DEVOPS WEEKLY ISSUE #394 - 15th July 2018
A fairly packed issue this week, with everything from moving back to monolithic applications, why tickets are the wrong way to manage IT operations, examples of CI/CD workflows and AWS Fargate configurations, some nifty ideas for security tooling for users of GitHub and more.
Sponsor
Take monitoring data and turn it into actionable alerts and insights. See how you can aggregate monitoring data and prioritize alerts to create visibility and improve workflows:
http://try.victorops.com/devopsweekly/monitoring-visibility
News
Tickets are the universal way of approaching operations work, but have lots of downsides. Worse, their existence is rarely considered. This post looks at approaches of reducing the number of ticket queues in your organisation.
https://www.rundeck.com/blog/tickets_make_operations_unnecessarily_miserable
A solid post about moving from microservices to a monolithic application and a monorepo. Good arguments for the growing complexity in certain architectures.
https://segment.com/blog/goodbye-microservices/
Kubernetes is part of the the puzzle of running applications, but how do you build an effective developer workflow around it? This slide deck explores patterns and problems to solve for.
https://www.slideshare.net/dbryant_uk/cncf-webinar-series-creating-an-effective-developer-experience-on-kubernetes
I think I disagree with the categorisation, but this post is a good exploration of some of the different approaches to platform management.
https://stacksense.io/krishnan/platforms/three-approaches-to-cloud-native-platforms/
A post looking at the differences between test infrastructure and build infrastructure. I’ve definitely seen this distinction first hand with certain types of project and I think it’s a good lens through which to view certain tools.
https://medium.com/@thomas.shaw78/splitting-build-and-test-infrastructure-d10a03f9c1f9
A look at consuming the event data from GitHub for later analysis, in this case when investigating potential security issues related to source code and interaction with it.
https://gdstechnology.blog.gov.uk/2018/07/13/building-a-tool-to-improve-our-github-security/
A walkthrough of setting up a highly available application, in this case a Grafana dashboard, when running on Fargate in AWS.
https://medium.com/247sports-engineering/highly-available-grafana-running-on-aws-fargate-and-rds-aurora-2bbea9d41b50
An in-depth good, bad and ugly analysis of Google Cloud Platform as it stands today. Useful for anyone looking at adopting any GCP services.
https://www.deps.co/blog/google-cloud-platform-good-bad-ugly/
A nice walkthrough of a continuous delivery setup using Jenkins, Docker, Kubernetes, BitBucket, Vault and more.
https://blog.apptension.com/2018/07/11/setup-software-deployment/
A quick comparison of the JIB build tool, for building Docker images for Java applications, with existing approaches like the Docker Maven plugin.
https://ro14nd.de/jib-vs-dmp
Events
| DevOps at Open Source Summit | August 29-31, Vancouver, BC |
The Open Source Summit offers DevOps professionals a unique opportunity to learn how to more efficiently manage environments, migrate legacy infrastructure, integrate the latest tools, and much more. Network with your community peers in the hallway track, or get educated on cutting edge technology - like Serverless and Kubernetes - in the 250+ sessions. Whether you’re interested in DevSecOps, infrastructure automation or container testing and orchestration tools - you will have the chance to learn about the latest advances and best practices.
View the schedule, and save 15% with discount code OSSDEVOPS15.
http://bit.ly/open-source-summit-2018
Tools
XAR lets you package many files into a single self-contained executable file which makes it easy to distribute and install. Impressive examples of using this to package standalone Python applications with fast boot times.
https://github.com/facebookincubator/xar/
https://code.fb.com/data-infrastructure/xars-a-more-efficient-open-source-system-for-self-contained-executables/
Dragonfly looks like an interesting building block. It’s an intelligent P2P based file distribution system which aims to resolve issues related to low-efficiency, low-success rate and waste of network bandwidth in file transferring process. Useful in large-scale file distribution scenarios such as application distribution, cache distribution, log distribution and image distribution.
https://github.com/alibaba/Dragonfly
runq is a hypervisor-based Docker runtime based on runc to run regular Docker images in a lightweight KVM/Qemu virtual machine.
https://github.com/gotoz/runq
Take monitoring data and turn it into actionable alerts and insights. See how you can aggregate monitoring data and prioritize alerts to create visibility and improve workflows:
http://try.victorops.com/devopsweekly/monitoring-visibility