Debugging complexity, databases (both large and small), keeping AWS tidy, Git at scale and more in-the-weeds posts this week.

StackHawk sponsors Devops Weekly

Already working in GitHub? Check out a sneak preview of the StackHawk and GitHub CodeQL integration to see how your team can add automated security testing to their development workflow.
https://sthwk.com/github-codeql-overview

News

A post on why SQLite is so great. Some notes on features in the latest release, along with lists of extensions and other distributions.
https://i-programmer.info/news/84-database/15609-in-praise-of-sqlite.html

Has debugging production systems become harder? This post says yes, and points to the increased architecture complexity.
https://www.honeycomb.io/blog/why-intuitive-troubleshooting-stopped-working/

Libyear is an interesting idea for tracking dependency freshness. It’s a single number that can be aggregated in different ways to make comparison easier.
https://libyear.com/

Git is the de facto choice for version control, and it’s used by everything from small projects with a single maintainer up to huge projects like Kubernetes. This post looks at some of the challenges at Kubernetes scale.
https://matt-rickard.com/how-kubernetes-broke-git/

Discussion of using stacked pull requests in Git. Keeping individual PRs small, but chaining them with dependencies on each other. The post doesn’t discuss this pattern in slower moving or more async environments though.
https://benjamincongdon.me/blog/2022/07/17/In-Praise-of-Stacked-PRs/

A post on the typical evolution of a startup’s architecture as it scales, with a focus on the database architecture as you deploy across multiple regions.
https://www.cockroachlabs.com/blog/startup-reference-architecture/

A brief introduction to some useful maths for understanding and making the most of monitoring systems.
https://dejanualexandru.medium.com/the-mathematics-behind-monitoring-ddbd9403c32c

A project tracking the (incredibly rare) breaking changes in the AWS API. Most of these are deprecation and then removal of features, rather than just breaking the APIs, which is credit to AWS’s approach here.
https://github.com/SummitRoute/aws_breaking_changes

Tools

A handy utility to automatically cleanup an AWS environment based on time-to-live and local configuration.
https://github.com/servian/aws-auto-cleanup