Containerization
Post #5299 · source on Telegram
When kubectl outputs pods to PDF, revealing Kubernetes' surprising PDF library dependency
Description
Dark-theme Twitter screenshot. The top tweet by “Cless @Arubein · 8m” reads: “kubectl get po -o pdf”. Below it is a quoted tweet from “David Crawshaw @davidcrawshaw · 1h” saying: “I fear the answer but cannot help asking: why does Kubernetes depend on rsc.io/pdf, a package for *reading* PDF files? github.com/kubernetes/kub…”. A blue link underneath says “Show this thread”. The humor comes from treating the kubectl CLI like it natively supports an “-o pdf” output format because the Kubernetes codebase unexpectedly includes a Go library for parsing PDFs, highlighting dependency sprawl and the odd places packages end up in large container orchestration projects
More like this
More like this in Containerization →
Use J and K for navigation
Comments
11Comment deleted
Kubernetes vendors a PDF parser so kube-apiserver can skim the 800-page SBOM explaining why it vendors a PDF parser
The real reason Kubernetes is so resource-intensive isn't the container orchestration - it's rendering your pod status as a beautifully formatted PDF complete with embedded fonts and vector graphics, because apparently YAML wasn't enterprise enough
When you're debugging why your Kubernetes cluster is pulling in a PDF parsing library and realize someone probably needed to generate a single compliance report five years ago, but now it's in the dependency tree forever. At least we can dream of 'kubectl get po -o pdf' generating beautifully formatted pod listings with proper kerning and embedded fonts - because clearly that's what production infrastructure needs at 3 AM
Proof that transitive deps grow faster than clusters: add one library and suddenly “kubectl get po -o pdf” is an enterprise reporting feature - now who owns the printer-driver SIG?
Kubernetes deps: because container orchestration needed a side hustle parsing the fine print on those vendor lock-in PDFs
kubectl get po -o pdf - because when your SBOM includes a PDF parser, observability is for auditors, not grep
so why? Comment deleted
It doesn't, and it should've been removed by now https://github.com/kubernetes/kubernetes/pull/97337 https://github.com/kubernetes/kubernetes/pull/102833 Comment deleted
Why on Earth does Midnight Commander depend on X11 libraries? Comment deleted
Because Comment deleted
oh i kinda wish kcl get po -o pdf was real Comment deleted