Welcome to a very special expanding brain edition of kubelist! This week, we're bringing you a collection of thought pieces across the full spectrum. Did Kubernetes win already? Is it too complex to win? Or maybe it doesn't matter, and you just want to get these gosh darn logs aggregated?
The kubelist editors have recently uncovered a massive conspiracy that permeates our entire industry: Companies that provide logging services all have names that are puns on the wood processing industry! Conspirator Mark Corwin describes setting up centralized logging for Kubernetes on the Timber blog. The kubelist editors are particularly fond of the pattern for using a sidecar to redirect logs from a file to stderr and stdout.
Matt Asay dissects the recent discussion over OpenShift's status as a fork (or not) of Kubernetes. We're likely to see this argument levelled against more projects and products in the future, as more things are built on top of and around Kubernetes. If OpenShift was built on API server aggregation instead of a copy of the Kubernetes code base but functionally identical, would it still be a fork? This is the kind of deep existential question we must ask ourselves.
Aside: The kubelist editors would love to get their hands on a copy of Matt's thesaurus.
Jorge Castro introduces the Kubernetes Discourse instance on the Kubernetes blog. Everyone learns and communicates differently; adding one more avenue for Kubernetes is a great move.
James Governor has read the Kubernetes tea leaves and provides us with insight on where Kubernetes and the surrounding ecosystem need to go. He has a few answers to the question, a fair evaluation of serverless and cloudevents, and a great first name, to boot!
Half off a Kubernetes book; sweet! Hopefully you're reading this email the same day we sent it out, or else you've missed your chance.