2 minute read

It’s the week of KubeCon EU in Barcelona, if you happen to be heading along do say Hi. If not, I’ll try and collect as many links from the event as possible for the next few issues.

From our sponsor, VictorOps

[You’re Invited] Puppet, Splunk and VictorOps are teaming up for a live webinar on powering continuous improvement by combining analytics, incident response and automation. Learn best practices for releasing better applications faster, without the fire drills.
http://try.victorops.com/devopsweekly/continuous-improvement

News

The word devops, and the resulting community, started with Devopsdays nearly 10 years ago. The organisers are arranging a 10 year anniversary event; Devopsdays Gent 2019 is happening on October 29th and 30th in Gent. The CFP is open and early bird tickets are available now.
https://devopsdays.org/blog/2019/05/10/10-years-of-devopsdays/

A great introduction to eBPF, described here as bringing super powers to Linux administrators. A quick explanation of what it is, along with examples of what you can do with it.
https://speakerdeck.com/lizrice/ebpf-superpowers

We generally agree that collecting metrics is a good thing, but how do you decide what to measure? This presentation explores part of that question, looking at what vanity metrics are and why you should avoid them.
https://speakerdeck.com/stack72/mirror-mirror-on-the-way-what-is-the-vainest-metric-of-them-all

A high-level set of observations about trends in devops and cloud. Some clear takeaways around CI/CD adoption, chaos engineering and kubernetes, and some discussion amongst the authors.
https://www.infoq.com/articles/devops-cloud-trends-2019

An interesting collection of papers focused on software package management.
https://github.com/ipfs/package-managers/issues/56

A counterpoint to the argument that serverless represents a new type of lockin. Some good points, but I’d love to see posts with more historical context about lockin in IT.
https://lumigo.io/blog/you-are-wrong-about-serverless-vendor-lock-in/

A good discussion of the challenges of cluster lifecyle management for Kubernetes, and an introduction to the new Cluster API which has just hit v1alpha1 and provides a declarative API for managing Kubernetes clusters.
https://blogs.vmware.com/cloudnative/2019/05/14/cluster-api-kubernetes-lifecycle-management/

A quick reminder that web logs have lots of useful information in them, and that many of them can be configured to provide more with simple configuration. The post provides a few nginx tips, but I think this is true of most good web servers.
https://www.honeycomb.io/blog/tell-me-more-nginx/

New storage solutions keep cropping up that take advantage of modern cloud infrastructure. This post provides lots of details for anyone interested in, or operating, Ceph or Rook.
https://www.cloudops.com/2019/05/the-ultimate-rook-and-ceph-survival-guide/

Events

KubeCon and CloudNativeCon Europe are this week! Lots of folks are already onroute to Barcelona, with events running from May 20th until the 23rd. Expect a fantastic range of technical talks and an unparalleled opportunity to discuss all things modern infrastructure, cloud native and open source. Hopefully see a few folks there.
https://bit.ly/2VTfMl9

Tools

KEDA is a Kubernetes-based event driven autoscaling component. It allows for exposing information useful for scaling Kubernetes workloads up and down, as well as exposing that information via a metrics server.
https://github.com/kedacore/keda

[You’re Invited] Puppet, Splunk and VictorOps are teaming up for a live webinar on powering continuous improvement by combining analytics, incident response and automation. Learn best practices for releasing better applications faster, without the fire drills.
http://try.victorops.com/devopsweekly/continuous-improvement

Updated: