Linux - Minikube

On this page

https://github.com/lftraining/MWC

linux foundation training

Terminal window
curl -LO https://storage.googleapis.com/minikube/releases/latest/minikube-latest.x86_64.rpm
sudo rpm -Uvh minikube-latest.x86_64.rpm
sudo dnf install podman-docker
Terminal window
minikube config set rootless true
minikube start --driver=podman

minikube start —driver=podman —cpus=4 —memory=8192

eBPF monitor

sudo dnf install bpftool

https://dev.to/maheshrayas/-intro-into-ebpf-and-rust-l2m

eBPF is a technology that allows developers to run safe, sandboxed programs in the Linux kernel. These programs can be used for networking, observability, and security, without modifying kernel source code or loading kernel modules.

eBPF runs inside a restricted virtual machine in the kernel — somewhat like how WebAssembly (WASM) or the JVM operates in userspace — ensuring safety and control over what the program can do.sudo

Terminal window
sudo dnf check-update
sudo dnf install kernel-tools kernel-headers cloud-utils

Probe BPF system capabilities. This command queries the kernel for supported BPF features; look for “supported” on key items like bpf_syscall, map types, and program types. Some may be restricted to only root users.

Terminal window
sudo bpftool feature probe | grep -E 'bpf|BPF'

Install Required eBPF Tools

Update package index and install core packages

Terminal window
sudo dnf install -y bpftrace clang llvm libbpf-devel bcc-tools

student@cp:$ sudo apt update student@cp:$ sudo apt install -y bpftrace clang llvm libbpf-dev linux-headers-$(uname -r) bpfcc-tools

These packages provide bpftrace (high-level scripting), bpftool (inspection and debugging), clang/LLVM (compilation), and BCC tools (classic BPF examples) needed throughout the course.

Verify installations student@cp:$ which bpftrace && bpftrace —version student@cp:$ which bpftool && bpftool version student@cp:~$ which clang && clang —version Successful output confirms the tools are available in PATH and ready for use.

Run eBPF “Hello World” with bpftrace

Execute simplest possible eBPF program. The BEGIN probe triggers immediately when the program loads, proving that the kernel accepted and executed a BPF bytecode program. If the message appears without errors, eBPF loading, verification, and JIT compilation are working correctly. student@cp:~$ sudo bpftrace -e ‘BEGIN { printf(“Hello, eBPF World!\n”); exit(); }’ Attaching 1 probe… Hello, eBPF World!