DEVOPS WEEKLY ISSUE #233 - 21st June 2015
With monitorama on this last week I have lots of monitoring content this week, including some interesting presentations, a few tools and a reminder to fill out the 2015 monitoring survey.
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 Event
DevOps: From Adoption to Performance
DevOps isn’t just for the “Unicorns.” Join us as Gene Kim explains how you can replicate the DevOps practices & successes of the so-called “Unicorns.” Discover the lessons he’s learned from studying these companies and how you can apply them at your company!
http://ow.ly/Lr1oo
News
With all the monitoring content in this issue it’s the perfect time to remind readers to take the 2015 Monitoring Survey.
http://kartar.net/2015/06/monitoring-survey-2015/
https://www.surveymonkey.com/s/monitoringsurvey2015
An excellent talk on the utility and importance of user research, in particular applied to making GitHub a better user experience.
https://speakerdeck.com/chrissiebrodigan/measuring-hard-to-measure-things
A good article on the curl-pipe-sudo-bash installation pattern, why it exists and what problems lead to it being used.
http://blog.digital-scurf.org/posts/in-defence-of-curl-pipe-sudo-bash/
A good reminder that monitoring is never done. I particularly liked the observation that everyone wants monitoring so they can say “it’s monitored”.
http://www.slideshare.net/melaniemj/monitoring-is-never-done
A good post explaining some of the potential in software defined networks, and a nice example of using Docker to build a powerful cloud bandwidth monitoring suite.
http://networkstatic.net/building-network-tools-using-docker/
https://github.com/nerdalert/cloud-bandwidth
Metrics are an important part of adopting devops, but determining what to measure when it comes to less technical subjects is tricky. This is a good post about metrics focused around incident management and wy to measure certain things.
https://www.pagerduty.com/blog/best-metrics-devops-culture/
A talk from Monitorama last week, initially looking at what the presenter wants and doesn’t want out of a modern monitoring setup and then providing pointers to some low-level tools coming in the future.
http://www.slideshare.net/brendangregg/monitorama-2015-netflix-instance-analysis
The last of the monitorama presentations, this is an excellent talk on running a game day to help plan and simulate dealing with real production outages.
http://www.daniellesucher.com/presentations/gamedays/#/
An interesting article looking at language choices for systems management applications. Based on a few personal opinions but some useful observations in there.
http://opsrevolution.com/language-choices-in-systems-management-apps/
A nice writeup of talks at the recent docscale conference. Only brief overviews of each talk but nice to see the range of topics covered.
http://marmelab.com/blog/2015/06/10/dotscale-2015-conference-summary.html
Jobs
SignalFx is an advanced monitoring platform for modern applications built on streaming analytics, offered as SaaS. We’re looking for SREs and Infrastructure Engineers to support the rapid expansion of our platform on multiple IaaS providers.
signalfx.com/careers.html or reach out to us at [email protected].
We are looking for a Senior DevOps Automation Engineer to join our Technology team at Hailo, ready to improve their skills, learning Go, and working in one of the coolest and most advanced platforms in the market.
https://hailo.workable.com/jobs/71676
Events
Devopsdays Chicago is coming up on the 25th and 26th of August and the CFP is currently open, closing on June 30th. The CFP site has several ideas for topics to cover too.
http://www.devopsdays.org/events/2015-chicago/
https://devopsdayschicfp-app.herokuapp.com/events/devopsdays-chicago-2015
Tools
Mesos-consul automatically registers/deregisters services run as Mesos tasks, so you can use consul for service discovery purposes in a Mesos cluster.
https://github.com/CiscoCloud/mesos-consul
Brubeck is a statsd compatible stats aggregator written in C. The initial post explains the problems with statsd, and in particular the use of UDP for monitoring.
http://githubengineering.com/brubeck/
https://github.com/github/brubeck
CollectdWin a system statistics collection daemon for Windows, inspired by ‘collectd’.
https://github.com/bloomberg/collectdwin
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