Kubernetes Upgrades — How Not to Mess It Up
A real-world guide for DevOps engineers who want zero-downtime upgrades and zero post-mortems The Story That Started This Guide The alert came in at 9:47 PM. “API server unreachable.” The on-call engineer opened his laptop to find that the cluster upgrade — which was “just a version bump from 1.27 to 1.28” — had gone completely sideways. The … Read more