DEVOPS WEEKLY ISSUE #417 - 23rd December 2018

2 minute read

Coming quickly to the end of 2018, if anyone has any summary blog posts or predictions for next year in the world of devops please let me know.

From our sponsor, Victorops

A DevOps-oriented organization focused on collaboration and transparency will drastically improve incident response and remediation speed. Learn all about the importance of building reliable services through a DevOps culture:
http://try.victorops.com/devopsweekly/why-devops-matters-guide

News

Having a full Private Key Infrastructure (PKI) setup has lots of advantages for securing infrastructure, but is typically regarded as a lot to understand. This post does a great job of explaining all the concepts, along with examples of the involved artefacts.
https://smallstep.com/blog/everything-pki.html

It’s always worth considering the business drivers behind any technology programme, especially one as pervasive as cloud adoption. This post, the first in a series, looks at why and when not to opt for a cloud migration.
https://container-solutions.com/whats-in-it-for-me-whats-the-business-case-for-cloud-native/

The State of Open Source Security Survey is open at the moment, if you’re using open source tools at all then please take a moment to complete it. I’ll post a link to the results when they’re published too.
https://snykteam.typeform.com/to/NaTJFW

A good list for anyone adopting devops practices. Part strategy, part tactics, this covers quite succinctly things like a maturity model and set of specific practices for evolving how an organisation runs software systems.
https://github.com/jdumars/agileops

A writeup of moving to adopt the envoy proxy application. A practical look at the advantages of service mesh technologies, and how to move from existing tools over time.
https://redditblog.com/2018/12/18/envoy-proxy-at-reddit/

Another detailed migration story, this time discussing moving from MongoDB to PostgreSQL. Lots of details about technology and architecture choices and tradeoffs.
https://www.theguardian.com/info/2018/nov/30/bye-bye-mongo-hello-postgres

A detailed analysis of the serverless space using an approach called Emerging Technology Analysis Canvas. It includes a little on definition but also delves into potential impact, ease of adoption, potential risks and likely timelines.
https://github.com/wso2/ETAC/blob/master/outlooks/serverless_outlook.md

Lots of interest at the moment in tools to bridge the local development loop with a Kubernetes cluster. This post explores one of those tools, Tilt, and shows how it’s enabling a fast feedback loop even with the cluster in the mix.
https://medium.com/@dbentley/28ddffe2d79f

Not quite the end of the year yet but here’s one prediction post for 2019. More consolidation, China and AI, multi-platform and more adoption amongst the largest organisations.
https://medium.com/@ab415/2019-the-year-lightning-strikes-the-cloud-market-6f5dc0683b0d

An interesting take on project management and system safety, as well as blame and understanding of complex systems.
https://itrevolution.com/project-product-software-safety-getting-plane/

Tools

Airship is a nice looking Terraform module distribution which provides a set of modules for creating a fully-managed AWS ECS cluster.
http://airship.tf/

Cloud Native Development (CND) is another inner-loop development tool which automatically syncs your local code with a Kubernetes cluster, minimising the manual commands you need to run when building your app.
https://github.com/okteto/cnd

fx is a new CLI parsing tool for JSON. Similar to jq, it supports using Javascript to write extensions and simple parsing programmes. Lots of interesting examples in this post.
https://medium.com/@antonmedv/discover-how-to-use-fx-effectively-668845d2a4ea

KubeDB is a high-level project providing a CLI tool and Kubernetes operator for running databases (including Redis, MongoDB, PostgreSQL, MySQL, ElasticSearch and Memcached) on your cluster.
https://kubedb.com/

A DevOps-oriented organization focused on collaboration and transparency will drastically improve incident response and remediation speed. Learn all about the importance of building reliable services through a DevOps culture:
http://try.victorops.com/devopsweekly/why-devops-matters-guide

Updated: