云原生CI/CD框架Tekton国内部署方式

Tekton 是一款功能非常强大而灵活的 CI/CD 开源的云原生框架。致力于提供全功能、标准化的云原生 CI/CD 解决方案。【本文主要是通过流水线自动化的将tekton镜像同步到腾讯云仓库，并部署tekton】

应用镜像

阿里云镜像仓库居然有限制...这次转到腾讯云镜像仓库了;ccr.ccs.tencentyun.com/tektons/dashboard

Pipeline

借助GitHub Actions:

1. 同步镜像并生成镜像映射文件(json):
2. 收集镜像映射文件为制品;

 This is a basic workflow to help you get started with Actions 
 
name: Get Tekton Images 
env: 
  VERSION: v0.29.0 
 
on: 
  push: 
    paths: 
      - '.github/workflows/tekton.yaml' 
      - 'tekton/**' 
 
jobs: 
  build: 
    runs-on: ubuntu-18.04 
    steps: 
    - uses: actions/checkout@v2 
    - name: build 
      run: | 
       
        curl https://storage.googleapis.com/tekton-releases/pipeline/previous/${{ env.VERSION }}/release.yaml -o release.yaml 
        grep  -v "#" release.yaml | grep -v "^$"  > release1.yaml  ; sed -i 's/\-\-\-/###/g' release1.yaml 
        python3  tekton/get_tekton_images.py ${{ secrets.DOCKER_USER}} ${{ secrets.DOCKER_PASSWD}} 
    - uses: actions/upload-artifact@v2 
      with:  
        name: ${{ env.VERSION }}-tekton-images 
        path: tekton_images.json 

 

部署文件解析

下载release部署yaml;

解析Deployments对象中的images;

a.tekton-pipelines-controller

b.tekton-pipelines-webhook

c.tekton-dashboard(最新tag)

gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/controller:v0.29.0@sha256:72f79471f06d096cc53e51385017c9f0f7edbc87379bf415f99d4bd11cf7bc2b 
gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/kubeconfigwriter:v0.29.0@sha256:6d058f2203b9ab66f538cb586c7dc3b7cc31ae958a4135dd99e51799f24b06c9 
gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/git-init:v0.29.0@sha256:c0b0ed1cd81090ce8eecf60b936e9345089d9dfdb6ebdd2fd7b4a0341ef4f2b9 
gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/entrypoint:v0.29.0@sha256:66958b78766741c25e31954f47bc9fd53eaa28263506b262bf2cc6df04f18561 
gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/nop:v0.29.0@sha256:6a037d5ba27d9c6be32a9038bfe676fb67d2e4145b4f53e9c61fb3e69f06e816 
gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/imagedigestexporter:v0.29.0@sha256:e38dd0d32253fce5aaf1e501c0bc71facc3720564b7e97055921cc5390d612e0 
gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/pullrequest-init:v0.29.0@sha256:d28202fb8b33a1d4c05f261ef8dcbcdcf3b469887d4dad256ce91f73c917420e 
gcr.io/google.com/cloudsdktool/cloud-sdk@sha256:27b2c22bf259d9bc1a291e99c63791ba0c27a04d2db0a43241ba0f1f20f4067f 
gcr.io/distroless/base@sha256:aa4fd987555ea10e1a4ec8765da8158b5ffdfef1e72da512c7ede509bc9966c4 
mcr.microsoft.com/powershell:nanoserver@sha256:b6d5ff841b78bdf2dfed7550000fd4f3437385b8fa686ec0f010be24777654d6 
gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/webhook:v0.29.0@sha256:46d5b90a7f4e9996351ad893a26bcbd27216676ad4d5316088ce351fb2c2c3dd 

用Python编写一个数据解析脚本：

import yaml 
import json 
import sys 
import os 
 
class Tekton : 
    def __init__(self, file_name, registry_user, registry_passwd): 
        self.yaml_file = file_name 
        self.arg_imgs = ["gcr.io/tekton-releases/github.com/tektoncd/dashboard/cmd/dashboard@sha256:95f71a2568ced67ec370b5360f88bec3280601908cac9e62dfbb801114480437"] 
        self.split_str = "###" 
        self.deployments = ["tekton-pipelines-controller", "tekton-pipelines-webhook"] 
        self.kind_type = "Deployment" 
        self.target_registry = "ccr.ccs.tencentyun.com/tektons/" 
        self.repos = [  "controller", "kubeconfigwriter", "git-init", 
                        "entrypoint","nop","imagedigestexporter",  
                        "pullrequest-init", "cloud-sdk", "base", "powershell", "webhook"] 
        self.result = [] 
        self.registry_user = registry_user 
        self.registry_passwd = registry_passwd 
 
