Some good long reads this week, on everything from counter arguments to SRE practices, a few posts on successfully adopting devops and cloud practices in large organisations and several posts looking at the latest tools.

From our sponsor, VictorOps

Ever been on-call? Then you know it can suck. Check out some of our tips and tricks to see how DevOps teams are maintaining composure during a critical incident and making on-call suck less:
http://try.victorops.com/devopsweekly/on-call-template

News

Some good tips for adopting devops practices in large organisations. Discussion of silos, standardized tooling, communication channels and more.
https://medium.com/devopslinks/how-to-approach-the-challenges-of-devops-in-large-organizations-b1a31e0562a6

It’s always good to hear counter arguments to popular or emergring practices I find. This post looks at why error budgets might not be a good idea when building resilient systems and teams.
http://willgallego.com/2019/02/23/resilience-engineering-and-error-budgets/

A detailed look at using Open Policy Agent. Discussion of where RBAC alone breaks down for some usecases and examples of specific policies for a shared cluster environment.
https://itnext.io/kubernetes-authorization-via-open-policy-agent-a9455d9d5ceb

An interesting look at which cloud migration projects, and adopting cloud native approaches, sometimes fail - and just as importantly what you can do to avoid that fate.
https://container-solutions.com/why-do-cloud-migrations-fail/

A nice write up of debugging a database problem. Stepping through the steps taken, and providing a good introduction to the Postgres query planner in the process.
https://gocardless.com/blog/debugging-the-postgres-query-planner/

A look at testing Helm charts using new functionality in Terratest. The post explains the why as well as how to write and run integration and unit tests for charts and Kubernetes configuration in general.
https://blog.gruntwork.io/automated-testing-for-kubernetes-and-helm-charts-using-terratest-a4ddc4e67344

A post asking the question what is a good metric? Good tips for any working out what to measure.
https://medium.com/@fmrigueiro/what-is-a-good-metric-bec313bbc8c7

The State of Open Source Security report is out. Some interesting details on the growth of third-party package ecosystems and on how developers are testing code for security vulnerabilities.
https://snyk.io/opensourcesecurity-2019/

An extensive report on what’s new in Salt 2019.2 Fluoride. Lots of new networking device support, interop with Terraform and Ansible and lots more.
https://salt.tips/the-coolest-features-in-salt-fluorine/

A walkthrough of Kapitan, a tool for templating configuration files with jsonnet and Jinja templates that can be applied to Kubernetes.
https://medium.com/@alessandro.demaria/introduction-to-kapitan-adb6a488cd77

Tools

K3s is a stripped down Kubernetes implementation which removes lots of optional features aiming for a Kubernetes that is useful in IoT and Edge environments.
https://k3s.io/
https://github.com/rancher/k3s

Ever been on-call? Then you know it can suck. Check out some of our tips and tricks to see how DevOps teams are maintaining composure during a critical incident and making on-call suck less:
http://try.victorops.com/devopsweekly/on-call-template