2 minute read

Lots of talk of testing and of organisation design at the moment in the devops community, at least based on posts I read this week. Also I’m seeing more people talk about how we change to adopt devops practices, rather than just what those practices are. Anyone with a good stories please write them up and send me the links.


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

#ChefConf is less than a month away and the workshops are filling up fast, get your ticket today! Save 10% with discount code DEVOPSWEEKLY


I’ve mentioned serverspec quite a lot recently and here’s another post with lots of good advanced tips and examples. The approach to listing hosts is a nice counter example to the default directory setup.

For all statsd’s simplicity it’s quite a powerful little tool for collecting metrics. This post does a good job of explaining what it’s actually doing and includes a few common misunderstandings and gotchas.

Lots of devops is about communication and sharing, and in most organisations this is often driven by the structure of that organisation. This post makes a number of good points about organisation design and provides a number of examples of team composition.

Scale Summit was a one day unconference held in London last week, all about operating and building systems at scale. Lots of topics were discussed and this post has detailed notes about some of them. Monitoring, mentoring, what happened in the last year and a run down of the lightning talk topics.

Interesting observations from attending an ITIL event followed by a Devops event. The call for the devops community to focus more on how to change, as well as what to change to, is a good one.

One of the advantages of continuous integration is having one shared place to do things - for this next post that’s to collect useful metrics. Quite a lot of detail about the types quality metrics you can collect as part of a pipeline, including some examples of why it’s a useful idea.

Adding performance testing to continuous integration pipelines can be tricky - deciding when to run them, for how long and determining what to assert on isn’t as simple as lots of other types of testing. This presentation shows an example from one organisation doing this successfully.


Skyscape Cloud Services is recruiting a whole host of developer, DevOps and Platform-as-a-Service roles as it quickly expands its footprint in the UK Public Sector. It’s the place to be if you want to help build, automate and scale a massive secure cloud platform or help introduce new products based on Cloud Foundry.


Flowcon, the continuous delivery conference, is in September in San Francisco. The call for proposals closes tonight at midnight PST and the organisers are looking for talks on using lean product development and adopting continuous design, delivery, and devops.


A very handy tool for anyone using Nginx, ngxtop will give you a real time view of traffic, popular URLs, status codes, requests per second, popular source addresses and more. See the sample output for examples.

Graphite-api is a minimalistic API server that replicates the behavior of the graphite web interface. Useful for building dashboards or other tooling on top of, without having to run a database or larger web application.