Three ways to create tasks in batches by Jenkins

Recently, to build multiple test environments, it is necessary to batch copy all tasks under the dev view in Jenkins to views such as sit.

Explanation

Jenkins task name [测试环境标识]-[工程名称]，如：dev-daodaotest，sit-daodaotestrules: .

Show Tasks view in regular expressions: [测试环境标识]-.* ，如：dev-.*，sit-.*.

The first type: bulk copy under the directory

Jenkins tasks are based on xmldocuments stored all can be copied xmlto create a batch mode.

# 进入 jobs 目录下
$ cd ~/.jenkins/jobs

# 创建批量复制 shell 脚本
$ vi copyViewJobs.sh
#!/bin/bash

# 视图名称
viewName=$1
# 新视图名称
newViewName=$2

# 循环复制任务
for jobName in `ls /home/jenkins/.jenkins/jobs/`
 do
    # 判断文件存在并且是目录
    if test -d $jobName
    then
        # 目录为 $viewName 开头，则进行复制
        if [[ $jobName == *$viewName* ]]; then
           # 截取工程名称
           name=`echo $jobName|awk 'BEGIN{FS="'$viewName'-"} {print $2}'`
           newJobName=$newViewName-$name
           echo $newJobName
           # 复制 config.xml
           mkdir $newJobName && cp $jobName/config.xml $newJobName/
        fi
     fi
 done
 
 # 执行批量复制脚本，dev 视图下的任务负责到 sit 视图下
 $ sh copyViewJobs.sh dev sit

Note : After copying, Jenkins needs to reload the configuration to take effect. Operation menu path: Manage Jenkins-》Reload Configuration from Disk.

The second kind: jenkins-cli

The implementation steps are similar to the first one, you can implement it according to your own scripting language. The key commands are briefly described below.

jenkins-cliFor usage, see: http: // localhost: 8080 / cli

# 下载 jenkins-cli.jar
$ wget http://localhost:8080/jnlpJars/jenkins-cli.jar

# 获取视图下的所有任务
$  java -jar jenkins-cli.jar -s http://localhost:8080/ -auth daodaotest:daodaotest list-jobs dev

# 复制任务
$ java -jar jenkins-cli.jar -s http://localhost:8080/ -auth daodaotest:daodaotest copy-job dev-daodaotest sit-daodaotest

Third: REST API

Like the second, only key commands are introduced. Take python-jenkinsapi as an example here .

python-jenkinsOfficial website address: https://opendev.org/jjb/python-jenkins

Install Python Jenkins

# 安装 pip
$ sudo yum install epel-release && sudo yum install python-pip

# 安装 python-jenkins
$ pip install python-jenkins

Get the task name in the view

import jenkins

server = jenkins.Jenkins('http://localhost:8080', username='daodaotest', password='daodaotest')

# 查询 dev 视图下的所有任务
jobs = server.get_jobs(folder_depth=0, view_name='dev')

# 循环打印任务名称
for job in jobs:
    print(job['fullname'])

Copy task

import jenkins

server = jenkins.Jenkins('http://localhost:8080', username='daodaotest', password='daodaotest')

# 任务是否存在，True 为存在，Fasle 为不存在
print(server.job_exists('dev-daodaotest'))

# 复制任务
server.copy_job('dev-daodaotest','sit-daodaotest')

# 打印任务信息
jobinfo = server.get_job_info('sit-daodaotest')
print(jobinfo)

请求报错 "Error 403 No valid crumb was included in the request"

Cause of error : jenkins placed a token named .crumb in the http request header. After using a reverse proxy and checking "Prevent Cross Site Request Forgery exploits" in the jenkins settings, this token will be removed by the forwarding server nginx as an illegal header, resulting in a jump failure.

Solution : Cancel "Prevent Cross Site Request Forgery exploits" in Jenkins' security settings.