem):
Installed following the current guide: https://sustainable-computing.io/kepler/installation/guide/#1-helm-chart-installation-recommended-for-kubernetes
helm install kepler https://github.com/sustainable-computing-io/kepler/releases/download/v0.10.2/kepler-helm-0.10.2.tgz --namespace kepler --create-namespace

In the podInformer logs, Kepler detects pods and containers. Example:
time=2025-08-11T12:45:36.949Z level=DEBUG source=k8s/pod/pod.go:190 msg="containers for pod" service=podInformer pod=a1mediator-649dfbd6f9-rmf8r containers=930bcc596d8b5b53255189f7c7d9b2800e217be54a8bdc9ee549dce70732d873

However, another log shows 0 pods reported and only 8 containers, while there should be over 112 pods and containers running in the environment:
time=2025-08-11T12:45:37.249Z level=DEBUG source=internal/resource/informer.go:397 msg="Resource information collected" service=resource-informer process.running=2011 process.terminated=0 container.running=8 container.terminated=0 vm.running=0 vm.terminated=0 pod.running=0 pod.terminated=0 container.no-pod=8 duration=328.61002ms

As a result, there are no metrics per pod/container.
I’m attaching Kepler logs, Prometheus metrics (in text format), and example metrics exported to a CSV file.
kepler_metrics_new.txt
kepler_log_new.log
csv_metrics_new.zip

Request for help:
I’m not sure what causes the missing pod and container metrics in Kepler v0.10.2. I wonder if some additional configuration is required that I might have missed, or if this could be a known bug in this version. There might also be something else I overlooked. I would appreciate any guidance or help in resolving this issue.