GitLabCI模板库的流水线优化实践

[[441815]]

还是按照之前的目录结构完成的，jobs目录用于存放作业模板。templates目录用于存放流水线模板。这次使用default-pipeline.yml作为所有作业的基础模板。

作业模板

作业分为Build、test、codeanalysis、artifactory、deploy部分，在每个作业中配置了rules功能开关，由变量控制最终作业的运行。

jobs/build.yml 构建作业模板

包含两个作业模板，分别是普通的构建模板(maven/npm/gradle)和docker 镜像构建模板。

## build相关作业 
 
.build: 
  stage: build 
  script:  
    - | 
      ${BUILD_SHELL} 
  variables: 
    GIT_CHECKOUT: "true" 
  rules: 
    - if: " $RUN_PIPELINE_BUILD == 'no' " 
      when: never 
    - when: always 
 
 
## 构建镜像 
.build-docker: 
  stage: buildimage 
  script: 
    - docker login -u $CI_REGISTRY_USER -p $CI_REGISTRY_PASSWD  $CI_REGISTRY 
    - docker build -t ${IMAGE_NAME} -f ${DOCKER_FILE_PATH} . 
    - docker push ${IMAGE_NAME}  
    - docker rmi ${IMAGE_NAME}  
  rules: 
    - if: " $RUN_BUILD_IMAGE == 'no' " 
      when: never 
    - when: always 

jobs/test.yml 测试作业模板

主要用于运行项目单元测试，例如maven、gradle、npm单元测试。

#单元测试 
.test: 
  stage: test 
  script: 
    - $TEST_SHELL 
  artifacts: 
    reports: 
      junit: ${JUNIT_REPORT_PATH} 
  rules: 
    - if: " $RUN_PIPELINE_TEST == 'no' " 
      when: never 
    - when: always 

jobs/codeanalysis.yml 代码扫描模板

包含两个作业模板，分别为扫描作业和获取扫描结果。这里面将扫描参数进行了分类，通用的项目参数、特殊的合并请求参数、自定义的项目参数。

##代码扫描 
## 
## 
## 
 
.code_analysis: 
  variables: 
    GLOBAL_PROJECT_ARGS: "-Dsonar.projectKey=${CI_PROJECT_NAME}  
                          -Dsonar.projectName=${CI_PROJECT_NAME}  
                          -Dsonar.projectVersion=${CI_COMMIT_REF_NAME}  
                          -Dsonar.projectDescription=${CI_PROJECT_TITLE}" 
    GLOBAL_SERVER_ARGS:  "-Dsonar.ws.timeout=30  
                          -Dsonar.links.homepage=${CI_PROJECT_URL}  
                          -Dsonar.host.url=${SONAR_SERVER_URL}  
                          -Dsonar.login=${SONAR_SERVER_LOGIN} 
                          -Dsonar.sourceEncoding=UTF-8 " 
    GLOBAL_MR_ARGS: " -Dsonar.pullrequest.key=${CI_MERGE_REQUEST_IID}  
                      -Dsonar.pullrequest.branch=${CI_MERGE_REQUEST_SOURCE_BRANCH_NAME}  
                      -Dsonar.pullrequest.base=${CI_MERGE_REQUEST_TARGET_BRANCH_NAME}  
                      -Dsonar.gitlab.ref_name=${CI_COMMIT_REF_NAME}  
                      -Dsonar.gitlab.commit_sha=${CI_COMMIT_SHA}  
                      -Dsonar.gitlab.project_id=${CI_PROJECT_PATH}  
                      -Dsonar.pullrequest.gitlab.repositorySlug=${CI_PROJECT_ID} " 
    MULTI_BRANCH_ARGS: "-Dsonar.branch.name=${CI_COMMIT_REF_NAME}" 
  stage: code_analysis 
  script: 
    - echo ${GLOBAL_PROJECT_ARGS} ${GLOBAL_SERVER_ARGS} ${SONAR_SCAN_ARGS} ${GLOBAL_MR_ARGS} 
    #sonar-scanner $GLOBAL_PROJECT_ARGS $GLOBAL_SERVER_ARGS $SCAN_JAVA_ARGS 
    - | 
        if [ $CI_PIPELINE_SOURCE == 'merge_request_event' ]  
         
