Several interesting posts on Installation and packaging of software this week. Building tools for installation and packaging is a common activity for operations and platform teams everywhere I feel.
StackHawk sponsors Devops Weekly
ICYMI: The StackHawk & GitHub CodeQL Integration webinar is available on-demand! See how you can add automated security testing to your pipeline without throwing a wrench in your roadmap. Watch it here:
A good discussion of scaling installers for complex software, in this case moving from Helm to a custom installer written in Go.
Packagers play a hugely important but often hidden role. Here’s a nice behind the scenes look at building reproducible packages for Go software in Arch Linux.
A nice two part detailed series on cost optimisation for cloud infrastructure. Everything from spot instances and processor choice to cloud provider tooling and more.
More and more organisations are starting to use ML, which often comes with a decent infrastructure footprint. This post covers how one organisation ensure good observability for that ML infrastructure.
A good post on the theory behind building a platform team and providing opinionated platforms to scale development teams.
A detailed set of posts on configuring AWS services, in this case how to create service credentials without exposing them to users.
Still an occasional topic as SRE and Devops reaches a wider audience, is SRE just traditional operations with a new name (and does it matter?).
BindPlane OP is a new open source observability pipeline that gives you the ability to collect, refine, and ship metrics, logs, and traces to any destination.
One of the best things about Kubernetes is the ability to extend it via custom resources. This repo collects together over 100 custom resources and there schemas.