本文围绕Ansible在Linux服务器运维中的应用，详细介绍了Ansible的循环功能，如with_items迭代列表、with_dict迭代字典等，还阐述了判断功能，包括判断运算符、执行结果、路径、字符串等tests，同时提及判断与错误处理方法，最后给出相关拓展实例。
循环
with_items迭代列表
比如Linux中依次需要安装多个软件包时，我们可以使用with_items迭代功能进行实现
例如：
安装httpd、samba、samba-client软件包时
[student@ansible ansible]$ vim b.yml
---
- name: install packages
hosts: node1
tasks:
- name: yum_repo1
yum_repository:
file: server
name: baseos
description: rhel8
baseurl: file:///mnt/BaseOS
enabled: yes
gpgcheck: no
- name: yum_repo2
yum_repository:
file: server
name: appstream
description: appstream
baseurl: file:///mnt/AppStream
enabled: yes
gpgcheck: no
- name: mount cdrom
mount:
src: /dev/cdrom
path: /mnt
fstype: iso9660
state: mounted
- name: install pks
yum:
name: "{{item}}"
state: present
with_items:
- httpd
- samba
- samba-client
with_dict迭代字典
item.key对应着是字典的键，item.value对应着字典的值
[student@ansible ansible]$ vim c.yml
---
- name: test
hosts: node1
tasks:
- name: debug
debug:
msg: "{{item.key}} & {{item.value}}"
with_dict:
address: 1
netmask: 2
gateway: 3
with_fileglob迭代文件
比如拷贝多个文件到受控主机上时，可以使用
[student@ansible ansible]$ vim d.yml
---
- name: test
hosts: node1
tasks:
- name: cp file
copy:
src: "{{item}}"
dest: /tmp/
with_fileglob:
- /tmp/*.sh
- /tmp/*.py
with_lines迭代行
With_lines可以将命令行的输出结果按行迭代
[student@ansible ansible]$ vim e.yml
---
- name: test
hosts: node1
tasks:
- name: cp file
copy:
src: "{{item}}"
dest: /tmp/
with_lines:
- find /etc/ansible -name "*.yml"
with_nested嵌套迭代
a和b分别与1、2、3连接组合
[student@ansible ansible]$ vim f.yml
---
- name: test
hosts: node1
tasks:
- name: debug
debug:
msg: "{{item[0]}} & {{item[1]}}"
with_nested:
- [a,b]
- [1,2,3]
with_sequence 排序列(start是从什么开始，end结束 stride每隔多少)
[student@ansible ansible]$ vim g.yml
---
- name: test
hosts: node1
tasks:
- name: debug
debug:
msg: "{{item}}"
with_sequence:
start=1
end=5
stride=1
with_random_choice 随机获得列表中的一个值
[student@ansible ansible]$ vim h.yml
---
- name: test
hosts: node1
tasks:
- name: debug
debug:
msg: "{{item}}"
with_random_choice:
- 1
- 2
- a
- b
- c
Loop
现在loop已经替代了with，更多的是loop配合过滤器进行使用
过滤器
常用字符串有关的过滤器
---
- name: test
hosts: node1
vars:
testvar: "abc123ABC 666"
testvar1: " abc "
tasks:
- name: debug1
debug:
#upper将字符串转换成纯大写
msg: "{{testvar | upper}}"
- name: debug2
debug:
#lower将字符串转换成纯小写
msg: "{{testvar | lower}}"
- name: debug3
debug:
#trim将字符串的首尾空格去掉
msg: "{{testvar1 | trim}}"
- name: debug4
debug:
#length求字符串的长度
msg: "{{testvar | length}}"
使用加密算法对字符串进行hash加密
创建一个用户chenyu，并且设置密码为redhat，密码采用SHA512哈希格式
---
- name: create user
hosts: node1
tasks:
- name: create user
user:
name: haha
password: "{{'redhat' | password_hash('sha512')}}"
补充过滤器
[student@ansible ansible]$ cat filterstr.yml
---
- name: 过滤器
hosts: servera
vars:
testvar: "abc123ABC 666"
testvar1: " abc "
testvar2: "123456789"
testvar3: "1a2b,@#$%^&"
tasks:
- name: 将字符串转换成纯大写
