Python automated operation and maintenance and deployment project tool Fabric usage examples

Fabric is a Python library. As long as the target machine supports ssh access, you can use fabric to perform remote operations (such as running shell commands remotely on host1 on host2). Obviously, since fabric is a Python package, other Python packages are It can be imported into fabric's unique fabfile.py script

Fabric is a good tool for automated operation and maintenance and deployment projects developed using Python. It can automatically interact with remote servers through SSH, such as The local file is transferred to the server and the shell command is executed on the server.

The following is an example of automated deployment of a Django project

# -*- coding: utf-8 -*-
# 文件名要保存为 fabfile.py

from __future__ import unicode_literals
from fabric.api import *

# 登录用户和主机名：
env.user = 'root'
# 如果没有设置，在需要登录的时候，fabric 会提示输入
env.password = 'youpassword'
# 如果有多个主机，fabric会自动依次部署
env.hosts = ['www.example.com']

TAR_FILE_NAME = 'deploy.tar.gz'

def pack():
  """
  定义一个pack任务, 打一个tar包
  :return:
  """
  tar_files = ['*.py', 'static/*', 'templates/*', 'vue_app/', '*/*.py', 'requirements.txt']
  exclude_files = ['fabfile.py', 'deploy/*', '*.tar.gz', '.DS_Store', '*/.DS_Store',
           '*/.*.py', '__pycache__/*']
  exclude_files = ['--exclude=\'%s\'' % t for t in exclude_files]
  local('rm -f %s' % TAR_FILE_NAME)
 
  local('tar -czvf %s %s %s' % (TAR_FILE_NAME, ' '.join(exclude_files), ' '.join(tar_files)))
  print('在当前目录创建一个打包文件: %s' % TAR_FILE_NAME)


def deploy():
  """
  定义一个部署任务
  :return:
  """
  # 先进行打包
  pack()

  # 远程服务器的临时文件
  remote_tmp_tar = '/tmp/%s' % TAR_FILE_NAME
  run('rm -f %s' % remote_tmp_tar)
  # 上传tar文件至远程服务器, local_path, remote_path
  put(TAR_FILE_NAME, remote_tmp_tar)
  # 解压
  remote_dist_base_dir = '/home/python/django_app'
  # 如果不存在, 则创建文件夹
  run('mkdir -p %s' % remote_dist_dir)

 # cd 命令将远程主机的工作目录切换到指定目录 
  with cd(remote_dist_dir):
    print('解压文件到到目录: %s' % remote_dist_dir)
    run('tar -xzvf %s' % remote_tmp_tar)
    print('安装 requirements.txt 中的依赖包')
    # 我使用的是 python3 来开发
    run('pip3 install -r requirements.txt')
    remote_settings_file = '%s/django_app/settings.py' % remote_dist_dir
    settings_file = 'deploy/settings.py' % name
    print('上传 settings.py 文件 %s' % settings_file)
    put(settings_file, remote_settings_file)

    nginx_file = 'deploy/django_app.conf'
    remote_nginx_file = '/etc/nginx/conf.d/django_app.conf'
    print('上传 nginx 配置文件 %s' % nginx_file)
    put(nginx_file, remote_nginx_file)
 
 # 在当前目录的子目录 deploy 中的 supervisor 配置文件上传至服务器
  supervisor_file = 'deploy/django_app.ini'
  remote_supervisor_file = '/etc/supervisord.d/django_app.ini'
  print('上传 supervisor 配置文件 %s' % supervisor_file)
  put(supervisor_file, remote_supervisor_file)
 
 # 重新加载 nginx 的配置文件
  run('nginx -s reload')
  run('nginx -t')
  # 删除本地的打包文件
  local('rm -f %s' % TAR_FILE_NAME)
  # 载入最新的配置文件，停止原有进程并按新的配置启动所有进程
  run('supervisorctl reload')
  # 执行 restart all，start 或者 stop fabric 都会提示错误，然后中止运行
  # 但是服务器上查看日志，supervisor 有重启
  # run('supervisorctl restart all')

Execute pack task

fab pack<br>

Execute deploy task

fab deploy

Let me share with you an automated deployment of code using Fabric

#coding=utf-8
from fabric.api import local, abort, settings, env, cd, run
from fabric.colors import *
from fabric.contrib.console import confirm

env.hosts = ["root@115.28.×××××"]
env.password = "×××××"


def get_git_status():
  git_status_result = local("git status", capture=True)
  if "无文件要提交，干净的工作区" not in git_status_result:
    print red("****当前分支还有文件没有提交")
    print git_status_result
    abort("****已经终止")


def local_unit_test():
  with settings(warn_only=True):
    test_result = local("python manage.py test")
    if test_result.failed:
      print test_result
      if not confirm(red("****单元测试失败，是否继续？")):
        abort("****已经终止")


def server_unit_test():
  with settings(warn_only=True):
    test_result = run("python manage.py test")
    if test_result.failed:
      print test_result
      if not confirm(red("****单元测试失败，是否继续？")):
        abort("****已经终止")


def upload_code():
  local("git push origin dev")
  print green("****代码上传成功")


def deploy_at_server():
  print green("****ssh到服务器进行下列操作")
  with cd("/var/www/××××××"):
    #print run("pwd")
    print green("****将在远程仓库下载代码")
    run("git checkout dev")
    get_git_status()
    run("git pull origin dev")
    print green("****将在服务器上运行单元测试")
    server_unit_test()
    run("service apache2 restart", pty=False)
    print green("****重启apache2成功")
    print green("********代码部署成功********")


def deploy():
  get_git_status()
  local("git checkout dev", capture=False)
  print green("****切换到dev分支")
  get_git_status()
  print green("****将开始运行单元测试")
  local_unit_test()
  print green("****单元测试完成，开始上传代码")
  upload_code()
  deploy_at_server()

fabric can solidify the commands for automated deployment or multi-machine operation into a script, thereby reducing manual operations. The above is written after I came into contact with this thing for the first time today. It is indeed very practical. Just run fab deploy.

The main logic is to run the unit test on the local dev branch, then submit it to the server, log in to the server via ssh, pull it down, run the unit test again, and then restart apache2. The first time I write it, it may be relatively simple, and I will continue to improve it.

For more Python automated operation and maintenance and deployment project tool Fabric usage examples related articles, please pay attention to the PHP Chinese website!