DEVOPS WEEKLY ISSUE #464 - 17th November 2019

1 minute read

It’s the weekend before KubeCon, which means the next issue just might have far too much Kubernetes content for some folks. Fair warning. This issue we have discussions of pipelines as code, runbooks, resilience engineering, insecure development environments and more.

From our sponsor, VictorOps

As DevOps and IT teams ingest more alerts and respond to more incidents, they collect more information and historical context. Today, teams are using this data to optimize incident response through constant automation and machine learning. Learn how:
https://go.victorops.com/devopsweekly-incident-response-automation-and-machine-learning

News

A good argument that having pipeline configuration files in each source repo works well for testing, but less well for complex deployments.
https://medium.com/@raxwunter/whats-wrong-with-pipeline-as-code-or-why-i-don-t-like-gitlab-ca70c39eb52b

A useful post on the importance of thinking about the problems with unsecured developer user access and insecure development environments.
https://www.fugue.co/blog/unsecured-dev-accounts-put-production-at-risk

Solid advice on what makes a good runbook; actionable, accessible, accurate, authoritative and adaptable.
https://www.transposit.com/blog/2019.11.14-what-makes-a-good-runbook/

A look at different failure modes in complex systems, and how to ensure a continuously resilient system. Introduces the failure modes and effects analysis methodology.
https://medium.com/@adrianco/failure-modes-and-continuous-resilience-6553078caad5

A post looking at what we mean by portability, compatibility and supportability when it comes to OCI container images.
http://crunchtools.com/deeply-understanding-the-different-between-portability-compatibility-and-supportability/

An interesting talk looking at service mesh capabilities and in particularly how much of the functionality already resides inside Kubernetes. Some good thoughts about the future direction of the platform too.
https://speakerdeck.com/thockin/weve-made-quite-a-mesh

As you become more of an advanced user of Kubernetes it’s common to want to write your own controllers. This post explores building a controller using Buck, and integrating it with Brigade and Porter and other component tools.
https://cloudblogs.microsoft.com/opensource/2019/11/15/in-cluster-cnab-management-brigade/
https://github.com/brigadecore/buck

A useful getting started guide for NATs, the high performance messaging system.
https://itnext.io/getting-started-with-nats-b752cbb17f74

Tools

Comby is a powerful tool for manipulating source code, suited to repository-wide refactoring and changes.
https://comby.dev/

As DevOps and IT teams ingest more alerts and respond to more incidents, they collect more information and historical context. Today, teams are using this data to optimize incident response through constant automation and machine learning. Learn how:
https://go.victorops.com/devopsweekly-incident-response-automation-and-machine-learning

Updated: