Kubernetes Troubleshooting (LFWS313)
This one-day course introduces the tools and techniques for troubleshooting a Kubernetes cluster and the applications running on it.
Suggested prerequisites: practical Linux administration skills and proficiency with the Linux command line, including editing files with a command-line text editor and working in multiple terminal sessions. You also need hands-on experience with command-line utilities such as kubeadm and kubectl and a solid understanding of Kubernetes administration.
Audience: DevOps and SRE engineers, cloud engineers, Kubernetes administrators and anyone who manages and monitors clusters.
When something breaks in Kubernetes, the people who can find the cause quickly are the ones everyone relies on. This course is for those who manage and monitor clusters and want to deepen their troubleshooting knowledge, so they can tackle even the toughest issues with confidence.
After a short recap of Kubernetes concepts, you learn the principles of troubleshooting and the essential tools that go with them, from kubectl and the dashboard to auditing, Prometheus and Grafana, Fluentd and Jaeger. Together with observability solutions, these help you reach faster and more accurate resolutions. You then work through common errors and their solutions: kubeadm and upgrade problems, image and pod errors, scheduling, storage, Secrets and ConfigMaps, namespaces, authorization and RBAC, resource quotas, networking, node errors and performance issues.
The course is built around hands-on labs and simulations, with extra self-paced practice labs for further study. You leave better able to diagnose and resolve issues in pods, services, networking, storage and nodes, and better prepared for advanced roles such as Kubernetes Administrator, DevOps Engineer or Cloud Infrastructure Architect.