Fixing ImagePullBackOff in Kubernetes
A pod can't start because Kubernetes can't pull its container image — the fix depends entirely on which of a handful of specific causes is actually responsible, from a typo to a private registry auth problem.
tag
17 posts
A pod can't start because Kubernetes can't pull its container image — the fix depends entirely on which of a handful of specific causes is actually responsible, from a typo to a private registry auth problem.
Founded June 22, 2015 by Docker, CoreOS, and a broad industry coalition, the OCI set out to make container images and runtimes portable across tools and vendors rather than tied to any one implementation.
A complete walkthrough creating and running a lightweight systemd-nspawn container — useful for isolated testing environments without the overhead of a full container runtime.
Docker Desktop configured to use the WSL2 backend suddenly can't see your distro, or containers fail to start with confusing errors. Here's how to work through the specific integration points that commonly break.
A complete walkthrough getting Docker Desktop configured to use WSL2 as its backend, integrating specific distros, and verifying containers actually run using WSL2's real Linux kernel rather than a separate VM.
How the OCI runtime and image specs standardized what a 'container' actually is, and how containerd/CRI-O/runc fit together beneath Docker and Kubernetes.
How Podman's daemon-less, fork-exec architecture differs from Docker's client-daemon model, and what that means for rootless containers in production.
A pod repeatedly crashes and restarts, sitting in CrashLoopBackOff. Here's a systematic way to find out why instead of just deleting and recreating the pod.
Docker builds and pulls start failing with ENOSPC, even though the host's regular disk usage doesn't look that high. Docker's own storage accumulates in places df alone won't clearly show you.
How vulnerability scanners actually inspect container image layers, how to read a scan report, and the practices that reduce real supply-chain risk.
How pod-to-pod networking, Services, and kube-proxy's packet rewriting fit together to make Kubernetes' flat network model actually work.
The two-phase filter-and-score process the Kubernetes scheduler uses to decide which node a pod lands on, and how to influence it.
How multi-stage builds, distroless base images, and layer discipline combine to produce smaller, more secure container images without sacrificing developer ergonomics.
At PyCon on March 15, 2013, dotCloud co-founder Solomon Hykes introduced Docker to the world, ahead of the company's later pivot to focus on it entirely.
On October 29, 2013, dotCloud announced it was scaling back its original PaaS business and renaming the company entirely around its container tooling.
How cgroup v2's unified hierarchy replaces v1's tangled controller mounts, and how to read and write limits directly through the filesystem.
How each of the Linux kernel's namespace types isolates a specific global resource, and why containers are just processes with a curated set of them.