运维工作太忙？一文详解Ansible自动化运维，轻松提升工作效率！
什么是自动化批量管理
在日常企业运维工作中，经常遇到有多台主机需要进行管理操作，并且操作的任务还都是相同的，这样重复性的操作任务会大大增加运维人员的工作量。
因此可以通过软件工具，将重复性的工作任务，进行批量的自动化完成，从而形成高效运维的管理体系。
注意事项：
想要自动化一定要先标准化(所有的环境、软件、目录、一致)；
尽量进行分组(比如:所有web服务器是1组).
自动化工具选择
Ansible架构
Ansible极速使用指南
环境准备
[root@m01 ~]# cat /server/scripts/ssh_check.sh
#!/bin/bash
#author: oldboy lidao996
#desc 批量检查脚本
ip_list="10.0.0.7 10.0.0.31 10.0.0.41 10.0.0.51"
echo '--------------------------------------------'
echo '批量执行命令'
echo '--------------------------------------------'
for ip in $ip_list
do
ssh root@$ip $@
done
[root@m01 ~]# cat /server/scripts/fenfa.sh
#!/bin/bash
#author: oldboy lidao996
#desc 一键自动化创建和分发公钥
ip_list="10.0.0.7 10.0.0.31 10.0.0.41 10.0.0.51"
echo '--------------------------------------------'
echo '1. 创建 key'
echo '--------------------------------------------'
ssh-keygen -f ~/.ssh/id_rsa -P ''
echo '--------------------------------------------'
echo '2. 分发 pub key'
echo '--------------------------------------------'
for ip in $ip_list
do
sshpass -p1 ssh-copy-id -i ~/.ssh/id_rsa.pub -o StrictHostKeyChecking=no root@$ip
done
[root@m01 ~]# sh /server/scripts/ssh_check.sh hostname
--------------------------------------------
批量执行命令
--------------------------------------------
web01
nfs01
backup
db01
Ansible部署
yum install -y ansible
#epel源中的软件包
初步配置主机清单（inventory）
[root@m01 ~]# cat /etc/ansible/hosts
[oldboy]
172.16.1.7
172.16.1.31
172.16.1.41
172.16.1.51与Ansible的第1次接触
ansible oldboy -m pingansible oldboy -m ping
ansible oldboy -m command -a 'hostname'
ansible oldboy -m command -a 'hostname -I'
ansible oldboy -m command -a 'crontab -l'Ansible配置文件
/etc/ansible/ansible.cfg #ansible配置文件 configure
[root@m01 ~]# grep -n '^host_key' /etc/ansible/ansible.cfg
71:host_key_checking = False
/etc/ansible/hosts #ans主机清单,默认是.Ansible主机清单
基本格式
[root@m01 ~]# cat /etc/ansible/hosts
[oldboy] #[组的名字]
172.16.1.7
172.16.1.31
172.16.1.41
172.16.1.51
[web]
172.16.1.7
[db]
172.16.1.51
[nfs]
172.16.1.31
[backup]
172.16.1.41
指定用户名、密码、端口
[web]
172.16.1.7 ansible_ssh_port=22 ansible_ssh_user=root ansible_ssh_pass='1'
[db]
172.16.1.51
[nfs]
172.16.1.31
[backup]
172.16.1.41子组
我想把多个主机分组进行合并,比如把db,nfs,backup合并成一个组叫data组.
[web]
172.16.1.7
[db]
172.16.1.51
[nfs]
172.16.1.31
[backup]
172.16.1.41
[data:children] #data:children表示 data是创建的子组 组里面包含 db,nfs,backup3个组.