        then 
           echo "sonar-scanner ${GLOBAL_PROJECT_ARGS} ${GLOBAL_SERVER_ARGS} ${SONAR_SCAN_ARGS} "  
           sonar-scanner ${GLOBAL_PROJECT_ARGS} ${GLOBAL_SERVER_ARGS} ${SONAR_SCAN_ARGS}  
        else  
           echo "sonar-scanner ${GLOBAL_PROJECT_ARGS} ${GLOBAL_SERVER_ARGS} ${SONAR_SCAN_ARGS} ${MULTI_BRANCH_ARGS}" 
           sonar-scanner ${GLOBAL_PROJECT_ARGS} ${GLOBAL_SERVER_ARGS} ${SONAR_SCAN_ARGS}  ${MULTI_BRANCH_ARGS} 
        fi  
  rules: 
    - if: " $RUN_CODE_ANALYSIS == 'no' " 
      when: never 
    - when: always 
 
 
#### 获取代码扫描结果 
.get_analysis_result: 
  stage: get_analysis_result 
  script: 
    - | 
        SONAR_REPORT_URL=$(grep "ceTaskUrl" .scannerwork/report-task.txt  | awk -F = '{OFS="=";print $2,$3}') 
        echo ${SONAR_REPORT_URL} 
         
         
        for i in {1..10} 
        do 
          curl -k -u "${SONAR_SERVER_LOGIN}":"" ${SONAR_REPORT_URL}  -o sonar_result.txt -s 
          grep '"status":"SUCCESS"' sonar_result.txt  && SONAR_SCAN_RESULT='SUCCESS' 
         
          if [ ${SONAR_SCAN_RESULT} == 'SUCCESS' ] 
            then 
              echo "${SONAR_SCAN_RESULT}" 
              SONAR_SCAN_RESULT=SUCCESS 
 
              curl -k -u "${SONAR_SERVER_LOGIN}":"" "${SONAR_SERVER_URL}/api/qualitygates/project_status?projectKey=$CI_PROJECT_NAME&branch=$CI_COMMIT_REF_NAME" -o result.txt  -s 
              echo "result info ---->>>>>" 
              cat result.txt 
              result=`cat result.txt | awk -F ':' '{print $3}' | awk -F '"' '{print$2}'` 
               
              echo $result 
 
              if [ $result == 'ERROR' ]                  
                then  
                  echo "${result}" 
                  exit 122 
                  break; 
              else 
                  echo "success!" 
                  break; 
              fi 
             
          else 
            SONAR_SCAN_RESULT='ERROR' 
            echo "第$i次获取结果信息，不是成功状态，睡眠10秒！" 
            cat sonar_result.txt 
            sleep 10 
          fi 
        done 
         
  rules: 
    - if: " $RUN_CODE_ANALYSIS == 'no' " 
      when: never 
    - when: always 

jobs/artifactory.yml 制品管理作业

包含两个作业，制品上传与下载。使用artifactory制品库接口。

## 制品库相关 
## 
 
.deploy-artifact: 
  stage: deploy-artifact 
  script: 
    - echo "curl -u${ARTIFACT_USER}:${ARTIFACT_PASSWD} -T ${ARTIFACT_PATH} $ARTIFACTORY_URL/$ARTIFACTORY_NAME/$TARGET_FILE_PATH/$TARGET_ARTIFACT_NAME" 
    - curl -u${ARTIFACT_USER}:${ARTIFACT_PASSWD} -T ${ARTIFACT_PATH} "$ARTIFACTORY_URL/$ARTIFACTORY_NAME/$TARGET_FILE_PATH/$TARGET_ARTIFACT_NAME" 
  rules: 
    - if: " $RUN_DEPLOY_ARTIFACTS == 'no' " 
      when: never 
    - when: always 
 
