Cilium BPF Conntrack Map Exhaustion: Random Resets While conntrack Looks Fine
Random resets with Cilium? Learn how eBPF conntrack (CT) maps fill up, why netfilter conntrack looks fine, and how to size + verify fixes in Kubernetes.
3 posts
Random resets with Cilium? Learn how eBPF conntrack (CT) maps fill up, why netfilter conntrack looks fine, and how to size + verify fixes in Kubernetes.
CPU is at 20% but latency is 500ms. Standard profilers show nothing. The app is waiting, not computing. I show how to use eBPF to find what it's waiting for.
CPU utilization is low but requests are slow. The hidden culprit: time spent waiting in the scheduler run-queue, invisible to traditional profilers but visible with eBPF off-CPU analysis.