db
nfs
backup主机清单案例
[root@m01 ~]# cat /etc/ansible/hosts
[web]
172.16.1.7
[nfs]
172.16.1.31
[backup]
172.16.1.41
[db]
172.16.1.51
[data:children]
db
nfs
backup
案例01 查看与使用指定的子组 data
[root@m01 ~]# ansible data -m command -a 'hostname'
172.16.1.31 | CHANGED | rc=0 >>
nfs01
172.16.1.51 | CHANGED | rc=0 >>
db01
172.16.1.41 | CHANGED | rc=0 >>
backup
案例02：查看指定的组db
[root@m01 ~]# ansible db -m command -a 'hostname'
172.16.1.51 | CHANGED | rc=0 >>
db01
案例03：查看所有主机情况
[root@m01 ~]# ansible all -m command -a 'hostname'
172.16.1.51 | CHANGED | rc=0 >>
db01
172.16.1.41 | CHANGED | rc=0 >>
backup
172.16.1.31 | CHANGED | rc=0 >>
nfs01
172.16.1.7 | CHANGED | rc=0 >>
web01
[root@m01 ~]#
案例04：查看某一台机器
[root@m01 ~]# ansible 172.16.1.7 -m command -a 'hostname'
172.16.1.7 | CHANGED | rc=0 >>
web01
[root@m01 ~]#Ansible-命令与脚本类模块
1）command模块
仅支持简单命令，不支持特殊符号，管道。。。
这个模块是默认模块，ansible不加上模块，默认就使用这个模块
ansible all -m command -a '命令'
ansible all -a 'hostname' #相当于省略 -m command
2）shell模块
[root@m01 ~]# ansible all -m command -a 'ip a s eth0 |sed -n 3p'
172.16.1.7 | FAILED | rc=255 >>
Error: either "dev" is duplicate, or "|sed" is a garbage.non-zero return code
172.16.1.51 | FAILED | rc=255 >>
Error: either "dev" is duplicate, or "|sed" is a garbage.non-zero return code
172.16.1.31 | FAILED | rc=255 >>
Error: either "dev" is duplicate, or "|sed" is a garbage.non-zero return code
172.16.1.41 | FAILED | rc=255 >>
Error: either "dev" is duplicate, or "|sed" is a garbage.non-zero return code
[root@m01 ~]# ansible all -m shell -a 'ip a s eth0 |sed -n 3p'
172.16.1.31 | CHANGED | rc=0 >>
inet 10.0.0.31/24 brd 10.0.0.255 scope global eth0
172.16.1.41 | CHANGED | rc=0 >>
inet 10.0.0.41/24 brd 10.0.0.255 scope global eth0
172.16.1.7 | CHANGED | rc=0 >>
inet 10.0.0.7/24 brd 10.0.0.255 scope global eth0
172.16.1.51 | CHANGED | rc=0 >>
inet 10.0.0.51/24 brd 10.0.0.255 scope global eth0
长按扫码免费领取
配套PDF版资料
3）script模块-传输脚本到被管理端并执行脚本
安装ipvsadm
[root@m01 ~]# cat /server/scripts/yum.sh
yum install -y ipvsadm
使用script 模块执行脚本.
ansible db -m script -a '/server/scripts/yum.sh'4）命令与脚本模块
Ansible-文件与目录管理模块
1）file模块
管理文件或目录、软链接。
案例01-创建目录/oldboy/目录
ansible all -m file -a 'path=/oldboy state=directory'
ansible all -a 'ls -ld /oldboy/'案例02-创建文件/oldboy/oldboy.txt文件
ansible all -m file -a 'path=/oldboy/oldboy.txt state=touch'
ansible all -a 'ls -l /oldboy/'案例03-创建软连接 /oldboy/oldboy.txt 到/tmp/oldboy.txt.soft
ansible all -m file -a 'src=/oldboy/oldboy.txt path=/tmp/oldboy.txt.soft state=link'
[root@m01 ~]#
[root@m01 ~]#
[root@m01 ~]# ansible all -a 'ls -l /tmp/oldboy.txt.soft'
172.16.1.31 | CHANGED | rc=0 >>
lrwxrwxrwx 1 root root 18 Apr 12 12:01 /tmp/oldboy.txt.soft -> /oldboy/oldboy.txt
172.16.1.41 | CHANGED | rc=0 >>
lrwxrwxrwx 1 root root 18 Apr 12 12:01 /tmp/oldboy.txt.soft -> /oldboy/oldboy.txt
172.16.1.51 | CHANGED | rc=0 >>
lrwxrwxrwx 1 root root 18 Apr 12 12:01 /tmp/oldboy.txt.soft -> /oldboy/oldboy.txt
172.16.1.7 | CHANGED | rc=0 >>
lrwxrwxrwx 1 root root 18 Apr 12 12:01 /tmp/oldboy.txt.soft -> /oldboy/oldboy.txt
案例04-删除文件/目录/软连接
ansible all -m file -a 'path=/oldboy/oldboy.txt state=absent ' #删除文件
ansible all -m file -a 'path=/oldboy state=absent ' #删除目录
ansible all -m file -a 'path=/tmp/oldboy.txt.soft state=absent ' #删除软连
案例05-创建文件/tmp/oldboy.txt,所有者root,用户组root,权限755
ansible all -m file -a 'path=/tmp/oldboy.txt owner=root group=root mode=755 state=touch'
ansible all -a 'ls -l /tmp/oldboy.txt'
2）copy远程传输模块
案例01-传输/etc/hosts文件到/etc/hosts
ansible all -m copy -a 'src=/etc/hosts dest=/etc/hosts '
案例02-传输/etc/hosts文件到/etc/hosts-先备份然后修改
ansible all -m copy -a 'src=/etc/hosts dest=/etc/hosts backup=yes'
[root@m01 ~]# ansible all -m shell -a 'ls -l /etc/hosts*'
172.16.1.41 | CHANGED | rc=0 >>
-rw-r--r-- 1 root root 328 Apr 12 12:24 /etc/hosts
-rw-r--r-- 1 root root 327 Apr 12 12:21 /etc/hosts.81057.2022-04-12@12:24:20~
-rw-r--r--. 1 root root 370 Jun 7 2013 /etc/hosts.allow
-rw-r--r--. 1 root root 460 Jun 7 2013 /etc/hosts.deny
172.16.1.7 | CHANGED | rc=0 >>
-rw-r--r-- 1 root root 328 Apr 12 12:24 /etc/hosts
-rw-r--r-- 1 root root 327 Apr 12 12:21 /etc/hosts.18115.2022-04-12@12:24:20~
-rw-r--r--. 1 root root 370 Jun 7 2013 /etc/hosts.allow
-rw-r--r--. 1 root root 460 Jun 7 2013 /etc/hosts.deny
172.16.1.31 | CHANGED | rc=0 >>
-rw-r--r-- 1 root root 328 Apr 12 12:24 /etc/hosts
-rw-r--r-- 1 root root 327 Apr 12 12:21 /etc/hosts.46979.2022-04-12@12:24:20~
-rw-r--r--. 1 root root 370 Jun 7 2013 /etc/hosts.allow
-rw-r--r--. 1 root root 460 Jun 7 2013 /etc/hosts.deny
172.16.1.51 | CHANGED | rc=0 >>
-rw-r--r-- 1 root root 328 Apr 12 12:24 /etc/hosts
-rw-r--r-- 1 root root 327 Apr 12 12:21 /etc/hosts.21455.2022-04-12@12:24:20~
-rw-r--r--. 1 root root 370 Jun 7 2013 /etc/hosts.allow
-rw-r--r--. 1 root root 460 Jun 7 2013 /etc/hosts.deny
3）文件管理与传输模块小结
Ansible-服务管理模块
1）systemd
案例01-关闭firewalld
ansible all -m systemd -a 'name=firewalld enabled=no state=stopped'
ansible all -a 'systemctl status firewalld'
案例02-开启sshd服务
ansible all -m systemd -a 'name=sshd enabled=yes state=started'
ansible all -a 'systemctl status sshd'
案例03-重启backup这台机器上面的rsync服务
ansible backup -m systemd -a 'name=rsyncd state=restarted'
2）service 了解
3）服务管理模块小结
Ansible-软件包管理模块
1）yum源配置管理模块
[root@m01 ~]# cat /etc/yum.repos.d/epel.repo
案例01-批量添加nginx-yum源
[nginx]
name=nginx stable repo
baseurl=http://nginx.org/packages/centos/$releasever/$basearch/
gpgcheck=1
enabled=1
gpgkey=https://nginx.org/keys/nginx_signing.key
ansible web -m yum_repository -a 'name=nginx description="nginx stable repo" baseurl="http://nginx.org/packages/centos/$releasever/$basearch/" gpgcheck=no enabled=yes'
#注意未来也可以通过,copy模块实现.
