首页 > 编程语言 > Python-jenkins模块获取jobs的执行状态操作
2020
09-29

Python-jenkins模块获取jobs的执行状态操作

1 获取jobs的当前任务状态

server_1 = jenkins.Jenkins('http://%s:%s@192.168.37.134:8081/',username, password)

获取状态前先确认2019文件夹下的get_node_list任务是否存在:

server_1.assert_job_exists('2019/get_node_list')

获取最后一次完成(不包括执行中的)的job任务执行number:

server_1.get_job_info('2019/get_node_list')['lastCompletedBuild']['number']

查看job状态(SUCCESS/FAILURE/ABORTED):

server_1.get_build_info('2019/get_node_list',3)['result']
server_1.get_build_console_output('2019/get_node_list',7).split('\n')[-2].split(':')[-1].strip()

启动jobs:

server_1.build_job('2019/get_node_list')

在job执行结束前使用server_1.get_build_console_output(‘2019/get_node_list',7).split('\n')[-2].split(':')[-1].strip()获取的状态信息不符合预期。

job状态应该还包括running,pending状态,那么获取job的当前状态正确姿势如下:

job_name = '2019/get_node_list'
def get_jobs_status(job_name,server):
  try:
    server.assert_job_exists(job_name)
  except Exception as e:
    print(e)
    job_statue = '1'
  #判断job是否处于排队状态
  inQueue = server.get_job_info(job_name)['inQueue']
  if str(inQueue) == 'True':
    job_statue = 'pending'
    running_number = server.get_job_info(job_name)['nextBuildNumber']  
  else:
    #先假设job处于running状态,则running_number = nextBuildNumber -1 ,执行中的job的nextBuildNumber已经更新
    running_number = server.get_job_info(job_name)['nextBuildNumber'] -1
    try:
      running_status = server.get_build_info(job_name,running_number)['building']
      if str(running_status) == 'True':
        job_statue = 'running'
      else:
        #若running_status不是True说明job执行完成
        job_statue = server.get_build_info(job_name,running_number)['result']
    except Exception as e:
      #上面假设job处于running状态的假设不成立,则job的最新number应该是['lastCompletedBuild']['number']
      lastCompletedBuild_number = server.get_job_info(job_name)['lastCompletedBuild']['number']
      job_statue = server.get_build_info(job_name,lastCompletedBuild_number)['result']

  return job_statue,running_number

注意:

可能还存在下图的情况,这个时候获取的是26的状态,这时候也许你想获取25的状态,26是不小心误操作触发的,这个时候任务的最新状态也许就无法满足预期要求,或者是支持并发构建的job场景中就不适用了,关键还是需要结合应用场景制定对应的方案。

2 统计jobs的执行成功率和平均执行时间

统计场景说明:

设计了一个统计job执行成功率的工程,主要从执行时间以及视图两个维度来划定需要统计的jobs及jobs对应的运行范围。

在这里我在job里面添加了DAYS和VIEWS两个参数:

**DAYS:**默认统计最近一天的运行情况,如果执行的时候输入的是0则代表统计所有的运行情况。

**VIEWS:**对应的是视图名称,“2019-1,test”代表统计这两个视图的运行情况

对应的视图如下:

执行成功后以表格形式列出统计的数据,表头如下

列出了序号、视图名称、job名称、job执行成功的平均执行时间、job执行成功次数、总的执行时间、job执行成功率

job执行演示:

执行构建时配置的参数如下

job_data任务的主要执行内容如下:

执行成功后查看HTML_Report统计的数据如下:

get_job_data.py源码如下:

#!/usr/bin/env python
# -*- coding:utf-8 -*-
# author: Sudley
# ctime: 2020/02/12

import sys
import jenkins
import time
from dominate.tags import *

def Count_the_success_rate_of_jobs(days,views):
  username = 'sudley'
  password = '******'

  with open('//home/Sudley/python-jenkins/get_job_data.txt','w') as f:
    print('create a new file //home/Sudley/python-jenkins/get_job_data.txt')
  serial_number = 0  #统计任务的累计序号

