FOSDEM for the last few days and then Configuration Management Camp tomorrow and Tuesday should mean lots of content for next weekends issue. For this week we have talk of monitoring standards, service discovery, talking continuous deployment for business and profiling unikernels.

Sponsor

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…
http://brightbox.com/devopsweekly

Sponsored

[OnDemand Webinar] Managing Technical Debt While Innovating on DevOps

How do you “keep the lights on” for your legacy systems, while also enabling your DevOps teams to launch new features faster? Watch this webinar to learn tips to manage technical debt in legacy code, plus efficient ways to close loops while enabling DevOps to optimize innovation and releases.
http://ow.ly/U3Jqn

News

An excellent explanation of why continuous deployment matters to business. Includes well reasoned lists for advantages for both developers and business folks, ideal if you’re trying to convince your organisation to adopt.
https://medium.com/@crowquine/why-continuous-deployment-matters-to-business-6a79b5602145#.wcei2lsoi

An excellent slide deck and video from a talk from the recent Scale14x conference, all about adopting service discovery, and the various problems encountered and overcome as they scaled usage.
https://blog.froese.org/2016/01/23/service-discovery-in-the-cloud-with-consul/

I’ve spoken previously about the importance of standards emerging, in particular around monitoring. So OpenTracing looks like a great endeavour, aiming to offer consistent, expressive, vendor-neutral APIs for popular tracing platforms.
http://opentracing.io/

There has been some conversations about what debugging might look like for Unikernel based systems recently. This post takes a deep dive into profiling Mirage, both as a unix binary using standard tools and a proof-of-concept flame graph for MIrage running on Xen.
http://www.brendangregg.com/blog/2016-01-27/unikernel-profiling-from-dom0.html

A nice early experience report of running Kubernetes in production. Some points about the use of environments and namespaces, monitoring and deployment workflows.
http://labs.unacast.com/2016/01/27/three-lessons-from-running-k8s-in-production/

FOSDEM spans two days, at least 20 tracks and a huge number of talks so consider this just a sample, but these raw notes give you a good idea of how some of the talks went this year.
http://tech.mattbostock.com/2016/01/30/fosdem-day-one/
http://tech.mattbostock.com/2016/01/31/fosdem-day-two/

Managing public keys in a large infrastructure is a subtle problem, and this post does a good job of asking and answering some of the common questions.
http://tfitch.com/chef/artifactory/2016/01/26/remote-copying-of-generated-keys.html

I’m not sold on the term microcontainer, but this post is a good introduction to the use and advantages of smaller base operating system images like Alpine for Docker containers.
http://www.iron.io/blog/2016/01/microcontainers-tiny-portable-containers.html

An exploration of what an end-to-end pipeline might look like for a docker based workflow. Covering everything from running tests to static analysis, a good reference for related tools.
http://capgemini.github.io/devops/docker-ci-workflows/

Jobs

DemystData is hiring an Infrastructure Automation Engineer to join our small, but growing, engineering team. Demyst is a financial technology startup headquartered in New York City. Candidates must have extensive AWS experience, be fully on board with Infrastructure as Code, and be security-minded. This position is full-time remote.
https://demystdata.workable.com/j/89E56BF9EA

We’re looking for a versatile DevOps engineer to work primarily in Linux using Docker, Terraform, Chef and AWS. Our team is seasoned, humble, and forward thinking. We’re using the latest tools to deliver a 100% uptime clustered web application across many production environments. We value continuous integration, high-quality testing, and infrastructure automation so that we can focus on building new features, not mucking with overhead.

Stormpath is an authentication and user management service that helps developers quickly and securely create web and mobile applications. We hire smart, fun, and passionate people who love what they do and deliver quality work. Interested in learning more?
https://boards.greenhouse.io/stormpath/jobs/158512#.VqlnlFMrLOQ

Tools

Builds servers can become complex quickly, sometimes because the problem you’re trying to solve is itself complicated. But if you have highly opinionated (git and GitHub only) and simpler CI requirements then Surf might be worth a look.
https://github.com/surf-build/surf

Goad is a nice demonstration of building a useful tool using Apache Lambda. It’s a Go based distributed load testing tool that takes advantage of the serverless architecture of Lambda to spin up on demand.
https://goad.io/

Keeping up with what service is running on what port is always tricky, and the commands to get the information tend to be somewhat cryptic. Enter whatportis, a simple interface for getting the ports in use by a service, or the service running on a specified port.
https://github.com/ncrocfer/whatportis

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…
http://brightbox.com/devopsweekly