2）yum模块
案例01-安装lrzsz
ansible all -a 'rpm -e lrzsz '
ansible all -m yum -a 'name=lrzsz state=installed '
案例02-安装sl、cowsay、aalib
ansible web -m yum -a 'name=sl,cowsay,aalib state=installed'
3）get_url模块(wget)
案例01-下载tengine源码包到/server/tools/(不存在)目录下
下载地址: https://tengine.taobao.org/download/tengine-2.3.3.tar.gz
ansible web -m file -a 'path=/server/tools/ state=directory'
ansible web -m get_url -a 'url=https://tengine.taobao.org/download/tengine-2.3.3.tar.gz dest=/server/tools/'
ansible web -a 'tree /server/'
4）软件包管理模块小结
系统管理模块
1）mount模块
案例01-挂载案例
web01把 nfs共享的目录/data目录挂载到 web01的/upload_video
步骤01_web01上面创建挂载点/upload_video
ansible web -m file -a 'path=/upload_video state=directory'
步骤02_挂载nfs
ansible web -m mount -a 'fstype=nfs src="172.16.1.31:/data" path=/upload_video state=mounted '
ansible web -a 'df -h'
ansible web -a 'tail -2 /etc/fstab'
2）cron定时任务模块
案例01-添加自动同步时间的定时任务
#1. sync time lidao996
*/2 * * * * /sbin/ntpdate ntp1.aliyun.com &>/dev/null
步骤01_备份数据
ansible all -a 'cp /var/spool/cron/root /tmp/'
ansible all -a 'ls -l /tmp/root'
ansible all -a 'crontab -r'
步骤02_书写定时任务
ansible all -m cron -a 'name="sync time by lidao996 20221111" minute="*/2" job="/sbin/ntpdate ntp1.aliyun.com &>/dev/null" state=present'
用户管理模块
1）user
案例01-创建用户lidao996
[root@m01 ~]# ansible web -m user -a 'name=lidao996'
案例02-创建虚拟用户tengine,指定uid为10086
useradd -u 10086 -s /sbin/nologin -M tengine
[root@m01 ~]# ansible web -m user -a 'name=tengine uid=10086 shell=/sbin/nologin create_home=no state=present'
172.16.1.7 | CHANGED => {
"ansible_facts": {
"discovered_interpreter_python": "/usr/bin/python"
},
"changed": true,
"comment": "",
"create_home": false,
"group": 10086,
"home": "/home/tengine",
"name": "tengine",
"shell": "/sbin/nologin",
"state": "present",
"system": false,
"uid": 10086
}
[root@m01 ~]# ansible web -a 'id tengine'
172.16.1.7 | CHANGED | rc=0 >>
uid=10086(tengine) gid=10086(tengine) groups=10086(tengine)
[root@m01 ~]# ansible web -a 'grep tengine /etc/passwd'
172.16.1.7 | CHANGED | rc=0 >>
tengine:x:10086:10086::/home/tengine:/sbin/nologin
2）group模块
模块总结
Ansible模块实战-部署rsync服务端
##rsyncd.conf start##
##rsyncd 20221111
fake super = yes
uid = rsync
gid = rsync
use chroot = no
max connections = 200
timeout = 300
pid file = /var/run/rsyncd.pid
lock file = /var/run/rsync.lock
log file = /var/log/rsyncd.log
ignore errors
read only = false
list = false
hosts allow = 172.16.1.0/24
#hosts deny = 0.0.0.0/32
auth users = rsync_backup
secrets file = /etc/rsync.password
[data]
comment = "backup dir by oldboy lidao996"
path = /data
[backup]
comment = "backup dir by oldboy lidao996"
path = /backup
[nfsbackup]
comment = 'nfsbackup dir by lidao996'
path = /nfsbackup
长按扫码免费领取
配套PDF版资料
根据命令及对应操作选择ansible模块
根据选择的模块实现对应的功能
