1 minute read

Posts this week covering the emergence of platform teams, journald, AWS and a writeup from a devops conference held on a virtual island populated by animals.


Many organisations adopting devops practices form platform teams. This post looks at some of the challenges and approaches to managing the product of an internal platform.

A nice opinionated post about AWS services that you might avoid, at least until you know how best to use them.

A look at using GitHub Actions to deploy your application based on a pull request. Some good tips for interacting with the GitHub API in your Actions too.

A long writeup from the recent Deserted Island Devops conference (which took place in Animal Crossing). Lots of the discussion and links to slides and videos from the talks.

A post on combining Helm templates with Kustomize for overriding specific attributes, useful when you want to use third-party content with some minor changes.

An solid introduction to service mesh specifically for infrastructure teams.

Istio is a powerful example of a service mesh, which has lots of features and things which can be configured. This collection of learning resources should get you started.

Journald underpins logging on many modern Linux operating systems. This post explores in detail how it’s configured, how it stores logs and in particular how you can access and make use of those logs.


Tinkerbell is a new open source bare metal provisioning engine. It’s based around a concept of workflows which can be stored as templates in source control and aims to scale down to devices as well as up to large datacenter deployments.

Terraform Compliance is a tool for writing BDD-style tests for Terraform code.

A new set of buildpacks, built with Google Cloud Platform in mind, but suitable for any environment. Supports Go, Python, Node, Java and .NET. Good documentation around configuration too.