debug:
#将字符串转换成纯大写
msg: "{{ testvar | upper }}"
- name: 将字符串转换成纯小写
debug:
#将字符串转换成纯小写
msg: "{{ testvar | lower }}"
- name: 将字符串首字母大写，之后的所有字母纯小写
debug:
#将字符串首字母大写，之后的所有字母纯小写
msg: "{{ testvar | capitalize }}"
- name: 返回字符串的第一个字符
debug:
#返回字符串的第一个字符
msg: "{{ testvar | first }}"
- name: 返回字符串的最后一个字符
debug:
#返回字符串的最后一个字符
msg: "{{ testvar | last }}"
- name: 将字符串开头和结尾的空格去除
debug:
#将字符串开头和结尾的空格去除
msg: "{{ testvar1 | trim }}"
- name: 将字符串放在中间，并且设置字符串的长度为30，字符串两边用空格补齐30位长
debug:
#将字符串放在中间，并且设置字符串的长度为30，字符串两边用空格补齐30位长
msg: "{{ testvar1 | center(width=30) }}"
- name: 返回字符串长度,length与count等效,可以写为count
debug:
#返回字符串长度,length与count等效,可以写为count
msg: "{{ testvar2 | length }}"
- name: 将字符串转换成列表，每个字符作为一个元素
debug:
#将字符串转换成列表，每个字符作为一个元素
msg: "{{ testvar3 | list }}"
- name: 将字符串转换成列表，每个字符作为一个元素，并且随机打乱顺序
debug:
#将字符串转换成列表，每个字符作为一个元素，并且随机打乱顺序
#shuffle的字面意思为洗牌
msg: "{{ testvar3 | shuffle }}"
[root@foundation0 ansible]# ansible-playbook filterstr.yml
和数字操作有关的过滤器
[student@ansible ansible]$cat filterdata.yml
---
- name: this playbook
hosts: servera
vars:
testvar4: -1
tasks:
- name: 将对应的值转换成int类型
debug:
#将对应的值转换成int类型
#ansible中，字符串和整形不能直接计算，比如{{ 8+'8' }}会报错
#所以，我们可以把一个值为数字的字符串转换成整形后再做计算
msg: "{{ 8+('8' | int) }}"
- name: 将对应的值转换成int类型,如果无法转换,默认返回0
debug:
#将对应的值转换成int类型,如果无法转换,默认返回0
#使用int(default=6)或者int(6)时，如果无法转换则返回指定值6
msg: "{{ 'a' | int(default=6) }}"
- name: 将对应的值转换成浮点型，如果无法转换，默认返回'0.0'
debug:
#将对应的值转换成浮点型，如果无法转换，默认返回'0.0'
msg: "{{ '8' | float }}"
- name: 当对应的值无法被转换成浮点型时，则返回指定值’8.88‘
debug:
#当对应的值无法被转换成浮点型时，则返回指定值’8.88‘
msg: "{{ 'a' | float(8.88) }}"
- name: 获取对应数值的绝对值
debug:
#获取对应数值的绝对值
msg: "{{ testvar4 | abs }}"
- name: 四舍五入
debug:
#四舍五入
msg: "{{ 12.5 | round }}"
- name: 取小数点后五位
debug:
#取小数点后五位
msg: "{{ 3.1415926 | round(5) }}"
- name: 从0到100中随机返回一个随机数
debug:
#从0到100中随机返回一个随机数
msg: "{{ 100 | random }}"
- name: 从5到10中随机返回一个随机数
debug:
#从5到10中随机返回一个随机数
msg: "{{ 10 | random(start=5) }}"
- name: 从5到15中随机返回一个随机数,步长为3
debug:
#从5到15中随机返回一个随机数,步长为3
#步长为3的意思是返回的随机数只有可能是5、8、11、14中的一个
msg: "{{ 15 | random(start=5,step=3) }}"
- name: 从0到15中随机返回一个随机数,这个随机数是5的倍数
debug:
#从0到15中随机返回一个随机数,这个随机数是5的倍数
msg: "{{ 15 | random(step=5) }}"
文件或目录类过滤器
[student@ansible ansible]$ cat filterfile.yml
---
- name: 文件或目录类的过滤器
hosts: servera
tasks:
- name: 使用sha1算法对字符串进行哈希
debug:
msg: "{{ '123456' | hash('sha1') }}"
- name: 使用md5算法对字符串进行哈希
debug:
msg: "{{ '123456' | hash('md5') }}"
- name: 获取到字符串的校验和,与md5哈希值一致
debug:
msg: "{{ '123456' | checksum }}"
- name: 使用sha256算法对字符串进行哈希,哈希过程中会生成随机"盐",以便无法直接对比出原值
debug:
msg: "{{ '123456' | password_hash('sha256') }}"
- name: 使用sha256算法对字符串进行哈希,并使用指定的字符串作为"盐"
debug:
msg: "{{ '123456' | password_hash('sha256','mysalt') }}"
- name: 使用sha512算法对字符串进行哈希,哈希过程中会生成随机"盐",以便无法直接对比出原值
debug:
msg: "{{ '123123' | password_hash('sha512') }}"
- name: 使用sha512算法对字符串进行哈希,并使用指定的字符串作为"盐"