1）服务部署：yum安装或更新
ansible backup -m yum -a 'name=rsync state=lastest'
2）配置文件分发
mkdir -p /server/ans/pro-rsync
准备配置文件存放在 上面目录中 rsyncd.conf
ansible backup -m copy -a 'src=/server/ans/pro-rsync/rsyncd.conf dest=/etc/rsyncd.conf backup=yes'
3）虚拟用户 rsync
ansible backup -m user -a 'name=rsync shell=/sbin/nologin create_home=no state=present'
4）密码文件和权限
创建文件
ansible backup -m file -a 'path=/etc/rsync.password mode=600 state=touch'
增加
ansible backup -m lineinfile -a 'path=/etc/rsync.password line="rsync_backup:1"'
5）模块对应目录，改所有者
ansible backup -m file -a 'path=/data owner=rsync group=rsync state=directory'
6）重启服务
ansible backup -m systemd -a 'name=rsyncd enabled=yes state=started'
7）命令行测试
[root@m01 /server/ans/pro-rsync]# rsync -av /etc/hostname [email protected]::data
8）指定hosts文件的位置
[root@m01 /server/ans/pro-rsync]# tree /server/ans/pro-rsync/
/server/ans/pro-rsync/
├── hosts
└── rsyncd.conf
0 directories, 2 files
[root@m01 /server/ans/pro-rsync]# ansible -i hosts all -m ping
Ansible-playbook（剧本）
ansible ad-hoc vs playbook区别
playbook vs shell脚本
Playbook剧本极速使用指南
剧本格式叫yaml格式yml
缩进，不要用tab键，空格
核心格式剧本中所有的内容要对齐
对齐的时候不能使用tab键
只能使用空格，2个空格
hosts用于指定在哪些主机执行指令
tasks: 用于对于这些主机，运行什么模块及选项
案例01-在所有机器的/tmp下面创建lidao.txt
[root@m01 /server/ans/playbook]# cat 01.touch.yml
- hosts: all
vars:
filename: lidao.txt
tasks:
- name: touch file
shell: touch /tmp/{{ filename }}
cp /etc/ansible/hosts .
ansible-playbook -i hosts 01.touch.yml
[root@m01 /server/ans/playbook]# tree /server/ans/playbook/
/server/ans/playbook/
├── 01.touch.yml
└── hosts
0 directories, 2 files
- hosts: all
tasks:
- name: touch file
shell: touch /tmp/lida.txt
案例02-添加定时同步时间的定时任务
原始命令行的ansible命令：
ansible all -m cron -a 'name="sync time by lidao996 20221111" minute="*/2" job="/sbin/ntpdate ntp1.aliyun.com &>/dev/null" state=present'
ansible -i hosts all -a 'crontab -l'
修改为剧本之后
#简单粗暴版本
---
- hosts: all
tasks:
- name: add cron sync time
cron: name="sync time by lidao996 20221111" minute="*/2" job="/sbin/ntpdate ntp1.aliyun.com &>/dev/null" state=present
#格式优化后
---
- hosts: all
tasks:
- name: add cron sync time
cron:
name: "sync time by lidao996 20221111"
minute: "*/2"
job: "/sbin/ntpdate ntp1.aliyun.com &>/dev/null"
state: present
在剧本中使用模块和选项，选项最好是一行一个选项，选项后面跟着冒号，选项要对其与缩进。
[root@m01 /server/ans/playbook]# ansible -i hosts all -a 'crontab -l'
172.16.1.7 | CHANGED | rc=0 >>
#Ansible: sync time by lidao996 20221111
*/2 * * * * /sbin/ntpdate ntp1.aliyun.com &>/dev/null
172.16.1.31 | CHANGED | rc=0 >>
#Ansible: sync time by lidao996 20221111
*/2 * * * * /sbin/ntpdate ntp1.aliyun.com &>/dev/null
172.16.1.41 | CHANGED | rc=0 >>
#Ansible: sync time by lidao996 20221111
*/2 * * * * /sbin/ntpdate ntp1.aliyun.com &>/dev/null
172.16.1.51 | CHANGED | rc=0 >>