 
.down-artifact: 
  stage: down-artifact 
  script: 
    - curl -u${ARTIFACT_USER}:${ARTIFACT_PASSWD} -O "$ARTIFACTORY_URL/$ARTIFACTORY_NAME/$TARGET_FILE_PATH/$TARGET_ARTIFACT_NAME" 
    - ls 

jobs/deploy.yml 发布作业模板

发布基于k8s的发布和回滚配置。

## 应用发布 
 
 
## 使用kubectl镜像发布 
.deploy_k8s: 
  stage: deploy 
  script: 
    - echo $KUBE_TOKEN 
    - kubectl config set-cluster my-cluster --server=${KUBE_URL} --certificate-authority="${KUBE_CA_PEM_FILE}" 
    - kubectl config set-credentials admin --token=${KUBE_TOKEN} 
    - ls -a  
    - sed -i "s#__namespace__#${NAMESPACE}#g" ${DEPLOY_FILE} 
    - sed -i "s#__appname__#${APP_NAME}#g" ${DEPLOY_FILE} 
    - sed -i "s#__containerport__#${CONTAINER_PORT}#g" ${DEPLOY_FILE}  
    - sed -i "s#__nodeport__#${NODE_PORT}#g" ${DEPLOY_FILE}  
    - sed -i "s#__imagename__#${IMAGE_NAME}#g" ${DEPLOY_FILE}  
    - sed -i "s#__CI_ENVIRONMENT_SLUG__#${CI_ENVIRONMENT_SLUG}#g" ${DEPLOY_FILE} 
    - sed -i "s#__CI_PROJECT_PATH_SLUG__#${CI_PROJECT_PATH_SLUG}#g" ${DEPLOY_FILE} 
    - sed -i "s#__ingressdomain__#${ENV_URL}#g" ${DEPLOY_FILE} 
    - cat ${DEPLOY_FILE} 
    - "kubectl create secret docker-registry ${APP_NAME} \ 
            --docker-server=${CI_REGISTRY} \ 
            --docker-username=$CI_REGISTRY_USER \ 
            --docker-password=${CI_REGISTRY_PASSWD} \ 
            --docker-email=test@test.com -n ${NAMESPACE} || echo 'secrets already exists'" 
    - kubectl apply -f ${DEPLOY_FILE} 
  rules: 
    - if: " $RUN_DEPLOY_K8S == 'no'" 
      when: never 
    - when: manual 
  environment: 
    name: "${ENV_NAME}" 
    url: "http://${ENV_NAME}.${CI_PROJECT_NAMESPACE}.${CI_PROJECT_NAME}.devops.com" 
 
   
 
 
## 回滚 
.rollout_k8s: 
  stage: deploy 
  script: 
    - rm -rf $HOME/.kube 
    - kubectl config set-cluster my-cluster --server=${KUBE_URL} --certificate-authority="${KUBE_CA_PEM_FILE}" 
    - kubectl config set-credentials admin --token=${KUBE_TOKEN} 
    - kubectl rollout history deployment ${APP_NAME} -n ${NAMESPACE} 
    - kubectl rollout undo deployment ${APP_NAME} -n ${NAMESPACE} 
  rules: 
    - if: " $RUN_DEPLOY_K8S == 'no' " 
      when: never 
    - when: manual 
  environment: 
    name: "${ENV_NAME}" 
    action: stop 

default流水线模板

templates/default-pipeline.yml 模板分成个部分：

• include导入作业模板
• variables 定义全局变量
• workflow 定义流水线控制
• jobs 构建与发布作业

include导入作业模板

include: 
  - project: 'cidevops/cidevops-newci-service' 
    ref: master 
    file: 'jobs/build.yml' 
  - project: 'cidevops/cidevops-newci-service' 
    ref: master 
    file: 'jobs/test.yml' 
  - project: 'cidevops/cidevops-newci-service' 
    ref: master 
    file: 'jobs/codeanalysis.yml' 
  - project: 'cidevops/cidevops-newci-service' 
    ref: master 
    file: 'jobs/deploy.yml' 
  - project: 'cidevops/cidevops-newci-service' 
    ref: master 
    file: 'jobs/artifactory.yml' 

variables 定义全局变量

variables: 
  ## 全局配置 
  GIT_CLONE_PATH: ${CI_BUILDS_DIR}/builds/${CI_PROJECT_NAMESPACE}/${CI_PROJECT_NAME}/${CI_PIPELINE_ID}     
  GIT_CHECKOUT: "false" 
  CACHE_DIR: "" 
 