debug:
msg: "{{ '123123' | password_hash('sha512','ebzL.U5cjaHe55KK') }}"
ansible判断
When
判断运算符
“==” “!=” “>” “<” “>=” “<=” “and” “or” “not” is in
每次执行完一个任务，不管成功与失败，都会将执行的结果进行注册，可以使用这个注册的变量来when
判断变量的一些tests
Defined：判断变量是否已经定义，已定义则返真
Undefined：判断变量是否已经定义，未定义则返真
None：判断变量值是否为空，如果变量已经定义，但是变量值为空，则返真
[student@ansible ansible]$ vim test.yml
---
- name: test
hosts: node1
vars:
aa: 11
cc:
tasks:
- name: create debug1
debug:
msg: a
when: aa is defined
- name: create debug2
debug:
msg: ab
when: bb is undefined
- name: create debug3
debug:
msg: abc
when: cc is none
判断执行结果的一些tests
Success/successed:通过任务的返回信息判断执行状态，任务执行成功返回真
Failure/failed：通过执行任务的返回信息判断执行状态，任务执行失败则返回真
Change/changed：通过任务的返回信息判断执行状态，任务返回状态为changed则返回真
Skip/skipped：通过任务的返回信息判断执行状态，当任务没有满足条件，而被跳过执行，则返回真。
[student@ansible ansible]$ vim test2.yml
---
- name: test
hosts: node1
vars:
aa: 11
tasks:
- name: shell
shell:
cmd: ls /mnt
when: aa == 11
register: dd
- name: create debug1 success
debug:
msg: haha success
when: dd is success
- name: create debug2 fail
debug:
msg: haha failed
when: dd is failed
- name: create debug3 change
debug:
msg: haha changed
when: dd is changed
- name: create debug4 skip
debug:
msg: haha skip
when: dd is skip
判断路径的一些tests
注意：以下tests的判断均对ansible主机中的路径，与目标主机无关。
file：判断路径是否是一个文件
directory：判断路径是否是一个目录
link：判断路径是否是一个软连接
mount：判断路径是否是一个挂载点
exists：判断路径是否存在
[student@ansible ansible]$ vim test.yml
---
- name: test
hosts: node1
vars:
a1: /test/file1
a2: /test/
a3: /test/softlinka
a4: /test/hardlinka
a5: /boot/
tasks:
- name: debug1
debug:
msg: this is file
when: a1 is file
- name: debug2
debug:
msg: "this is directory"
when: a2 is directory
- name: debug3
debug:
msg: "this is softlink"
when: a3 is link
- name: debug4
debug:
msg: "this is hardlink"
when: a4 is link
- name: debug5
debug:
msg: "this is mount directory"
when: a5 is mount
- name: debug6
debug:
msg: "this is exists"
when: a1 is exists
判断字符串的一些tests
lower：判断包含字母的字符串中的字母是否纯小写
upper：判断包含字母的字符串中的字母是否纯大写
[student@ansible ansible]$ vim test.yml
---
- name: test
hosts: node1
vars:
a1: abc
a2: ABC
a3: a1b
tasks:
- name: debug1
debug:
msg: this string is all lower
when: a1 is lower
- name: debug2
debug:
msg: this string is all upper
when: a2 is upper
- name: debug3
debug:
msg: chenyu
when: a3 is lower
其他的一些tests
string：判断对象是否是一个字符串
number：判断对象是否一个数字
[student@ansible ansible]$ vim test.yml
---
- name: test
hosts: node1
vars:
a1: 1
a2: "1"
a3: a
tasks:
- name: debug1
debug:
msg: this is number
when: a1 is number
- name: debug2
debug:
msg: this is string
when: a2 is string
- name: debug3
debug:
msg: this is string
when: a3 is string
block/rescue/always: 限制性block，如果执行失败，则执行rescue，无论是block还是rescue执行失败还是成功，在最后都执行always
例题：
创建一个名为/etc/ansible/lv.yml 的playbook，它将在所有受管节点上运行以执行下列任务：
创建符合以下要求的逻辑卷：
逻辑卷创建在research卷组中
逻辑卷名称为data