#Ansible: sync time by lidao996 20221111
*/2 * * * * /sbin/ntpdate ntp1.aliyun.com &>/dev/null
案例03-企业案例-批量下载安装zabbix-agent2-6.0客户端并启动
---
- hosts: all
tasks:
- name: 1. download zabbix agent2 rpm
get_url:
ur : https://mirrors.tuna.tsinghua.edu.cn/zabbix/zabbix/6.0/rhel/7/x86_64/zabbix-agent2-6.0.0-1.el7.x86_64.rpm
dest: /tmp/
validate_certs: no
- name: 2. install zabbix agent2 rpm
yu :
name: /tmp/zabbix-agent2-6.0.0-1.el7.x86_64.rpm
state: installed
- name: 3. start zabbix agent2 service
s stemd:
name: zabbix-agent2
enabled: yes
state: started
案例04-部署rsync服务端（剧本）
### 1) 服务部署:yum 安装或更新
ansible backup -m yum -a 'name=rsync state=lastest'
### 2) 配置文件分发
mkdir -p /server/ans/pro-rsync
准备配置文件存放在 上面目录中 rsyncd.conf
ansible backup -m copy -a 'src=/server/ans/pro-rsync/rsyncd.conf dest=/etc/rsyncd.conf backup=yes'
### 3) 虚拟用户 rsync
ansible backup -m user -a 'name=rsync shell=/sbin/nologin create_home=no state=present'
### 4)密码文件和权限
创建文件
ansible backup -m file -a 'path=/etc/rsync.password mode=600 state=touch'
增加
ansible backup -m lineinfile -a 'path=/etc/rsync.password line="rsync_backup:1"'
### 5)模块对应目录,改所有者
ansible backup -m file -a 'path=/data owner=rsync group=rsync state=directory'
### 6) 重启服务
ansible backup -m systemd -a 'name=rsyncd enabled=yes state=started'
[root@m01 /server/ans/playbook]# ansible -i hosts backup -m ping
172.16.1.41 | SUCCESS => {
"ansible_facts": {
"discovered_interpreter_python": "/usr/bin/python"
},
"changed": false,
"ping": "pong"
}
[root@m01 /server/ans/playbook]# cat 05-backup-resyncd.yml
---
- hosts: backup
tasks:
# - name: 1) 服务部署:yum 安装或更新
# yum:
# name: rsync
# state: latest
- name: 2) 配置文件分发
copy:
src: /server/ans/playbook/rsyncd.conf
dest: /etc/rsyncd.conf
backup: yes
- name: 3) 虚拟用户 rsync
user:
name: rsync
shell: /sbin/nologin
create_home: no
state: present
- name: 4) 密码文件和权限
lineinfile:
path: /etc/rsync.password
mode: 0600
line: "rsync_backup:1"
create: yes
- name: 5) 模块对应目录,改所有者
file:
path: /data/
owner: rsync
group: rsync
state: directory
- name: 6) 重启服务
systemd:
name: rsyncd
enabled: yes
state: started
案例05-部署nfs服务端全流程
Ansible变量系列
变量详解
1）vars、vars_files、group_vars
[root@m01 /server/ans/playbook]# cat 07.vars_dir.yml
---
- hosts: all
vars:
dir_name: /oldboy_lidao
file_name: lidao996.icu
tasks:
- name: 01. mkdir
file:
path: "{{ dir_name }}"
state: directory
- name: 02. touch
file:
path: "{{ dir_name }}/{{ file_name }}"
state: touch
把变量存放到一个文件中. 剧本比较大的时候.
[root@m01 /server/ans/playbook]# cat 07.vars_dir.yml
---
- hosts: all
vars:
dir_name: /oldboy_lidao
file_name: lidao996.icu
tasks:
- name: 01. mkdir
file:
path: "{{ dir_name }}"
state: directory
- name: 02. touch
file:
path: "{{ dir_name }}/{{ file_name }}"
state: touch
创建一个变量文件,给某个组共用
用法:需要创建一个group_vars目录.
目录下面创建以主机组命名的目录.
