I’m in Austin all this week for KubeCon and CloudNativeCon so please say Hi if you happen to be here too. Lots of service management content this week, along with several posts on skills and team organisation.

Sponsor

Integrate VictorOps into DevOps workflows to deploy faster and maintain higher quality. Get your free trial started today:
http://try.victorops.com/DevOpsWeekly/FreeTrial

News

Knowing how and when to adopt new technology is a key skill given the potential improvements and risks associated with doing so. This post has some great tips for knowing when, and where, to say No.
https://blog.codeship.com/growing-tech-stack-say-no/

There are a number of different ways of deploying software, each with different tradeoffs. This post does a great job of describing several, giving them names and discussing pros and cons.
http://blog.itaysk.com/2017/11/20/deployment-strategies-defined

A nice set of habits, or principles, for anyone working as a site reliability engineer. A good discussion around the need for big-picture thinking, analysis, trust in process and more.
https://blog.newrelic.com/2017/11/13/site-reliability-engineer-sre-habits/

Everyone is probably familiar with the idea of a three-tier support structure, and some of the problems it can cause. But what alternatives exist? This post and talk introduce the concept of swarming for dealing with support issues.
https://medium.com/@JonHall_/itsm-devops-and-why-the-three-tier-structure-must-be-replaced-with-swarming-91e76ba22304
https://www.slideshare.net/JonHall7/it-in-the-park-2017-why-we-should-ditch-the-3tier-support-model-and-start-swarming

I’ve always been interested in the automation of local development environments. This post explains some of the benefits, and goes into the details of a node.js setup using Docker.
http://activelamp.com/blog/devops/docker-and-development-environments-setting-up-for-nodejs-on-osx/

A solid argument that sometimes focusing on bottom up change at the team level misses the larger challenge of how multiple teams and the rest of the organisation work together. Definitely relevant to any organisation adopting devops principles and practices.
https://hackernoon.com/stop-obsessing-over-the-teams-454fae465d63

A quick introduction to the concept of a service portfolio, why it can be a good way of managing priorities across a larger organisation, and how that fits with devops. I don’t agree with all of the points, but I do thing portfolios have a place.
https://itsm.tools/2017/11/28/four-reasons-need-service-portfolio-effective-devops/

A detailed walkthrough for setting up a CI pipeline for basic testing of ansible roles. I’d love to see more detailed assertions that just whether it runs cleany though.
https://blog.travis-ci.com/2017-11-30-testing-ansible-roles-using-docker-on-travis

High-level DSLs can provide a powerful way of managing configuration without requiring too much programming. This post introduces NaCl which is used for network rules for the includeOS firewall.
http://www.includeos.org/blog/2017/introducing-nacl.html

Tools

MetalLB is a Kubernetes-native load balancer for environments where the cloud-provider load balancer integration isn’t suitable, for instance when running a bare-metal cluster.
https://github.com/google/metallb

Integrate VictorOps into DevOps workflows to deploy faster and maintain higher quality. Get your free trial started today:
http://try.victorops.com/DevOpsWeekly/FreeTrial