  ## 作业控制 
  RUN_PIPELINE_BUILD: ""     #是否运行构建 yes/no  
  RUN_PIPELINE_TEST: ""      #是否运行测试 yes/no  
  RUN_CODE_ANALYSIS: ""      #是否代码扫描 yes/no  
  RUN_BUILD_IMAGE: ""        #是否生成镜像 yes/no 
  RUN_DEPLOY_ARTIFACTS: ""   #是否上传制品 yes/no 
  RUN_DEPLOY_K8S: ""         #是否发布K8S yes/no 
 
  ## 依赖容器镜像 
  BUILD_IMAGE: "" 
  CURL_IMAGE: "curlimages/curl:7.70.0" 
  SONAR_IMAGE: "sonarsource/sonar-scanner-cli:latest" 
  KUBECTL_IMAGE: "lucj/kubectl:1.17.2" 
   
   
  ## 构建测试参数 
  MAVEN_OPTS: ""                #maven构建参数  
  GRADLE_OPTS: ""               #gradle构建参数 
  BUILD_SHELL: ''               #构建命令 
 
  ## 单元测试参数 
  TEST_SHELL : 'mvn test  --settings=./settings.xml '       #测试命令 
  JUNIT_REPORT_PATH: 'target/surefire-reports/TEST-*.xml'   #单元测试报告 
 
  ## 代码扫描 
  SONAR_SOURCE_DIR : "src"                                          #项目源码目录 
  SONAR_SERVER_URL: "http://192.168.1.200:30090"                    #SonarQube服务器信息 
  SONAR_SERVER_LOGIN: "ee2bcb37deeb6dfe3a07fe08fb529559b00c1b7b"    #Sonar Token最好在项目中定义。 
  SONAR_SCAN_ARGS: "-Dsonar.sources=${SONAR_SOURCE_DIR}"            #项目扫描参数 
 
  ## 构建镜像 
  CI_REGISTRY: 'registry.cn-beijing.aliyuncs.com'               #镜像仓库地址               
  CI_REGISTRY_USER: 'xxxxxx'                               #仓库用户信息 
  #CI_REGISTRY_PASSWD: 'xxxxxxxx.'                              #仓库用户密码 
  IMAGE_NAME: "${CI_REGISTRY}/${CI_PROJECT_PATH}:${CI_COMMIT_REF_NAME}-${CI_COMMIT_SHORT_SHA}"     #镜像名称 
  DOCKER_FILE_PATH: "./Dockerfile"                              #Dockerfile位置 
 
  ## 上传制品库(artifactory) 
  ARTIFACTORY_URL: "http://192.168.1.200:30082/artifactory"     #制品库地址 
  ARTIFACTORY_NAME: "${CI_PROJECT_NAMESPACE}"                   #制品库名称 
  ARTIFACT_PACKAGE: "jar"                                       #制品类型 
  ARTIFACT_PATH: "target/*.${ARTIFACT_PACKAGE}"                 #制品位置 
  TARGET_FILE_PATH: "${CI_PROJECT_NAMESPACE}/${CI_PROJECT_NAME}/${CI_COMMIT_REF_NAME}-${CI_COMMIT_SHORT_SHA}"   #目标制品位置(目录结构) 
  TARGET_ARTIFACT_NAME: "${CI_PROJECT_NAME}-${CI_COMMIT_REF_NAME}-${CI_COMMIT_SHORT_SHA}.${ARTIFACT_PACKAGE}"   #目标制品名称 
 
  ## 部署应用k8s 
  APP_NAME: "$CI_PROJECT_NAME"                #应用名称 <--> deploymentName 
  CONTAINER_PORT: "8081"                      #服务端口 <--> servicesPort 
  NAMESPACE: "$CI_PROJECT_NAME-$CI_PROJECT_ID-$CI_ENVIRONMENT_SLUG"             #名称空间 
  ENV_URL: "${ENV_NAME}.${CI_PROJECT_NAMESPACE}.${CI_PROJECT_NAME}.devops.com"  #IngressHosts 