存放变量文件vars.yml
group_vars/ 目录
lb/vars.yml #存放lb组的变量
web/vars.yml #存放web组的变量
data/vars.yml #存放xxx组的变量
all/vars.yml #所有主机共用的变量
案例01-根据不同的主机组创建对应的目录
group_vars/ 目录
web/vars.yml #存放web组的变量
data/vars.yml #存放xxx组的变量
all/vars.yml #所有主机共用的变量
web服务器创建 /app/code/目录
dir_name: /app/code/
data服务端创建 /data/目录
dir_name: /data/
#参考:
##变量文件内容
[root@m01 /server/ans/playbook]# cat group_vars/data/vars.yml
dir_name: /datav2/
[root@m01 /server/ans/playbook]# cat group_vars/web/vars.yml
dir_name: /app/code/
[root@m01 /server/ans/playbook]# tree group_vars/
group_vars/
├── data
│ └── vars.yml
└── web
└── vars.yml
3 directories, 3 files
#剧本内容
[root@m01 /server/ans/playbook]# cat 09.vars_group_vars_dir.yml
- hosts: all
tasks:
- name: 根据主机创建不同的目录
file:
path: "{{ dir_name }}"
state: directory
案例02-使用group_vars的all组定义变量
[root@m01 /server/ans/playbook]# tree group_vars/
group_vars/
├── all
│ └── vars.yml
├── data
│ └── vars.yml
└── web
└── vars.yml
3 directories, 3 files
[root@m01 /server/ans/playbook]# cat group_vars/all/vars.yml
dir_name_code: /app/code/
dir_name_data: /data/
[root@m01 /server/ans/playbook]# cat 09.vars_group_vars_dir.yml
- hosts: all
tasks:
- name: 01 {{ dir_name_code }}
file:
path: "{{ dir_name_code }}"
state: directory
- name: 02 {{ dir_name_data }}
file:
path: "{{ dir_name_data }}"
state: directory
剧本目录,目录结构
[root@m01 /server/ans/playbook]# tree
.
├── 01.touch.yml
├── 02.add-cron.yml
├── 03.add-cron-you.yml
├── 04.install-zabbix-agent2.yml
├── 05-backup-resyncd.yml
├── 06-nfs-server.yml
├── 07.vars_dir.yml
├── 08.vars_files_dir.yml
├── 09.vars_group_vars_dir.yml
├── group_vars
│ ├── all
│ │ └── vars.yml
│ ├── data
│ │ └── vars.yml
│ └── web
│ └── vars.yml
├── hosts
├── rsyncd.conf
└── vars.yml
4 directories, 15 files
变量基础定义小结
2）ansible-facts变量
facts变量说明== : ansible内置变量,执行剧本，有个默认的任务(task)，收集每个主机的基本信息。
#查看 ansible facts变量内容
ansible -i hosts web -m setup
常用fact变量
ansible_hostname #主机名
ansible_memtotal_mb #内存大小(总计) 单位mb
ansible_processor_vcpus #cpu数量
ansible_default_ipv4.address #默认的网卡ip eth0
ansible_distribution #系统发行版本名字 CentOS Ubuntu Debian ...
ansible_processor_cores #核心总数
ansible_date_time.date #当前时间 年-月-日
案例01-系统巡检-获取所有机器的基础信息保存到/tmp/主机名命名文件中
步骤:
01.创建文件
02.写入内容[root@m01 /server/ans/playbook]# cat 10.vars_sys_info.yml
---
- hosts: all
tasks:
- name: 创建文件并写入系统基本信息
lineinfile:
path: /tmp/{{ ansible_hostname }}
create: yes
line: "主机名: {{ ansible_hostname }}\n
ip地址: {{ ansible_default_ipv4.address }}\n
内存总计: {{ ansible_memtotal_mb }}"- hosts: all
gather_facts: no
vars:
dir_name: /oldboy_lidao
file_name: lidao996.icu
tasks:
- name: 01. mkdir
file:
path: "{{ dir_name }}"
state: directory
- name: 02. touch
file:
path: "{{ dir_name }}/{{ file_name }}"
state: touch
3）ansible-register变量
创建压缩包压缩包名字包含时间，tar打包压缩，date获取时间。
tar zcf /tmp/etc-`date +%F`.tar.gz /etc/
案例01-创建以主机名命名文件/opt/主机名
步骤:
01.获取主机名:hostname
02.创建文件,使用上一步的结果
register: 变量名字 #这个变量的内容,叫json格式.
register: hostname #json格式,只想要输出标准输出 stdout standard output 标准输出.
hostname.stdout #取出命令的结果 `hostname`
[root@m01 /server/ans/playbook]# cat 12.vars_register.yml
---
- hosts: all
tasks:
- name: 01.获取主机名
shell: hostname
register: hostname
- name: 输出变量内容
debug:
msg: "{{ hostname }}"
[root@m01 /server/ans/playbook]# cat 12.vars_register.yml
---
- hosts: all
tasks:
- name: 01.获取主机名
shell: hostname
register: hostname
- name: 输出变量内容
debug:
msg: "{{ hostname.stdout }}"
- name: 02. 创建文件
file:
path: /opt/{{ hostname.stdout }}
state: touch
register变量输出结果
{
"msg": {
"changed": true,
"cmd": "hostname",
"delta": "0:00:00.008150",
"end": "2022-04-14 12:32:14.587547",
"failed": false,
"rc": 0, #命令的返回值,0表示正确,非0错误.
