Tekton实战案例--S2I

案例环境说明

• 示例项目：

  代码仓库：https://gitee.com/mageedu/spring-boot-helloWorld.git

  构建工具maven

• pipeline各Task

  • git-clone：克隆项目的源代码

  • build-to-package: 代码测试，构建和打包

  • generate-build-id：生成build id

  • image-build-and-push：镜像构建和推送

  • deploy-to-cluster：将新版本的镜像部署到kubernetes集群

• Workspace

  • 基于PVC，跨task数据共享

2.2.5.2 pipeline完成Image构建，推送和部署

1. 01-git-clone的Task

   apiVersion: tekton.dev/v1beta1
   kind: Task
   metadata:name: git-clone
   spec:description: Clone code to the workspaceparams:- name: urltype: stringdescription: git url to clonedefault: ""- name: branchtype: stringdescription: git branch to checkoutdefault: "main"workspaces:- name: sourcedescription: The code repo will clone in the workspacesteps:- name: git-cloneimage: alpine/git:v2.36.1script: git clone -b $(params.branch) -v $(params.url) $(workspaces.source.path)/source

2. 02–build-to-package.yaml

   apiVersion: tekton.dev/v1beta1
   kind: Task
   metadata:name: build-to-package
   spec:workspaces:- name: sourcedescription: The code repo in the workspacessteps:- name: buildimage: maven:3.8-openjdk-11-slimworkingDir: $(workspaces.source.path)/sourcevolumeMounts:- name: m2mountPath: /root/.m2script: mvn clean install# 定义volume提供maven cache，但是前提得创建出来maven-cache的pvcvolumes:- name: m2persistentVolumeClaim:claimName: maven-cache
   

3. 03-generate-build-id.yaml

   apiVersion: tekton.dev/v1beta1
   kind: Task
   metadata:name: generate-build-id
   spec:params:- name: versiondescription: The version of the applicationtype: stringresults:- name: datetimedescription: The current date and time- name: buildIddescription: The build IDsteps:- name: generate-datetimeimage: ikubernetes/admin-box:v1.2script: |#!/usr/bin/env bashdatetime=`date +%Y%m%d-%H%M%S`echo -n ${datetime} | tee $(results.datetime.path)- name: generate-buildidimage: ikubernetes/admin-box:v1.2script: |#!/usr/bin/env bashbuildDatetime=`cat $(results.datetime.path)`buildId=$(params.version)-${buildDatetime}echo -n ${buildId} | tee $(results.buildId.path)

4. 04-build-image-push.yaml

   要想能推送镜像到镜像仓库，必须创建一个secret对象，挂在到kaniko的/kaniko/.docker目录下，具体创建secret的方法有两种：

   1、先在一台机器上login镜像仓库，这里以dockerhub为例，将会把认证文件保存在~/.docker/config.json:
   

5. 基于config,json创建sectet，这里的secret的类型选择generic

   kubectl create secret generic docker-config --from-file=/root/.docker/config.json
   

   2、先基于user/password创建一个base64：

   echo -n USER:PASSWORD | base64
   

   创建一个config.json，然后将创建出来的base64替换到下面xxxxxxxxxxxxxxx

   {"auths": {"https://index.docker.io/v1/": {"auth": "xxxxxxxxxxxxxxx"}}
   }
   

   最后创建一个secret

   kubectl create secret generic docker-config --from-file=<path to .docker/config.json>
   

6. 05-deploy-task.yaml

   apiVersion: tekton.dev/v1beta1
   kind: Task
   metadata:name: deploy-using-kubectl
   spec:workspaces:- name: sourcedescription: The git repoparams:- name: deploy-config-filedescription: The path to the yaml file to deploy within the git source- name: image-urldescription: Image name including repository- name: image-tagdescription: Image tagsteps:- name: update-yamlimage: alpine:3.16command: ["sed"]args:- "-i"- "-e"- "s@__IMAGE__@$(params.image-url):$(params.image-tag)@g"- "$(workspaces.source.path)/source/deploy/$(params.deploy-config-file)"- name: run-kubectlimage: lachlanevenson/k8s-kubectlcommand: ["kubectl"]args:- "apply"- "-f"- "$(workspaces.source.path)/source/deploy/$(params.deploy-config-file)"
   

7. 06-pipelinerun-s2i.yaml

   apiVersion: tekton.dev/v1beta1
   kind: Pipeline
   metadata:name: source-to-image
   spec:params:- name: git-url- name: pathToContextdescription: The path to the build context, used by Kaniko - within the workspacedefault: .- name: image-urldescription: Url of image repository- name: deploy-config-filedescription: The path to the yaml file to deploy within the git sourcedefault: all-in-one.yaml- name: versiondescription: The version of the applicationtype: stringdefault: "v0.10" workspaces:- name: codebase- name: docker-configtasks:- name: git-clonetaskRef:name: git-cloneparams:- name: urlvalue: "$(params.git-url)"workspaces:- name: sourceworkspace: codebase- name: build-to-packagetaskRef:name: build-to-packageworkspaces:- name: sourceworkspace: codebaserunAfter:- git-clone- name: generate-build-idtaskRef:name: generate-build-idparams:- name: versionvalue: "$(params.version)"runAfter:- git-clone- name: image-build-and-pushtaskRef:name: image-build-and-pushparams:- name: image-urlvalue: "$(params.image-url)"- name: image-tagvalue: "$(tasks.generate-build-id.results.buildId)"workspaces:- name: sourceworkspace: codebase- name: dockerconfigworkspace: docker-configrunAfter:- generate-build-id- build-to-package- name: deploy-to-clustertaskRef:name: deploy-using-kubectlworkspaces:- name: sourceworkspace: codebaseparams:- name: deploy-config-filevalue: $(params.deploy-config-file)- name: image-urlvalue: $(params.image-url)- name: image-tagvalue: "$(tasks.generate-build-id.results.buildId)"runAfter:- image-build-and-push

8. 07-rbac.yaml

   因为06task的容器要执行kubectl，所以，给这个pod要指定一个serviceaccount，这样才能操作集群的资源

   ---
   apiVersion: v1
   kind: ServiceAccount
   metadata:name: helloworld-admin
   ---
   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRoleBinding
   metadata:name: helloworld-admin
   roleRef:apiGroup: rbac.authorization.k8s.iokind: ClusterRolename: cluster-admin
   subjects:
   - kind: ServiceAccountname: helloworld-adminnamespace: default

9. 08-pipelinerun-s2i.yaml

   apiVersion: tekton.dev/v1beta1
   kind: PipelineRun
   metadata:name: s2i-buildid-run-00002
   spec:serviceAccountName: defaulttaskRunSpecs:- pipelineTaskName: deploy-to-clustertaskServiceAccountName: helloworld-adminpipelineRef:name: source-to-imageparams:- name: git-urlvalue: https://gitee.com/mageedu/spring-boot-helloWorld.git- name: image-urlvalue: icloud2native/spring-boot-helloworld- name: versionvalue: v0.1.2workspaces:- name: codebasevolumeClaimTemplate:spec:accessModes:- ReadWriteOnceresources:requests:storage: 1GistorageClassName: nfs-csi- name: docker-configsecret:secretName: docker-config

   运行:

   kubectl apply -f .
   

   结果：

   1. 整个pipeline执行成功
      
      2、image推送到dockerhub
      
      3、查看部署
      
      更多关于tekton文章，后续更新。。。