流水线设置

## 流水线控制 
workflow: 
  rules: 
    - if: "$CI_MERGE_REQUEST_ID"          #过滤合并请求 
      when: never 
    - if: "$CI_PIPELINE_SOURCE == 'web'"    #允许在web页面发布 
    - if: "$CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ "  #过滤版本分支和主干分支提交代码触发 
      when: never 
    - if: "$CI_COMMIT_BEFORE_SHA == '0000000000000000000000000000000000000000'"     #过滤分支创建请求 
      when: never 
    ### 默认策略 
    - when: always 
 
 
## 运行阶段   
stages: 
  - build 
  - test 
  - parallel01 
  - get_analysis_result 
  - deploy-artifact 
  - deploy-feature 
  - rollout-feature 
  - deploy-uat 
  - rollout-uat 
  - deploy-stag 
  - rollout-stag 
  - deploy-prod 
  - rollout-prod 
 
cache: 
  paths: 
    - ${CACHE_DIR} 
     
before_script: 
  - export 

构建作业配置

################# Jobs Configure ##################### 
## 构建作业 
build: 
  image: ${BUILD_IMAGE} 
  extends: .build 
 
## 测试作业 
test: 
  image: ${BUILD_IMAGE} 
  extends: .test 
  before_script: 
    - ls  
    - ls ${CACHE_DIR} 
 
## 代码扫描 
code_analysis: 
  stage: parallel01 
  image: ${SONAR_IMAGE} 
  extends: .code_analysis 
 
## 获取构建结果 
get_analysis_result: 
  image: ${CURL_IMAGE} 
  extends: .get_analysis_result 
  needs: 
    - code_analysis 
 
 
## 构建镜像 
build_image: 
  image: docker:latest 
  services: 
    - name: docker:dind 
  stage: parallel01 
  extends: .build-docker 
 
 
## 上传制品 
deploy_artifact: 
  image: ${CURL_IMAGE} 
  stage: deploy-artifact 
  extends: .deploy-artifact 
 
## 下载制品   
#down_artifact:   
#  image: ${CURL_IMAGE} 
#  stage: down_artifact 
#  extends: .down-artifact 

发布部署作业配置

#################Deploy Feature Jobs Configure ##################### 
## feature发布应用 
deploy_feature: 
  variables: 
    DEPLOY_FILE: 'deployment.yaml' 
    ENV_NAME: 'feature' 
  stage: deploy-feature 
  image: ${KUBECTL_IMAGE} 
  extends: .deploy_k8s 
  environment: 
    on_stop: "rollout_feature" 
  rules: 
    - if: $RUN_DEPLOY_K8S == 'no' 
      when: never 
    - if: $CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ || $CI_COMMIT_TAG 
      when: never 
    - when: manual 
     
## 应用回滚 
rollout_feature: 
  variables: 
    DEPLOY_FILE: 'deployment.yaml' 
    ENV_NAME: 'feature' 
  stage: rollout-feature 
  image: ${KUBECTL_IMAGE} 
  extends: .rollout_k8s 
  needs: 
    - deploy_feature 
  rules: 
    - if: $RUN_DEPLOY_K8S == 'no' 
      when: never 
    - if: $CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ || $CI_COMMIT_TAG 
      when: never 
    - when: on_failure 
 
 
#################Deploy UAT Jobs Configure ##################### 
## UAT 
deploy_uat: 
  variables: 
    DEPLOY_FILE: 'config/deployment-uat.yaml' 
    ENV_NAME: 'uat' 
  stage: deploy-uat 
  image: ${KUBECTL_IMAGE} 
  extends: .deploy_k8s 
  environment: 
    on_stop: "rollout_uat" 
  rules: 
    - if: $RUN_DEPLOY_K8S == 'no' 
      when: never 
    - if: $CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ || $CI_COMMIT_TAG 
      when: manual 
    - when: never 
 
