2 minute read

One topic that comes up regularly in this newsletter, including in this issue, is testing infrastructure. I’m doing a bit of research so if you’re written anything on infrastructure testing, or built any interesting tools, let me know at [email protected].


Devops Weekly is sponsored by Brightbox Cloud - serious UK-based cloud infrastructure from only 1.5p per hour (£10.95/month)

Start your £20 free trial now: http://brightbox.com/devopsweekly

Sponsored Event

[Webinar] 5 Key Metrics to Release Better Software Faster

No matter how often you deploy - or how sophisticated your delivery pipeline is - you need to measure the quality of your software. But what exactly should you measure? Learn the 5 key metrics that will increase your confidence in producing a safe build for production. Register now.


Some tips on how and why to implement a continuous delivery pipeline for your infrastructure. Particular useful due to the strong definitions.

A video from the recent OpenStack summit, entitled OpenStack is Doomed. Some details of how OpenStack got here and some suggestions for how to fix some of the problems. Useful for anyone interested in large projects.

8 tenets to consider when building or choosing software, focused on operational concerns.

Fedora is replacing the Yum package manager with the similar but newer DNF. If you’re using a RedHat based system, or targeting the latest version of Fedora something to be aware of.

Some interesting points about the explosion of applications within large organisations, and questions about how we can consider IT asset management along-side all the other topics being discussed on the technology or service management side.

A nice example of taking a previously manual process and automating it. In this case a risk calculation around deployment in a tightly regulated environment.

Intel’s Clear Linux containers are an interesting way of attempting to combine the best of virtual machines with the best of containers. This write-up shows you how to get up and running to test them out.

It’s easy to look at stories of practices at Etsy or Facebook and think your organisation can’t do that. This post explains how to get started with accelerating your own organization’s software delivery.

A useful run through the different components that make up a Graphite metrics install, along with alternatives or commonly used supporting software.

A good introduction to a new MySQL feature in 5.6, Performance Schema, which exposes performance metrics for queries as a table in the database. The post has nice examples of how to use the data and why it’s proved useful in production.


The concept of the blameless postmortem is a popular one in devops circles, but how do you conduct them so they become the best ways to learn from both failures and successes? This workshop with Dave Zwieback in Seattle on June 15th is a great start.


Percheron is a command line and YAML based tool for describing a stack of related Docker containers.

A handy command line tool for interacting with the Marathon API. The README shows a simple text format example for describing applications running in Marathon.

Devops Weekly is sponsored by Brightbox Cloud - serious UK-based cloud infrastructure from only 1.5p per hour (£10.95/month)

Start your £20 free trial now: http://brightbox.com/devopsweekly