Architecture, new tools, operational exercises and tips for managing teams adopting devops practices.

Sponsor

Learn from failure to improve incident response. Start conducting post-incident reviews for deeper analyses and start making future incident remediation easier:
http://try.victorops.com/devopsweekly/post-incident-review

News

A discussion of the pursuit of minimalism when building infrastructure systems, looking at the development of Heroku over time with tips for retiring technology, avoiding newly released software, non-invented-here and more.
https://brandur.org/minimalism

The recent release of Knative for Kubernetes features a number of components, one of which, Knative Build, can be used to run container builds on your Kubernetes cluster.
https://www.morethanseven.net/2018/07/29/knative-build-with-docker-buildkit-img/

How do you know how failures will affect your system? This post introduces the idea of a disaster recovery exercise and discusses the benefits for onboarding new people to on-call rotas.
https://hackernoon.com/how-to-lead-a-disaster-recovery-exercise-for-your-on-call-team-37b29f5f77c0

A look at different ops organisational structures in terms of incentives. I don’t agree with the terminology/definitions but the models and points are interesting.
https://medium.com/@skirsch/dev-vs-ops-devops-and-sre-f160dc6180f9

A collection of tips for managers looking to adopt devops practices into their organisations and teams.
https://enterprisersproject.com/article/2018/7/how-be-stronger-devops-leader-9-tips

A quick look at using pre-commit hooks to check for issues with Terraform code before commiting changes.
https://www.unixdaemon.net/tools/terraform-precommit-hooks/

An interesting post on the role of Kubernetes as an operational data plane, with CRDs as a critical piece. Some comparisons with previous attempts at the same thing with ESBs as a nice warning.
https://medium.com/@allingeek/kubernetes-as-a-common-ops-data-plane-f8f2cf40cd59

IIS is a powerful web application server popular on Windows, which makes it important to monitor. This set of posts explain the basic moving parts and how to instrument and monitor them.
https://www.datadoghq.com/blog/iis-metrics/
https://www.datadoghq.com/blog/iis-monitoring-tools/

A post on how to square strict conformance requirements like separation of duties and adopting automation and devops practices.
https://blog.xebialabs.com/2018/07/24/separation-of-duties-how-to-conform-in-a-devops-world/#.W1c3lY9Z7CI.twitter

Jobs

Engineering Manager, Operations and Infrastructure

FreeAgent is one of the leading online accounting platforms, loved by over 65,000 freelancers, small business owners and their accountants. Our Operations and Infrastructure team is focused on the hosting, deployment, service availability and scalability of the FreeAgent platform, implementing key architectural improvements, introducing Systems Reliability Engineering principles and anticipating and measuring scalability issues. You will lead a dedicated team of extremely smart and collaborative engineers, ensuring they’re well directed, motivated, productive and engaged in their work. This is a technical team leadership and people-management role.
https://www.freeagent.com/company/careers/

Events

Come to Velocity (September 30-October 3 in New York, NY) to learn the strategies and tools you need to make your systems fast, resilient, and secure. You’ll hear from industry leaders like Jaana Burcu Dogan (Google), Alice Goldfuss (GitHub), and Stephanie Hurlburt (Binomial), who can help you stay ahead of important trends in distributed systems, production engineering, software architecture, security, and more. Early Price ends August 17. Save up to $599 with code DOW20.
https://oreil.ly/2z1aE4z

Tools

A handy tool for finding security issues in source code repositories, in particular secrets or security keys accidentally checked in.
https://github.com/UKHomeOffice/repo-security-scanner

Argo CD is a new continuous deployment tool for Kubernetes. It supports deploying with Ksonnnet, Helm charts or standard Kubernetes configuration and integrates with a source code repository for the source of truth.
https://github.com/argoproj/argo-cd

Learn from failure to improve incident response. Start conducting post-incident reviews for deeper analyses and start making future incident remediation easier:
http://try.victorops.com/devopsweekly/post-incident-review