Bringing the power of eBPF tools to your k8s web UI
summary
Built a kubernetes-aware eBPF workflow: Inspektor Gadget runs a gadget pod daemonset and a kubectl plugin to inject BCC/eBPF probes per node and translate pod selectors to PIDs, and Headlamp loads plugins that create a Trace CRD to start/stop probes via annotations. Demos show a block I/O latency histogram from a biolatency gadget and automatic seccomp rules generated from observed syscalls; integration works on minikube and returns output in the CR status for UI consumption.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 0:42Agenda
- 1:11Inspector Gadget
- 2:34What does it mean
- 3:53How does it work
- 4:37Wrappers
- 5:59What is Trace
- 7:15Demo
- 13:51Headlamp
- 16:58Frontend plugins
- 17:37How to run
- 17:59Live demo
- 24:17Future plans
related talks
28mDebug Your Clusters with eBPF-Powered Tools
24mMaking the Kernel Speak Kubernetes: Unlocking eBPF’s Power for Observability
25mBeyond the Buzzword: BPF’s Unexpected Role in Kubernetes
40mUsing the EBPF Superpowers To Generate Kubernetes Security Policies
30mObservability in Kubernetes: Unlocking the Power of eBPF
27mStep By Step Kubernetes Observability with eBPF
34mStep by Step Kubernetes Observability with eBPF
29mKubernetes Observability with eBPF