    def load_yaml(self, data): 
        content = yaml.load(data) 
        return content 
 
    def load_json(self, data): 
        content = json.loads(data) 
        return content 
 
    def get_images(self): 
        f = open(self.yaml_file, 'r').read() 
        for i in f.split("###")[:-1]: 
            try: 
                content = self.load_yaml(i.replace("###", "")) 
                if content["kind"] == self.kind_type: 
                    deploy_name = content["metadata"]["name"] 
                    # 获取image 
                    if deploy_name in self.deployments: 
                        img = content["spec"]["template"]["spec"]["containers"][0]["image"] 
                        self.arg_imgs.append(img) 
                    # 获取参数中的images 
                    if deploy_name == "tekton-pipelines-controller": 
                        arg_img =  content["spec"]["template"]["spec"]["containers"][0]["args"] 
                        for a in arg_img: 
                            if not a.startswith("-"): 
                                self.arg_imgs.append(a) 
            except Exception as e: 
                print(e) 
        return self.arg_imgs 
 
    def save_json_file(self, data, file_name): 
        for i in self.arg_imgs: 
            self.result.append({ 
                "s_image": i, 
                "t_image": self.target_registry + i.split("/")[-1].split("@")[0] 
                }) 
        newdata = json.dumps(self.result, indent=4) 
        a=open(file_name, 'w') 
        a.write(newdata) 
        a.close() 
 
    def sync_images(self): 
        f = open("tekton_images.json", 'r').read() 
        content = self.load_json(f) 
        docker_login_cmd = "docker login -u {0} -p {1} {2}".format( 
            self.registry_user, 
            self.registry_passwd, 
            self.target_registry.split("/")[0]) 
        os.system(docker_login_cmd) 
        for item in content: 
            print("[GetImages] {}".format(item)) 
            docker_pull_cmd = "docker pull {0}".format(item["s_image"]) 
            docker_tag_cmd = "docker tag {0} {1}".format(item["s_image"], item["t_image"]) 
            docker_push_cmd = "docker push {0}".format(item["t_image"]) 
            os.system(docker_pull_cmd + "&&" + docker_tag_cmd + "&&" + docker_push_cmd ) 
            print("[GetImagesDone] {}".format(item)) 
         
if __name__ == '__main__': 
    tekton = Tekton("release1.yaml", sys.argv[1], sys.argv[2]) 
    images = tekton.get_images() 
    tekton.save_json_file(images, "tekton_images.json") 
    tekton.sync_images() 

镜像映射文件

s_image 原始镜像名称， t_image 目标镜像名称; 这里使用腾讯云的镜像仓库;

[ 
    { 
        "s_image": "gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/controller:v0.29.0@sha256:72f79471f06d096cc53e51385017c9f0f7edbc87379bf415f99d4bd11cf7bc2b", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/controller:v0.29.0" 
    }, 
    { 
        "s_image": "gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/kubeconfigwriter:v0.29.0@sha256:6d058f2203b9ab66f538cb586c7dc3b7cc31ae958a4135dd99e51799f24b06c9", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/kubeconfigwriter:v0.29.0" 
    }, 
    { 
        "s_image": "gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/git-init:v0.29.0@sha256:c0b0ed1cd81090ce8eecf60b936e9345089d9dfdb6ebdd2fd7b4a0341ef4f2b9", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/git-init:v0.29.0" 
    }, 
    { 
        "s_image": "gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/entrypoint:v0.29.0@sha256:66958b78766741c25e31954f47bc9fd53eaa28263506b262bf2cc6df04f18561", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/entrypoint:v0.29.0" 
    }, 
    { 
        "s_image": "gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/nop:v0.29.0@sha256:6a037d5ba27d9c6be32a9038bfe676fb67d2e4145b4f53e9c61fb3e69f06e816", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/nop:v0.29.0" 
    }, 
    { 
        "s_image": "gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/imagedigestexporter:v0.29.0@sha256:e38dd0d32253fce5aaf1e501c0bc71facc3720564b7e97055921cc5390d612e0", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/imagedigestexporter:v0.29.0" 
    }, 
    { 
        "s_image": "gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/pullrequest-init:v0.29.0@sha256:d28202fb8b33a1d4c05f261ef8dcbcdcf3b469887d4dad256ce91f73c917420e", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/pullrequest-init:v0.29.0" 
    }, 
    { 
        "s_image": "gcr.io/google.com/cloudsdktool/cloud-sdk@sha256:27b2c22bf259d9bc1a291e99c63791ba0c27a04d2db0a43241ba0f1f20f4067f", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/cloud-sdk" 
    }, 
    { 
        "s_image": "gcr.io/distroless/base@sha256:aa4fd987555ea10e1a4ec8765da8158b5ffdfef1e72da512c7ede509bc9966c4", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/base" 
    }, 
    { 
        "s_image": "mcr.microsoft.com/powershell:nanoserver@sha256:b6d5ff841b78bdf2dfed7550000fd4f3437385b8fa686ec0f010be24777654d6", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/powershell:nanoserver" 
    }, 
    { 
        "s_image": "gcr.io/tekton-releases/github.com/tektoncd/pipeline/cmd/webhook:v0.29.0@sha256:46d5b90a7f4e9996351ad893a26bcbd27216676ad4d5316088ce351fb2c2c3dd", 
        "t_image": "ccr.ccs.tencentyun.com/tektons/webhook:v0.29.0" 
    }, 
    { 
    "s_image": "gcr.io/tekton-releases/github.com/tektoncd/dashboard/cmd/dashboard@sha256:95f71a2568ced67ec370b5360f88bec3280601908cac9e62dfbb801114480437", 
    "t_image": "ccr.ccs.tencentyun.com/tektons/dashboard" 
     
