DevOpsCon 2017 "Continuous Delivery with Containers"
1 like681 views
The document discusses the impact of container technology on continuous delivery (CD), emphasizing that container images should serve as the 'single source of truth' within the pipeline. It highlights the importance of adding metadata to container images, ensuring a dependable testing environment, and maintaining good architectural practices. The speaker advises on using existing CD practices while adapting to new tools and stresses that not all developers are operationally aware.
DevOpsCon 2017 "Continuous Delivery with Containers"
- 1. Continuous Delivery with Containers: The Good, the Bad, and the Ugly Daniel Bryant @danielbryantuk
- 3. Setting the scene… • Continuous delivery is a large topic • PaaS and Serverless are super interesting… • But I’m assuming you’re all-in on containers • Focusing today on the process and tooling • No live coding today • Mini-book contains more details (thanks nginx!) 11/06/2017 @danielbryantuk http://bit.ly/2jWDSF7
- 4. TL;DR – Containers and CD • Container image becomes the build pipeline ‘single binary’ • Adding metadata to containers images is vital, but challenging • Must validate container constraints (NFRs) • Cultivate container ‘mechanical sympathy’ 11/06/2017 @danielbryantuk
- 5. @danielbryantuk • Independent Technical Consultant, CTO at SpectoLabs • Agile, architecture, CI/CD, DevOps • Java, Go, JS, microservices, cloud, containers • Leading change through the application of technology and teams 11/06/2017 @danielbryantuk
- 7. Continuous Delivery • Produce valuable and robust software in short cycles • Optimising for feedback and learning • Not (necessarily) Continuous Deployment 11/06/2017 @danielbryantuk
- 11. Container technology (and CD) • OS-level virtualisation • cgroups, namespaces, rootfs • Package and execute software • Container image == ‘single binary’ 11/06/2017 @danielbryantuk
- 16. Make your dev environment like production • Develop locally or copy/code in container • Must build/test containers locally • Perform (at least) happy path tests 11/06/2017 @danielbryantuk
- 18. Make your dev environment like production • Develop locally or copy/code in container • Must build/test containers locally • Perform (at least) happy path tests • Use identical base images from production • With same configuration 11/06/2017 @danielbryantuk
- 19. Lesson learned: Dockerfile content is super important • OS choice • Configuration • Build artifacts • Exposing ports • Java • JDK vs JRE and Oracle vs OpenJDK? • Golang • Statically compiled binary in scratch? • Python • Virtualenv? 11/06/2017 @danielbryantuk
- 21. Different test and prod containers? • Create “test” version of container • Full OS (e.g. Ubuntu) • Test tools and data • Easy to see app/configuration drift • Use test sidecar containers instead • ONTEST proposal by Alexi Ledenev 11/06/2017 @danielbryantuk http://blog.terranillius.com/post/docker_testing/
- 24. Building images with Jenkins • My report covers this • Build as usual… • Build Docker Image • Cloudbees Docker Build and Publish Plugin • Push image to registry 11/06/2017 @danielbryantuk
- 26. Metadata – Beware of “latest” Docker Tag • Beware of the ‘latest’ Docker tag • “Latest” simply means • the last build/tag that ran without a specific tag/version specified • Ignore “latest” tag • Version your tags, every time • danielbryantuk/test:2.4.1 11/06/2017 @danielbryantuk
- 27. Lesson learned: Metadata is valuable • Application metadata • Version / GIT SHA • Build metadata • Build date • Image name • Vendor • Quality metadata • QA control • Security audited etc 11/06/2017 @danielbryantuk
- 28. Metadata - Adding Labels at build time • Docker Labels • Add key/value data to image 11/06/2017 @danielbryantuk
- 29. Metadata - Adding Labels at build time • Microscaling Systems’ Makefile • Labelling automated builds on DockerHub (h/t Ross Fairbanks) • Create file ‘/hooks/build’ • label-schema.org • microbadger.com 11/06/2017 @danielbryantuk
- 30. Metadata - Adding Labels at runtime 11/06/2017 @danielbryantuk $ docker run -d --label uk.co.danielbryant.lbname=frontdoor nginx • Can ’docker commit’, but creates new image • Not possible to update running container • Docker Proposal: Update labels #21721
- 39. Testing NFRs in the build pipeline • Performance and Load testing • Gatling / jmeter • Flood.io • Security testing • Findsecbugs / OWASP Dependency check • Bdd-security (OWASP ZAP) / Arachni • Gauntlt / Serverspec • Docker Bench for Security / CoreOS Clair 11/06/2017 @danielbryantuk
- 40. Delaying NFRs to the ‘Last Responsible Moment’ • Newsflash! • Sometimes the last responsible moment is up-front! • Containers/microservices don’t make this easier • Sometimes more difficult… 11/06/2017 @danielbryantuk
- 41. Mechanical sympathy: Docker and Java • Watch for JVM cgroup/taskset awareness • getAvailableProcessors() may incorrectly report the number of cpus in Docker (JDK-8140793) • Runtime.availableProcessors() ignores Linux taskset command (JDK-6515172) • …because default fork/join thread pool sizes is based from host CPU count etc • Set container memory appropriately • JVM requirements = Heap size (Xmx) + Metaspace + JVM overhead • Account for native thread requirements e.g. thread stack size (Xss) • Entropy • Host entropy can soon be exhausted by crypto operations 11/06/2017 @danielbryantuk 41
- 49. In summary • Continuous delivery is vitally important in modern architectures/ops • Container images must be the (single) source of truth within pipeline • And metadata added as appropriate… • Mechanical sympathy is important (assert properties in the pipeline) • Not all developers are operationally aware • The tooling is now becoming stable/mature • We need to re-apply existing CD practices with new technologies/tooling 11/06/2017 @danielbryantuk
- 51. Thanks for listening • Any questions? • Feel free to contact me • @danielbryantuk • daniel.bryant@specto.io 11/06/2017 @danielbryantuk http://bit.ly/2jWDSF7
- 53. Containerise an existing (monolithic) app? • For • We know the monolith well • Allows homogenization of the pipeline and deployment platform • Can be a demonstrable win for tech and the business • Against • Can be difficult (100+ line scripts) • Often not designed for operation within containers, nor cloud native • Putting lipstick on a pig? 11/06/2017 @danielbryantuk
- 54. Key lessons learned • Conduct an architectural review • Architecture for Developers, by Simon Brown • Architecture Interview, by Susan Fowler • Look for data ingress/egress • File system access • Support resource constraints/transience • Optimise for quick startup and shutdown • Evaluate approach to concurrency • Store configuration (secrets) remotely 11/06/2017 @danielbryantuk