Ansible模块开发-自定义模块
第一步创建ansible自定义模块路径
cd /data/db/playbooks/
mkdir -p library
vim ansible.cfg 增加如下内容:
[defaults]
library = ./library
下面我们开始第一个模块开发
创建第一个模块
vim library/info.py
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2018/11/12 12:00 PM
# @Author : biglittle
# @Contact : biglittleant@hotmail.com
# @Site :
# @File : info.py
# @Software: PyCharm
# @Desc : python file
# @license : Copyright(C), Your Company
from ansible.module_utils.basic import *
# 实例化一个module，因为不需要参数所以argument_spec初始化参数为空字典。
module = AnsibleModule(
argument_spec = dict(),
)
output="hello word!"
result = dict(module='myinfo',stdout=output,changed=False,rc=0)
module.exit_json(**result)
ansible -i inventory/devlop linux-node1 -m myinfo
linux-node1 | SUCCESS => {
"changed": false,
"module": "myinfo",
"rc": 0,
"stdout": "hello word!",
"stdout_lines": [
"hello word!"
]
}
创建一个带参数的脚本
vim library/myinfo_args.py
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2018/11/12 12:00 PM
# @Author : biglittle
# @Contact : biglittleant@hotmail.com
# @Site :
# @File : myinfo_args.py
# @Software: PyCharm
# @Desc : python file
# @license : Copyright(C), Your Company
from ansible.module_utils.basic import *
module = AnsibleModule(
argument_spec = dict(
msg=dict(required=True),
),
)
msg = module.params['msg']
result = dict(module='myinfo_args',stdout=msg,changed=False,rc=0)
module.exit_json(**result)
执行验证
ansible -i inventory/devlop linux-node1 -m myinfo_args -a "msg='new word'"
linux-node1 | SUCCESS => {
"changed": false,
"module": "myinfo_args",
"rc": 0,
"stdout": "new word",
"stdout_lines": [
"new word"
]
}
来个不带参数的的执行
ansible -i inventory/devlop linux-node1 -m myinfo_args
linux-node1 | FAILED! => {
"changed": false,
"msg": "missing required arguments: msg"
}
自己实现一个shell模块
vim library/myshell.py
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2020年01月10日16:39:09
# @Author : biglittle
# @Contact : biglittleant@hotmail.com
# @Site :
# @File : myshell.py
# @Software: PyCharm
# @Desc : python file
# @license : Copyright(C), Your Company
from ansible.module_utils.basic import AnsibleModule
import commands
def main():
"""
run shell
"""
changed = False
module = AnsibleModule(
argument_spec = dict(
cmd = dict(type='str', required=True),
),
)
cmd = module.params['cmd']
code,output = commands.getstatusoutput(cmd)
if code == 0:
# 按照ansible 的返回格式定义返回内容,stdout为标准输出,changed代表系统有没有东西被变更,rc=0代表执行成功
result = dict(stdout=output,changed=changed,rc=0)
# 使用ansible规则的module实例下的exit_json返回正常内容
module.exit_json(**result)
else:
# 当调用失败返回错误信息的时候,数据字典只要传递msg信息就可了,然后调用module实例的fail_json方法给返回
result = dict(msg=output,rc=code)
module.fail_json(**result)
if __name__ == '__main__':
main()
执行一个正确的命令
ansible -i inventory/devlop linux-node1 -m myshell -a cmd='pwd'
linux-node1 | SUCCESS => {
"changed": false,
"rc": 0,
"stdout": "/home/niu",
"stdout_lines": [
"/home/niu"
]
}
ansible -i inventory/devlop linux-node1 -m myshell -a cmd='pws'
linux-node1 | FAILED! => {
"changed": false,
"msg": "sh: pws: command not found",
"rc": 32512
}
如果不定义: result = dict(msg=output,rc=code) ansible 会有一个默认的返回，输出类似下面的情况。
ansible -i inventory/devlop linux-node1 -m myshell -a cmd='pws'
linux-node1 | FAILED! => {
"changed": false,
"module_stderr": "Shared connection to linux-node1 closed.\r\n",
"module_stdout": "\r\nTraceback (most recent call last):\r\n File \"/home/niu/.ansible/tmp/ansible-tmp-1578648957.53-161256530271090/AnsiballZ_myshell.py\", line 113, in <module>\r\n _ansiballz_main()\r\n File \"/home/niu/.ansible/tmp/ansible-tmp-1578648957.53-161256530271090/AnsiballZ_myshell.py\", line 105, in _ansiballz_main\r\n invoke_module(zipped_mod, temp_path, ANSIBALLZ_PARAMS)\r\n File \"/home/niu/.ansible/tmp/ansible-tmp-1578648957.53-161256530271090/AnsiballZ_myshell.py\", line 48, in invoke_module\r\n imp.load_module('__main__', mod, module, MOD_DESC)\r\n File \"/tmp/ansible_myshell_payload_XylgIF/__main__.py\", line 40, in <module>\r\n File \"/tmp/ansible_myshell_payload_XylgIF/__main__.py\", line 33, in main\r\nUnboundLocalError: local variable 'result' referenced before assignment\r\n",
"msg": "MODULE FAILURE\nSee stdout/stderr for the exact error",
"rc": 1
}
参数解释
argument_spec 支持的参数
例子：
module = AnsibleModule(
argument_spec = dict{
name = dict(type='str', required=True),
cwd = dict(type='str', required=False),
shell = dict(type='bool', default=True),
}
)
官方的ping模块分析
模块路径：https://github.com/ansible/ansible/blob/devel/lib/ansible/modules/system/ping.py
from ansible.module_utils.basic import AnsibleModule
def main():
# 实例化了一个类
module = AnsibleModule(
argument_spec=dict(
data=dict(type='str', default='pong'),
),
supports_check_mode=True
)
# 判断参数是否为crash，如果是就抛出异常
if module.params['data'] == 'crash':
raise Exception("boom")
# 正常情况下，定义个字典。ping=data,data默认是pong
result = dict(
ping=module.params['data'],
)
# 返回结果
module.exit_json(**result)
if __name__ == '__main__':
main()
ansible -i inventory/devlop linux-node1 -m ping
linux-node1 | SUCCESS => {
"changed": false,
"ping": "pong"
}
ansible -i inventory/devlop linux-node1 -m ping -a "data=abcd"
linux-node1 | SUCCESS => {
"changed": false,
"ping": "abcd"
}
ansible模块中的输出定制
changed: 平时我们使用别人的模块,会发现changed字段,有的时候是true,有的时候是false,其实changed并不代表什么,ansible里面一致认为的是对系统进行了更改的changed为true,未更改的为false,其实只是一个记录的值而已,要改变chnged的值,返回的字典里面只要changed=Fasle就好了result = dict(changed=False,stdout=ouput) 。
ansible模块中的退出状态处理
-
正常退出:module.exit_jons
-
错误退出:module.fail_json
错误退出比较不一样的是,你要传递的参数是msg: result = dict(msg=output,rc=code)
报错汇总
ERROR! this task 'myinfo_args' has extra params, which is only allowed in the following modules: shell, win_shell, include_vars, add_host, raw, include_role, meta, set_fact, include, import_tasks, script, import_role, include_tasks, group_by, command, win_command
命令执行错了 ansible -i inventory/devlop linux-node1 -m myinfo_args -a 'new word'
正确的命令：ansible -i inventory/devlop linux-node1 -m myinfo_args -a "msg='new word'"