Quite a few technical posts this week, with several on new developer or integration tools around Kubernetes. Alongside those we have content on learning, sharing ideas, trunk based development and more.

From our sponsor, VictorOps

Endorsing a culture of blameless transparency around post-incident reviews can lead to continuous improvement and more resilient services. Check out some techniques that DevOps teams are using to improve post-incident analysis and learn more from failure:
http://try.victorops.com/devopsweekly/fishbone-diagram-post-incident-reviews

News

A nice post on the importance of investing in learning when it comes to operations, both learning from running systems, from incidents and from the past.
https://blog.dbsmasher.com/2019/08/08/how-we-keep-learning.html

A post advocating trunk based development, and skipping pull request based workflows. Lots of good arguments for this in the abstract, but like the author I don’t see most teams working this way.
https://medium.com/@mattia.battiston/why-i-love-trunk-based-development-641fcf0b94a0

The Cloud Native Computing Foundation (CNCF) are running a wide ranging survey focused on usage of the various projects in the foundation and about the role of the organisation.
https://www.surveymonkey.com/r/cloudnativesurvey2019

Conversations around multi-cloud tend to fall a little into idealistic and pragmatic. This post outlines some of the FUD without over selling any particular advantage.
https://stacksense.io/krishnan/thought-leadership/busting-the-multi-cloud-drumbeat/

A post on investigating a production incident with Kubernetes and Istio and learning more about readiness probes, autoscaling and resource exhaustion in the process.
https://medium.com/kudos-engineering/increasing-resilience-in-kubernetes-b6ddc9fecf80

A user experience report of using the Garden development tool for Kubernetes. Looking forward to other reviews of tools in this space.
https://www.kartar.net/2019/08/garden/

A look at using Kind (Kubernetes in Docker) to test applications targeting Kubernetes, with an example of using Brigade to automate that testing on a Kubernetes cluster.
https://radu-matei.com/blog/kubernetes-e2e-kind-brigade/

One part of different disciplines working together is sharing ideas. This post has some specific tips for embracing devops and working across development, operations and QA teams on new ideas to improve the way you work.
https://medium.com/@daryn_74143/lets-get-dev-ops-and-qa-talking-about-ideas-d2d50d4ea976

BuildKit, the build library that powers docker build, supports custom frontends which allow you to build your own build UI. This post explores why that’s useful with a demo of a high-level declarative build config file for Ubuntu base images.
https://matt-rickard.com/building-a-new-dockerfile-frontend/

A good discussion about the role and pitfalls of CloudFormation, and more generally some useful thoughts about declarative infrastructure APIs and client side abstractions built on top of them.
https://read.acloud.guru/cloudformation-is-an-infrastructure-graph-management-service-and-needs-to-act-more-like-it-fa234e567c82

Most examples of writing tools for Kubernetes use Go, but I find it useful to highlight examples in other languages. Here’s a useful tutorial for writing a Kubernetes operator in Python.
https://medium.com/flant-com/kubernetes-operator-in-python-451f2d2e33f3

Events

The 3rd Sensu summit is coming up in Portland, Oregon on the 9th and 10th of September. It features technical talks as well hands-on workshops on all things monitoring related. You can get a sizeable discount of tickets with the following link as well.
https://ti.to/sensu/sensu-summit-2019/discount/DevOpsWeekly

Tools

Octant is a new Kubernetes dashboard which provides a rich plugin model for others to extend its functionality and add new components.
https://github.com/vmware/octant

A useful library for anyone writing Go-based HTTP services. Sundheit allows you to register async health checks for your dependencies and the service itself and provides a health endpoint that exposes status and health metrics.
https://medium.com/appsflyer/health-checks-like-a-pro-3fa362942883
https://github.com/AppsFlyer/go-sundheit

Endorsing a culture of blameless transparency around post-incident reviews can lead to continuous improvement and more resilient services. Check out some techniques that DevOps teams are using to improve post-incident analysis and learn more from failure:
http://try.victorops.com/devopsweekly/fishbone-diagram-post-incident-reviews