This article examines a recurring Kubernetes issue where a node becomes NotReady for exactly three minutes after a kubelet restart, explains the underlying PLEG and node‑status mechanisms, and offers a free, in‑depth Kubernetes handbook for further study.
This article examines a recurring Kubernetes node NotReady issue where nodes become ready for only three minutes after a kubelet restart, detailing the underlying PLEG mechanism, component interactions, and diagnostic steps to resolve the problem.
This article dissects the kubelet source code, explaining how the Pod Lifecycle Event Generator (PLEG) works in both GenericPLEG and EventedPLEG modes, how events are streamed, processed, and fed into syncLoopIteration, and how the kubelet reacts to container lifecycle changes.
When replacing dockershim with containerd, we observed that pods take over a minute to start because the GenericPLEG Relisting operation stalls for more than 30 seconds during node boot, caused by containerd’s UpdateContainerResources holding a bbolt lock and intensive image pulls; the article explains the root cause and provides a fix using the overlay volatile mount option.
A high‑load Kubernetes node entered NotReady due to repeated OOM‑killer activity, daemonset restarts, and PLEG health failures, and the article walks through diagnosis, log analysis, root‑cause explanation, and practical remediation steps to restore node readiness.
This article explains the role of the Pod Lifecycle Event Generator (PLEG) in Kubelet, why the “PLEG is not healthy” error causes nodes to become NotReady, common failure scenarios, and a step‑by‑step troubleshooting method that ultimately resolves the issue by upgrading systemd.
This article analyzes a recurring NotReady issue in Alibaba Cloud Kubernetes clusters, explaining how kubelet, NodeStatus, and the Pod Lifecycle Events Generator (PLEG) interact, why nodes become Ready for only three minutes after a kubelet restart, and how the underlying timeout mechanisms cause the problem.
The article analyzes a recurring Kubernetes node NotReady problem where restarting kubelet only restores readiness for three minutes, explains the underlying PLEG timeout mechanism, investigates unresponsive Terway CNI plugin behavior, and presents a netlink‑timeout fix to prevent the issue.
This article walks through a low‑probability NotReady issue in Alibaba Cloud Kubernetes clusters, detailing how systemd’s 32‑bit cookie overflow in its dbus handling caused container runtime failures, the step‑by‑step debugging process, and the eventual upstream fix.