    } 
] 

镜像映射文件可以在GitHubActions页面下载：

下载镜像脚本

解析上面生成的镜像文件，docker pull下载对应的镜像到本地;

import json  
import os 
 
class Tekton: 
    def __init__(self): 
        self.json_file = "tekton_images.json" 
        self.target_registry = "ccr.ccs.tencentyun.com/tektons/" 
        # self.registry_user = registry_user 
        # self.registry_passwd = registry_passwd 
 
    def load_json(self, data): 
        content = json.loads(data) 
        return content 
 
    def down_images(self): 
        f = open(self.json_file, 'r').read() 
        content = self.load_json(f) 
 
        # docker_login_cmd = "docker login -u {0} -p {1} {2}".format( 
        #             self.registry_user, 
        #             self.registry_passwd, 
        #             self.target_registry.split("/")[0]) 
        for item in content: 
            print("[GetImages] {}".format(item["t_image"])) 
            docker_pull_cmd = "docker pull {0}".format(item["t_image"]) 
            # docker_tag_cmd = "docker tag {0} {1}".format(item["t_image"], item["s_image"].split("@")[0]) 
            os.system(docker_pull_cmd + "&&" + docker_tag_cmd ) 
            print("[GetImagesDone] {}".format(item)) 
 
if __name__ == '__main__': 
    t = Tekton().down_images() 

部署Tekton

替换部署文件中的镜像:

1. 手动更新release.yaml中的镜像;然后kubectl apply release.yaml 部署(后续有时间再优化脚本，实现自动更新release.yaml)
2. 手动更新tekton-dashboard-release.yaml中的镜像;然后部署;

[root@master ~]# kubectl -n tekton-pipelines get pod 
NAME                                          READY   STATUS    RESTARTS   AGE 
tekton-dashboard-5c4b89d9-2z8g7               1/1     Running   0          21m 
tekton-pipelines-controller-b96f647bb-gff69   1/1     Running   0          13h 
tekton-pipelines-webhook-76bc9c97b9-cd2m4     1/1     Running   0          13h 

编写一个Ingress来暴露tekton dashboard：

apiVersion: extensions/v1beta1 
kind: Ingress 
metadata: 
  name: tekton-service 
  namespace: tekton-pipelines 
  annotations: 
    kubernetes.io/ingress.class: nginx 
    nginx.ingress.kubernetes.io/proxy-body-size: 256m 
spec: 
  rules: 
  - host: tekton.idevops.site 
    http: 
     paths: 
     - path: / 
       backend: 
          serviceName: tekton-dashboard 
          servicePort: 9097 

访问UI页面：

编写Pipeline

apiVersion: tekton.dev/v1beta1 
kind: Task 
metadata: 
  name: tektoncd-task 
spec: 
  resources: 
    inputs: 
    - name: repo 
      type: git 
  steps: 
  - name: run-test 
    image: maven:3-jdk-8 
    workingDir: /workspace/repo 
    command: ["mvn"] 
    args: ["clean", "package"] 
--- 
apiVersion: tekton.dev/v1alpha1 
kind: PipelineResource 
metadata: 
  name: tektoncd-resource 
spec: 
  type: git 
  params: 
    - name: url 
      value: http://192.168.1.200/devops/devops-maven-service.git 
    - name: revision 
      value: master 
--- 
apiVersion: tekton.dev/v1beta1 
kind: TaskRun 
metadata: 
  name: cdpipeline 
spec: 
  taskRef: 
    name: tektoncd-task 
  resources: 
    inputs: 
    - name: repo 
      resourceRef: 
        name: tektoncd-resource