逻辑卷大小为1500MiB
使用ext4文件系统格式化逻辑卷
如果无法创建请求的逻辑卷大小，应显示错误消息
Could not create logical volume of that size，并且应改为使用大小 800MiB。
如果卷组research 不存在 ，应显示错误消息
Volume group does not exist。
不要以任何方式挂载逻辑卷
前提：在node1、node2上添加一块硬盘，然后新建卷组
Node1的卷组大小为2G 卷组名为research
Node2的卷组大小为1G 卷组名为research
[student@ansible ansible]$ vim vg.yml
---
- name: create vg for node1
hosts: node1
tasks:
- name: create partition
parted:
device: /dev/sdb
number: 1
part_type: primary
part_start: 10MiB
part_end: 2058MiB
state: present
- name: create vg research
lvg:
vg: research
pvs: /dev/sdb1
- name: create vg for node2
hosts: node2
tasks:
- name: create partition for node2
parted:
device: /dev/sdb
number: 1
part_type: primary
part_start: 10MiB
part_end: 1034MiB
state: present
- name: create vg research for node2
lvg:
vg: research
pvs: /dev/sdb1
执行vg.yml来给node1、node2创建vg
Ansible-playbook vg.yml
新建lv.yml，满足题目需求
[student@ansible ansible]$ vim lv.yml
---
- name: create lvm
hosts: node1,node2
tasks:
- name: create lv
block:
- name: create lvm 1500M
lvol:
vg: research
lv: data
size: 1500M
rescue:
- name: output fail message
debug:
msg: Could not create logical volume of that size
- name: create lvm 800M
lvol:
vg: research
lv: data
size: 800M
always:
- name: format lvm
filesystem:
fstype: ext4
dev: /dev/research/data
when: "'research' in ansible_facts.lvm.vgs"
#也可以用when: "'research' in ansible_lvm.vgs"
- name: serach not exists
debug:
msg: Volume group does not exist
when: "'research' not in ansible_facts.lvm.vgs"
#也可以用when: "'research' not in ansible_lvm.vgs"
判断与错误处理
fail模块可用于中断剧本，但我们一般是不会无故中断，除非在满足条件的情况下可以中断，经常和when一起用
[student@ansible ansible]$ vim c.yml
---
- name: test
hosts: node1
tasks:
- name: shell
shell:
cmd: echo 'this is a string for testing--error'
register: return_value
- name: fail
fail:
msg: Conditions established,Interrupt running playbook
when: "'error' in return_value.stdout"
- name: debug
debug:
msg: I never execute,because the playbook has stopped
或者使用failed_when
---
- name: test
hosts: node1
tasks:
- name: debug
debug:
msg: I execute normally
- name: shell
shell:
cmd: echo 'this is a string testing--error'
register: return_value
failed_when: "'error' in return_value.stdout"
- name: debug2
debug:
msg: haha
ignore_errors: yes 跳过错误、忽略错误
例子
---
- name: test
hosts: node1
tasks:
- name: debug1
debug:
msg: "{{ansible_fqdn}}"
- name: debug2
debug:
msg: "{{ansible_ip}}"
ignore_errors: yes
- name: create file
file:
path: /tmp/abc
state: touch
Changed_when：可以修改任务执行后的最终状态
---
- name: test
hosts: node1
tasks:
- name: debug1
debug:
msg: "{{ansible_fqdn}}"
changed_when: true
或者可以让任务执行状态显示失败
---
- name: test
hosts: node1
tasks:
- name: shell
shell:
cmd: ls /tmp
changed_when: false
拓展
给node1主机添加一块5G的硬盘
给node2主机添加一块2G的硬盘
node3主机不添加硬盘
写一个partition.yml的playbook，满足如下要求：
给所有的受控主机创建分区，创建2500M的分区，然后对分区进行格式化，格式化为xfs的文件系统。
