A selection of posts this week on some of the harder non-technical aspects of devops; from sociotechnical systems to why the relationship between compliance and software development is often fraught. Plus posts on immutable infrastructure, automating the management of a large database cluster and more.

Sponsor

Learn how to accelerate your path to full-stack monitoring and alerting in this webinar. Register now:
http://try.victorops.com/DevOpsWeekly/ZenossWebinar

News

An excellent post on the divide between compliance teams and development teams in typical enterprise environments.
https://m.subbu.org/devops-and-governance-c9f2e73b7a70

An essay on the inter-relationship between technology and practice, explaining the field of sociotechnical systems and why particular technology choices won’t resolve wider cultural issues.
https://queue.acm.org/detail.cfm?id=3185224

A good post on building services which become part of the background, the importance of reliability and what UX research can learn from site reliability engineering.
https://medium.com/@aHev/why-ux-researchers-should-learn-sre-practices-a2b213e69a8a

A nice presentation on various common container patterns, from spoofing hosts to multi-stage builds and from sidecar containers to mounting sources at runtime. Nicely presented and explained with examples.
https://l0rd.github.io/containerspatterns/

Slides from my Configuration Management Camp talk, all about why application configuration should have a schema, and all the time and toil saving things we could do if it did.
https://speakerdeck.com/garethr/why-your-configuration-needs-a-schema

Another talks from Configuration Management Camp, on the reasons behind immutable infrastructure and how to adopt the pattern using LinuxKit.
https://www.cloudatomiclab.com/immutable-gent/

A deck covering how to sell serverless to your colleagues. Looking at inertia, the reality today vs potential in the future and embracing the service ecosystem.
https://www.slideshare.net/sallamar/selling-serverless-to-your-colleagues

A description of the automation behind managing a 150 node Cassandra clusters, using Chef, Rundeck and Terraform.
https://medium.com/teads-engineering/easy-cassandra-scaling-with-terraform-chef-rundeck-9443e0375aa7

A walkthrough of building a pipeline for building and testing a virtual machine, in this case using Packer, Ansible, InSpec and Pester.
https://medium.com/@trondhindenes/building-an-testing-vm-images-403d68b5eacf

A quick look at setting up a lab environment to test prometheus and grafana, using Docker Compose.
https://www.unixdaemon.net/sysadmin/prometheus-experiments-with-docker-compose/

When working with Kubernetes exposing a service to the outside world is a common requirement. This post covers how to go about doing so using Kubernetes Ingress.
http://container-solutions.com/production-ready-ingress-kubernetes/

Events

DotScale is back, bringing together speakers on devops, scalability and distributed systems in Paris from May 31st to June 1st. The organisers have kindly shared a 20% discount code for newsletters of DEVOPSWEEKLY.
https://2018.dotscale.io/tickets?promocode=DEVOPSWEEKLY

Tools

Alpinist provides a simple way of maintaining an Alpine APK package repository. Just upload packages to S3 and a Lambda function will build the repository metadata.
https://github.com/forward3d/alpinist

Learn how to accelerate your path to full-stack monitoring and alerting in this webinar. Register now:
http://try.victorops.com/DevOpsWeekly/ZenossWebinar