  for view in views.split(','):
    #根据视图名称拼接视图的URL,多个视图间用','分隔
    URL = ('http://%s:%s@192.168.37.134:8081/job/2019/view/%s/')%(username, password, view)
    server = jenkins.Jenkins(URL)
    #依次获取当前view视图中jobs的信息
    for num in range(0,len(server.get_all_jobs())):
      job_name = server.get_all_jobs()[num]['fullname']
      #获取最后一次完成构建的编号,用于划定时间范围(如果需要的话)
      try:
        lastCompletedBuild_num = server.get_job_info(job_name)['lastCompletedBuild']['number']
      except:
        #假如job下面一个构建记录都没有则补0
        print('There is not build number in',job_name)
        average_success_duration = success_count = all_count = success_rate = 0
        line = str(serial_number) + ' ' + view + ' ' + job_name + ' ' + str(int(average_success_duration)) + ' ' + str(success_count) + ' ' + str(all_count) + ' ' + str(success_rate) + '%'
        with open('//home/Sudley/python-jenkins/get_job_data.txt','a') as f:
          f.write(str(line))
          f.write('\n')
        serial_number = serial_number + 1
        continue
      #获取最后一次完成构建的时间戳,单位由毫秒转换为秒
      lastCompletedBuild_timestamp = server.get_build_info(job_name,lastCompletedBuild_num)['timestamp'] / 1000
      #将时间先由秒转化为元组在转化为字符串并取到天数
      lastCompletedBuild_date = time.strftime("%Y%m%d",time.localtime(lastCompletedBuild_timestamp))
      #print(lastCompletedBuild_date)
      #根据变量days和lastCompletedBuild_timestamp计算出days天前的日期,若days为0则没有日期限制,统计之前运行的所有任务
      if str(days) == '0':
        end_date = 'false'
      else:
        end_timestamp = float(lastCompletedBuild_timestamp) - float(days) * 24 * 3600
        end_date = time.strftime("%Y%m%d",time.localtime(end_timestamp))
      #print(end_date)

      #获取days天内job的执行情况
      success_count = 0        #job执行成功的总数
      success_duration = 0      #执行成功的job执行时间之和,单位是s
      for number in range(0,len(server.get_job_info(job_name)['builds'])):
        job_build_number = server.get_job_info(job_name)['builds'][number]['number']
        job_build_timestamp = server.get_build_info(job_name,job_build_number)['timestamp'] / 1000
        job_build_date = time.strftime("%Y%m%d",time.localtime(job_build_timestamp))
        #如果日期和end_date相同则终止此job数据的累计
        if job_build_date == end_date:
          number = number - 1
          break
        #累计执行成功的次数和duration执行时间
        job_build_result = server.get_build_info(job_name,job_build_number)['result']
        if str(job_build_result) == 'SUCCESS':
          job_build_duration = server.get_build_info(job_name,job_build_number)['duration']
          success_duration = success_duration + job_build_duration / 1000
          success_count = success_count + 1

      #计算执行成功的平均执行时间和成功率,打印关键信息
      all_count = number + 1
      success_rate = success_count * 1.0 / all_count * 100
      if success_count == 0:
        average_success_duration = success_duration
      else:
        average_success_duration = success_duration * 1.0 / success_count
      #将关心的数据按照一定的格式写到/home/Sudley/python-jenkins/get_job_data.txt文件中
      line = str(serial_number) + ' ' + view + ' ' + job_name + ' ' + str(int(average_success_duration)) + ' ' + str(success_count) + ' ' + str(all_count) + ' ' + str(round(success_rate,2)) + '%'
      with open('//home/Sudley/python-jenkins/get_job_data.txt','a') as f:
        f.write(str(line))
        f.write('\n')
      serial_number = serial_number + 1

def txt2xml():
  h = html()
  with h.add(body()):
    h2('job执行效率统计')
    caption('summary:')
    with table(border="2",cellspacing="0"):
      l = tr(bgcolor="#0000FF")
      l += th('序号')
      l += th('view_name')
      l += th('job_name')
      l += th('average_success_duration')
      l += th('success_count')
      l += th('all_count')
      l += th('success_rate')

      file=open('/home/Sudley/python-jenkins/get_job_data.txt')
      for line in file.readlines():
        curLine=line.strip().split(" ")
        l = tr()
        for i in range(0,len(curLine)):
          l += td(curLine[i])

  with open('/home/Sudley/python-jenkins/get_job_data.html','w') as f:
    f.write(h.render())

if __name__ == '__main__' :
  days = sys.argv[1]
  views = sys.argv[2]
  Count_the_success_rate_of_jobs(days,views)
  txt2xml()

以上这篇Python-jenkins模块获取jobs的执行状态操作就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持自学编程网。

编程技巧