## UAT应用回滚 
rollout_uat: 
  variables: 
    DEPLOY_FILE: 'config/deployment-uat.yaml' 
    ENV_NAME: 'uat' 
  stage: rollout-uat 
  image: ${KUBECTL_IMAGE} 
  extends: .rollout_k8s 
  needs: 
    - deploy_uat 
  rules: 
    - if: $RUN_DEPLOY_K8S == 'no' 
      when: never 
    - if: $CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ || $CI_COMMIT_TAG 
      when: on_failure 
    - when: never 
 
#################Deploy STAG Jobs Configure ##################### 
## STAG 
deploy_stag: 
  variables: 
    DEPLOY_FILE: 'config/deployment-stag.yaml' 
    ENV_NAME: 'stag' 
  stage: deploy-stag 
  image: ${KUBECTL_IMAGE} 
  extends: .deploy_k8s 
  environment: 
    on_stop: "rollout_stag" 
  needs: 
    - deploy_uat 
  rules: 
    - if: $RUN_DEPLOY_K8S == 'no' 
      when: never 
    - if: $CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ || $CI_COMMIT_TAG 
      when: manual 
    - when: never 
 
## STAG应用回滚 
rollout_stag: 
  variables: 
    DEPLOY_FILE: 'config/deployment-stag.yaml' 
    ENV_NAME: 'stag' 
  stage: rollout-stag 
  image: ${KUBECTL_IMAGE} 
  extends: .rollout_k8s 
  needs: 
    - deploy_stag 
  rules: 
    - if: $RUN_DEPLOY_K8S == 'no' 
      when: never 
    - if: $CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ || $CI_COMMIT_TAG 
      when: on_failure 
    - when: never 
 
#################Deploy PROD Jobs Configure ##################### 
## PROD 
deploy_prod: 
  variables: 
    DEPLOY_FILE: 'config/deployment-prod.yaml' 
    ENV_NAME: 'prod' 
  stage: deploy-prod 
  image: ${KUBECTL_IMAGE} 
  extends: .deploy_k8s 
  environment: 
    on_stop: "rollout_prod" 
  needs: 
    - deploy_stag 
  rules: 
    - if: $RUN_DEPLOY_K8S == 'no' 
      when: never 
    - if: $CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ || $CI_COMMIT_TAG 
      when: manual 
    - when: never 
 
## PROD应用回滚 
rollout_prod: 
  variables: 
    DEPLOY_FILE: 'config/deployment-prod.yaml' 
    ENV_NAME: 'prod' 
  stage: rollout-prod 
  image: ${KUBECTL_IMAGE} 
  extends: .rollout_k8s 
  needs: 
    - deploy_prod 
  rules: 
    - if: $RUN_DEPLOY_K8S == 'no' 
      when: never 
    - if: $CI_COMMIT_REF_NAME =~ /^RELEASE-*/ || $CI_COMMIT_REF_NAME =~ /master/ || $CI_COMMIT_TAG 
      when: on_failure 
    - when: never 

JAVA项目交付流水线

导入作业模板

include: 
  - project: 'cidevops/cidevops-newci-service' 
    ref: master 
    file: 'templates/default-pipeline.yml' 

配置项目参数

variables: 
  ## 全局配置 
  GIT_CLONE_PATH: ${CI_BUILDS_DIR}/builds/${CI_PROJECT_NAMESPACE}/${CI_PROJECT_NAME}/${CI_PIPELINE_ID}     
  GIT_CHECKOUT: "false" 
  CACHE_DIR: "target/" 
 
  ## 作业控制 
  RUN_PIPELINE_BUILD:   "yes"     #是否运行构建 yes/no  
  RUN_PIPELINE_TEST:    "yes"      #是否运行测试 yes/no  
  RUN_CODE_ANALYSIS:    "yes"     #是否代码扫描 yes/no  
  RUN_BUILD_IMAGE:      "yes"     #是否生成镜像 yes/no 
  RUN_DEPLOY_ARTIFACTS: "no"      #是否上传制品 yes/no 
  RUN_DEPLOY_K8S:       "yes"     #是否发布K8S yes/no 
 
