3 minute read

Karanbir Singh of the Centos project yesterday asked “Everyone wants to hire ops with 4+ yrs of experience. Where do they get the first 4 then ?” So started a conversation that spawned the #FirstTechJob hashtag which is still trending on Twitter. If you’ve not added your experience please do, and hopefully look for some analysis of the results in a future issue of the newsletter.


Devops Weekly is sponsored by Brightbox Cloud: the multi-zone cloud platform built for High Availability.

Try an SSD cloud server in 30 seconds with your £20 free credit…


7 Practices to Expand Performance and Effective Collaboration in DevOps

In DevOps, the product is only as good as the teamwork behind it. In this webcast Mark Tomlinson, performance engineering veteran and founder of the popular PerfBytes podcast, shares 7 practices to help you expand performance and effective collaboration into your DevOps team.


The start of an interesting looking series on applying some of the devops practices to the indie game development scene. The point about some industries having more room to innovate is well made, and it’s interesting to see successful practices applied outside that bubble.

An excellent detailed post looking at the specifics of the glibc vulnerability, and why it’s unusually bad. Some good points at the end of the post about the need for new approaches outside the way things are done today with vendors and researchers.

If devops involves all areas of an organisation, then making a compelling business case for implementing some of the practices is a likely first step in larger organisations. This post provides a clear guide on what to call out, and what to show to make the case.

With more security announcements this week several posts touch on the impact of security disclosures to how people are running containers. This post makes the case that simply scanning containers isn’t enough.

In a similar vein, this post explores the use of OverlayFS and LXC to make updating system dependencies in containers easier - focusing on the operational benefits.

Related, but making a different argument. This post looks at the advantage of using much smaller operating systems for containers, specifically looking at Alpine linux. Focuses mainly on the performance and complexity advantages.

A breakdown of 13 properties of a good build system, from performance to usability via repeatability and the ability to version configuration. Lots of good advice for anyone looking at products.

A handy tip for anyone using a WIndows machine. Git ships with a full SSH client on windows, which means you don’t need a separate terminal for connecting to remote nix systems.

A good list of things that traditional ITSM approaches need to adapt to, including the growing focus on security, software defined infrastructure, a wider range of often smaller suppliers and more.

A post introducing the concept of chaos engineering, and laying down a series of principles for those starting out adopting these ideas.

A useful overview of one way of securing credentials used to run Terraform locally, using GPG and passwordstore.

A useful summary of introducing security checks into a development pipeline, looking at both static and dynamic checks, and talking a little about who writes those tests.

Kubernetes provides a solid set of primitives for building distributed systems, which makes it an ideal platform for people interested in building higher-level interfaces. This post explores why, by showing how easy it is to built a simple PaaS on top of Kubernetes API.

A case for tracking work in progress and throughput as the main metrics when it comes to practices devops. Useful, although I think this focuses a little too much on the argument that devops is only about speed.


Want to work on new projects? Try a new stack? Join Hired and discover new opportunities today.


Devopsdays London is back on the 19th and 20th of April, with the topic of Devops in the Enterprise, and in particular in financial services. This type of focus should make for a great event. Tickets are on sale now.


Solo5 is a new base for unikernels, specifically providing a base to run other stacks on QEMU/KVM. The repository contains an example of running Mirage under KVM rather than Xen.

Linkerd is described as a Modern RPC proxy for microservices. It uses Finangle and supports runtime traffic routing, load balancing, service discovery and built-in instrumentation,

Devops Weekly is sponsored by Brightbox Cloud: the multi-zone cloud platform built for High Availability.

Try an SSD cloud server in 30 seconds with your £20 free credit…