如果不能满足创建2500M的分区，则显示错误消息为：disk size not enough，接着改为创建1000M。
如果没有/dev/sdb硬盘，则显示错误消息为：the device is not sdb
[student@ansible ansible]$ vim partition.yml
[student@ansible ansible]$ cat partition.yml
---
- name: create partition
hosts: all
tasks:
- name: create partition 2500
block:
- name: create
parted:
device: /dev/sdb
number: 1
part_type: primary
part_start: 10MiB
part_end: 2520MiB
state: present
rescue:
- name: output fail message
debug:
msg: disk size not enough
- name: create partition 1000
parted:
device: /dev/sdb
number: 1
part_type: primary
part_start: 10MiB
part_end: 1020MiB
state: present
always:
- name: filesystem
filesystem:
dev: /dev/sdb1
fstype: xfs
when: "ansible_devices.sdb is defined"
- name: not in sdb
debug:
msg: "the device is not sdb"
when: "ansible_devices.sdb is not defined"
[student@ansible ansible]$ ansible-playbook partition.yml
PLAY [create partition] *************************************************************************************
TASK [Gathering Facts] **************************************************************************************
ok: [node1]
ok: [node3]
ok: [node2]
TASK [create] ***********************************************************************************************
skipping: [node3]
fatal: [node2]: FAILED! => {"changed": false, "err": "Error: The location 2520MiB is outside of the device /dev/sdb.\n", "msg": "Error while running parted script: /sbin/parted -s -m -a optimal /dev/sdb -- unit KiB mklabel msdos mkpart primary 10MiB 2520MiB", "out": "", "rc": 1}
changed: [node1]
TASK [output fail message] **********************************************************************************
ok: [node2] => {
"msg": "disk size not enough"
}
TASK [create partition 1000] ********************************************************************************
changed: [node2]
TASK [filesystem] *******************************************************************************************
skipping: [node3]
changed: [node2]
changed: [node1]
TASK [not in sdb] *******************************************************************************************
skipping: [node1]
skipping: [node2]
ok: [node3] => {
"msg": "the device is not sdb"
}
PLAY RECAP **************************************************************************************************
node1 : ok=3 changed=2 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0
node2 : ok=4 changed=2 unreachable=0 failed=0 skipped=1 rescued=1 ignored=0
node3 : ok=2 changed=0 unreachable=0 failed=0 skipped=2 rescued=0 ignored=0
[student@ansible ansible]$ for i in node{1..3}; do ssh root@$i 'lsblk | grep sdb'; done
sdb 8:16 0 5G 0 disk
└─sdb1 8:17 0 2.5G 0 part
sdb 8:16 0 2G 0 disk
└─sdb1 8:17 0 1010M 0 part
扫一扫
3874
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