"start": "2022-04-14 12:32:14.579397",
"stderr": "", #错误信息
"stderr_lines": [],
"stdout": "backup02", #这个最常用. 命令的结果,输出.
"stdout_lines": [
"backup02"
]
}
}
register: hostname
hostname.stdout #正常输出信息
hostname.rc #取出返回值.
hostname.stderr #取出错误信息.
长按扫码免费领取
配套PDF版资料
Ansible-进阶-剧本调试方法
Debug模块
案例01-调试-nfs服务端部署剧本
[root@m01 /server/ans/playbook]# cat 13-debug-nfs-server.yml
---
- hosts: db
tasks:
- name: 01. 部署nfs服务端软件
yum:
name: nfs-utils
state: installed
- name: 02. 修改配置文件
lineinfile:
path: /etc/exports
line: "/data 172.16.1.0/24(rw)"
state: present
backup: yes
- name: 03. 创建对应的目录,权限
file:
path: /data/
owner: nfsnobody
group: nfsnobody
state: directory
register: file_jieguo
- name: 输出,显示这个过程
debug:
msg: "{{ file_jieguo }}"
- name: 04. 启动服务-rpc服务
systemd:
name: rpcbind
enabled: yes
state: started
- name: 05. 启动服务-nfs服务
systemd:
name: nfs
enabled: yes
state: started
tags标签
一般用于调试剧本，给剧本个每个task可以设置个标签
运行剧本的时候可以运行指定标签
运行剧本的时候排除某些标签
[root@m01 /server/ans/playbook]# cat 14-tags-nfs-server.yml
---
- hosts: db
tasks:
- name: 01. 部署nfs服务端软件
yum:
name: nfs-utils
state: installed
tags:
- install
- name: 02. 修改配置文件
lineinfile:
path: /etc/exports
line: "/data 172.16.1.0/24(rw)"
state: present
backup: yes
tags:
- conf
- conf_file
- name: 03. 创建对应的目录,权限
file:
path: /data/
owner: nfsnobody
group: nfsnobody
state: directory
tags:
- conf
- conf_dir
- name: 04. 启动服务-rpc服务
systemd:
name: rpcbind
enabled: yes
state: started
tags:
- start_srv
- name: 05. 启动服务-nfs服务
systemd:
name: nfs
enabled: yes
state: started
tags:
- start_srv
运行指定的标签
ansible-playbook -i hosts --tags conf 14-tags-nfs-server.yml
ansible-playbook -i hosts --tags conf_file,conf_dir 14-tags-nfs-server.yml
运行剧本的时候排除指定的标签
ansible-playbook -i hosts --skip-tags install,conf_file 14-tags-nfs-server.yml
忽略错误
用于运行剧本的时候，强制让某个任务(模块)运行即使出错了，也不要中断我们的剧本。
[root@m01 /server/ans/playbook]# cat 15-ignore-nfs-server.yml
---
- hosts: db
tasks:
- name: 01. 部署nfs服务端软件
yum:
name: nfs-util
state: installed
ignore_errors: yes
tags:
- install
- name: 02. 修改配置文件
lineinfile:
path: /etc/exports
line: "/data 172.16.1.0/24(rw)"
state: present
backup: yes
tags:
- conf
- conf_file
- name: 03. 创建对应的目录,权限
file:
path: /data/
owner: nfsnobody
group: nfsnobody
state: directory
tags:
- conf
- conf_dir
- name: 04. 启动服务-rpc服务
systemd:
name: rpcbind
enabled: yes
state: started
tags:
- start_srv
- name: 05. 启动服务-nfs服务
systemd:
name: nfs
enabled: yes
state: started
tags:
- start_srv
Ansible-进阶应用
include文件包含：把一个任务分成多个剧本来实现，书写个总剧本文件，通过include_tasks：引用子剧本文件。
子剧本文件中只需要些模块部分(task部分即可)