[20200720]cloud native develoment - Nelson Lin
- 1. Cloud Native Development Cloud code for example 20.07.2020 Nelson
- 2. Agenda Cloud-native & kubernetes Why do we develop on kubernetes? Difficulty and solution Cloud code and related components
- 3. About Nelson Lin Cloud Technical Account Manager in Red Hat. Help accounts adopt Red Hat cloud technologies well and help them succeed. Software engineer, building a cloud management system, experienced of frontend and backend programming having experience in CI/CD pipeline, interested in big data, distributed computing, and cloud-native technologies.
- 4. Cloud Native & Kuberenetes
- 5. Cloud-native definition ● An approach exploits the advantages of the cloud computing delivery model to building and running applications ● How applications are created and deployed, ● Cloud native is all about changing the way you think about constructing critical business systems. ● Cloud-native systems are designed to embrace rapid change, large scale, and resilience. https://tanzu.vmware.com/cloud-native https://docs.microsoft.com/en-us/dotnet/architecture/cloud-native/definition https://zh.data-driven.blog/2019/11/01/%E4%BB%80%E9%BA%BC%E6%98%AF-cloud-native%EF%BC%9F-%E7%82%BA%E4%BB%80%E9%BA%BC-cloud-native- %E9%80%99%E9%BA%BC%E7%86%B1%E9%96%80%EF%BC%9F/
- 6. Cloud-native foundational pillars While several container orchestrators exist, Kubernetes has become the de facto standard for the cloud-native world. It's a portable, extensible, open- source platform for managing containerized workloads.
- 7. Why Do We Develop On Kubernetes
- 9. Let’s go dive deeper
- 10. Let’s go dive deeper Development phase takes over 65%of your SDLC
- 11. Containerization/Cloud-nativeVirtualization(XaaS) From convention to virtualization and cloud-native Single-box
- 13. Developing k8s/cloud-native app. is frustrating Too much configuration Debugging is a pain Applications have so many moving parts Spend too much time outside of the codes
- 14. Infinite loop of pain & suffering Even “Hello, World” is complicated! Error prone People solve that with bash scripts, makefiles and ansible 1 4 2 53 6 Change code Run docker build Run docker push Patch yaml Run kubectl apply Verify 7 debugging
- 15. Challenge Config Chaos Slow & Error-prone Repetitive tasks Hard to debug Logging way
- 16. Here Comes Cloud Code
- 17. Cloud Code Set of development tools
- 18. Cloud Code IDE Skaffold JIB Cloud Code ● support for the full development cycle of Kubernetes applications ● create a cluster for development ● test running your finished application ● Build & deploy ● Tag & Test & cleanup ● File sync ● Log tailing ● Port forwarding https://cloud.google.com/code/docs/vscode/quickstart https://cloud.google.com/code/docs/intellij/quickstart-IDEA Out-of-the-box configuration snippets, tailored debugging experience Making developing with Kubernetes a whole lot easier
- 19. Jib is a set of plugins for maven and gradle for building optimized OCI-compliant container images for Java applications without a Docker daemon. JIB ● Fast ○ separates applications into multiple layers ○ splitting dependencies from classes ● Daemon-less ○ No need of docker CLI or counterparts ○ No dockerfile ● Reproducible https://github.com/GoogleContainerTools/jib/issues/101 https://github.com/GoogleContainerTools/jib
- 20. The flow difference Docker build flow JIB build flow
- 21. JIB makes build images and transfer them efficiently 200MB 2MB Enhanced by 10,000%
- 22. Skaffold Skaffold as a Kubernetes-native framework for developer workflows, portable to any IDE or CI system. It provides a very easy way to test local changes to your app on a Kubernetes cluster. And it supports multiple image builders (local Docker, GCB & Kaniko).
- 23. Build ● Dockerfile ● Bazel ● Jib ● buildpack That’s Skaffold Adapts to your tooling local + remote clusters Test ● container- structure- test Tag ● Git ● Sha256 ● Env Template ● DateTime Deploy ● Kubectl ● Helm ● kustomize
- 25. Skaffold debugger ● Built artifacts determine the underlying runtime technologies ● Artifacts transformed on-the-fly runtime’s debugger functionality ● Supported Language ○ Go ○ NodeJS ○ Java & JVM Language ○ Python ● Runtime specific images ○ .net Core https://skaffold.dev/docs/workflows/debug/
- 26. Skaffold pipeline stages Skaffold Pipeline stages Description Init generate a starting point for Skaffold configuration Build build images with different builders Tag tag images based on different policies Test test images with structure tests Deploy deploy with kubectl, kustomize or helm File Sync sync changed files directly to containers Log Tailing tail logs from workloads Port Forwarding forward ports from services and arbitrary resources to localhost Cleanup cleanup manifests and images https://skaffold.dev/docs/references/yaml/
- 27. After developing, GitOps plays the next stage.
- 28. Cloud Code for IDE plugins Step 1 Getting Started Step 2 Local Development Step 3 Deploy to remote cluster Step 4 Troubleshooting Debugging & logs
- 29. Private image registry https://kubernetes.io/docs/tasks/configure-pod-container/pull- image-private-registry/ Then #kubectl create secret docker-registry --docker-server=<uri> --docker-username=<id> --docker-password=<password> <name> pod/dc/deployment/ defines spec. spec.imagePullSecrets.name = <prior to name>
- 30. Workflow of cloud code ui(nodejs) title(java) game(go) Laptop Remote Cluster ui:v1 title:v1 game:v1 watch & sync watch & mvn jibBuild watch & docker build port-forwarding http://localhost:8080
- 31. Let’s Demo
- 32. Demo introduction nodejs-backend nodejs-frontend nodejs-backend nodejs-frontend monogo svc/mongo svc/frontend svc/backend backend frontend monogo svc/mongo svc/frontend svc/backend backend frontend secret/quay secret/quay docker-desktop OCP Port-forward & log Port-forward & log deploy deploy push push RemoteLocal intellij vscode
- 33. Show my screen
- 34. Problem No enough resources No managed services Changing IDE is arduous Have to learn new skills Support few tools How integrate with you existing circumstances/pipelines.
- 35. Take away Let your code deployed on kubernetes as early as possible if you want to run on K8S Cloud code is just a tool, but follow the concept and try to develop as cloud-native way ( odo, crc, ocp and acm) Try to reduce the gap between deployment engineer and software engineer With advent of k8s, go learn a new skill - k8s
- 36. Thank you and question?
Editor's Notes
- #16: Repetitive tasks of building, pushing and deploying images Difficult to set up debugging against a k8s