The Sched app allows you to build your schedule but is not a substitute for your event registration. You must be registered for KubeCon + CloudNativeCon Europe 2025 to participate in the sessions. If you have not registered but would like to join us, please go to the event registration page to purchase a registration.
Please note: This schedule is automatically displayed in British Summer Time (BST) (UTC +1). To see the schedule in your preferred timezone, please select from the drop-down menu to the right, above "Filter by Date." The schedule is subject to change and session seating is available on a first-come, first-served basis.
Understanding memory statistics in Kubernetes is critical for reducing an application’s memory impact or avoiding the Out-Of-Memory (OOM) killer. In this talk, we’ll decode the complexities behind Kubernetes memory metrics (did you say container_memory_working_set_bytes?), tracing them from the kubelet binary to the host’s memory control groups.
The memory metrics we observe, whether through kubectl top or Prometheus, are the result of a complex journey, from memory control group statistics, through calculations by libraries like opencontainers/libcontainer, to cAdvisor or the container runtime, and finally, to the kubelet. We’ll deep dive into the role of cAdvisor and the container runtime in memory tracking, the interaction with the OOM killer, and the impact of control groups (cgroups) versions on metric calculations. By the end of this session, you’ll be able to better interpret memory statistics and troubleshoot memory-related issues in your clusters.
Mahé is a security engineer at Isovalent and an active contributor to Kubernetes SIG Security. He was previously working as a security researcher and loves working with Linux, security, and Kubernetes!