  ## 依赖容器镜像 
  BUILD_IMAGE: "maven:3.6.3-jdk-8" 
  CURL_IMAGE: "curlimages/curl:7.70.0" 
  SONAR_IMAGE: "sonarsource/sonar-scanner-cli:latest" 
  KUBECTL_IMAGE: "lucj/kubectl:1.17.2" 
   
   
  ## 构建测试参数 
  MAVEN_OPTS: "-Dmaven.repo.local=/home/gitlab-runner/ci-build-cache/maven "  #maven构建参数 
  BUILD_SHELL: 'mvn clean package  -DskipTests  --settings=./settings.xml '   #构建命令 
  #GRADLE_OPTS: ""               #gradle构建参数 
 
  ## 单元测试参数 
  TEST_SHELL : 'mvn test  --settings=./settings.xml   '       #测试命令 
  JUNIT_REPORT_PATH: 'target/surefire-reports/TEST-*.xml'   #单元测试报告 
 
  ## 代码扫描 
  SONAR_SOURCE_DIR : "src"                                          #项目源码目录 
  SONAR_SERVER_URL: "http://192.168.1.200:30090"                    #SonarQube服务器信息 
  SONAR_SERVER_LOGIN: "ee2bcb37deeb6dfe3a07fe08fb529559b00c1b7b"    #Sonar Token最好在项目中定义。 
  SONAR_SCAN_ARGS: "-Dsonar.sources=${SONAR_SOURCE_DIR}  
                   -Dsonar.java.binaries=target/classes  
                   -Dsonar.java.test.binaries=target/test-classes  
                   -Dsonar.java.surefire.report=target/surefire-reports "                                     #项目扫描参数 
 
  ## 构建镜像 
  CI_REGISTRY: 'registry.cn-beijing.aliyuncs.com'               #镜像仓库地址               
  CI_REGISTRY_USER: 'xxxxxx'                               #仓库用户信息 
  #CI_REGISTRY_PASSWD: 'xxxxxxxx.'                              #仓库用户密码 
  IMAGE_NAME: "${CI_REGISTRY}/${CI_PROJECT_PATH}:${CI_COMMIT_REF_NAME}-${CI_COMMIT_SHORT_SHA}"     #镜像名称 
  DOCKER_FILE_PATH: "./Dockerfile"                              #Dockerfile位置 
 
  ## 上传制品库(artifactory) 
  #ARTIFACTORY_URL: "http://192.168.1.200:30082/artifactory"     #制品库地址 
  #ARTIFACTORY_NAME: "${CI_PROJECT_NAMESPACE}"                   #制品库名称 
  #ARTIFACT_PACKAGE: "jar"                                       #制品类型 
  #ARTIFACT_PATH: "target/*.${ARTIFACT_PACKAGE}"                 #制品位置 
  #TARGET_FILE_PATH: "${CI_PROJECT_NAMESPACE}/${CI_PROJECT_NAME}/${CI_COMMIT_REF_NAME}-${CI_COMMIT_SHORT_SHA}"   #目标制品位置(目录结构) 
  #TARGET_ARTIFACT_NAME: "${CI_PROJECT_NAME}-${CI_COMMIT_REF_NAME}-${CI_COMMIT_SHORT_SHA}.${ARTIFACT_PACKAGE}"   #目标制品名称 
 
  ## 部署应用k8s 
  APP_NAME: "$CI_PROJECT_NAME"                #应用名称 <-->deploymentName 
  CONTAINER_PORT: "8081"                      #服务端口 <--> servicesPort 
  NAMESPACE: "$CI_PROJECT_NAME-$CI_PROJECT_ID-$CI_ENVIRONMENT_SLUG"             #名称空间 
  ENV_URL: "${ENV_NAME}.${CI_PROJECT_NAMESPACE}.${CI_PROJECT_NAME}.devops.com"  #IngressHosts 

指定CI文件

运行流水线测试

docker镜像仓库

部署环境演示

Kubernetes集群中应用状态