Ansible的debug模块介绍，fact变量采集和缓存相关操作演示
目录
一.debug模块的使用方法
1.帮助文档给出的示例
2.主要用到的参数
（1）msg：主要用这个参数来指定要输出的信息
（2）var：打印指定的变量，一般是通过register注册了的变量
（3）verbosity：调试级别，默认是0表示全部显示
3.输出信息详解
4.在debug中使用when做条件判断
（1）通过rc的结果做判断
（2）通过是否failed做判断
二.fact变量
1.setup简单用法演示
（1）通过filter进行过滤
（2）--tree将信息输出到指定目录
2.手动设置fact
3.使用set_fact模块定义变量
4.手动采集变量
三.fact缓存
1.json方式
2.redis方式
3.memcached方式基本已弃用
一.debug模块的使用方法
ansible上playbook的debug是一个常用的调试模块，主要用于在playbook执行（调试、引用变量）过程输出一些关键信息，并且可以对这些关键信息进行一定的格式化输出和条件判断
1.帮助文档给出的示例
EXAMPLES:
# Example that prints the loopback address and gateway for each host
- debug:
msg: System {{ inventory_hostname }} has uuid {{ ansible_product_uuid }}
- debug:
msg: System {{ inventory_hostname }} has gateway {{ ansible_default_ipv4.gateway }}
when: ansible_default_ipv4.gateway is defined
# Example that prints return information from the previous task
- shell: /usr/bin/uptime
register: result
- debug:
var: result
verbosity: 2
- name: Display all variables/facts known for a host
debug:
var: hostvars[inventory_hostname]
verbosity: 4
# Example that prints two lines of messages, but only if there is an environment value set
- debug:
msg:
- "Provisioning based on YOUR_KEY which is: {{ lookup('env', 'YOUR_KEY') }}"
- "These servers were built using the password of '{{ password_used }}'. Please retain this for later use."2.主要用到的参数
（1）msg：主要用这个参数来指定要输出的信息
演示效果
[root@main ~]# cat iduser.yaml
---
- name: is su exist
hosts: webservers
tasks:
- name: test su
shell: id su
register: su
ignore_errors: yes
- name: echo it
debug:
msg: "用户存在"
[root@main ~]# ansible-playbook iduser.yaml
PLAY [is su exist] ******************************************************************************************************************************
TASK [Gathering Facts] **************************************************************************************************************************
ok: [servera]
TASK [test su] **********************************************************************************************************************************
changed: [servera]
TASK [echo it] **********************************************************************************************************************************
ok: [servera] => {
"msg": "用户存在"
}
PLAY RECAP **************************************************************************************************************************************
servera : ok=3 changed=1 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 （2）var：打印指定的变量，一般是通过register注册了的变量
演示效果
[root@main ~]# cat iduser.yaml
---
- name: is su exist
hosts: servera
tasks:
- name: test su
shell: id su
register: su
ignore_errors: yes
- name: echo it
debug:
var: su #打印前面已经注册了的“su”变量
[root@main ~]# ansible-playbook iduser.yaml
PLAY [is su exist] ******************************************************************************************************************************
TASK [Gathering Facts] **************************************************************************************************************************
ok: [servera]
TASK [test su] **********************************************************************************************************************************
changed: [servera]
TASK [echo it] **********************************************************************************************************************************
ok: [servera] => {
"su": {
"changed": true,
"cmd": "id su",
"delta": "0:00:00.002850",
"end": "2023-10-19 14:12:43.406662",
"failed": false,
"rc": 0,
"start": "2023-10-19 14:12:43.403812",
"stderr": "",
"stderr_lines": [],
"stdout": "uid=1000(su) gid=1000(su) groups=1000(su)",
"stdout_lines": [
"uid=1000(su) gid=1000(su) groups=1000(su)"
]
}
}
PLAY RECAP **************************************************************************************************************************************
servera : ok=3 changed=1 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 （3）verbosity：调试级别，默认是0表示全部显示
3.输出信息详解
以下面代码段为例：
"su"：表示变量，在输出信息中它是一个字典类型
changed：根据此值来判断是否发生了状态改变
cmd：过程中调用的命令
failed：是否运行失败
rc：返回值，为0成功，非0失败或异常
stderr：出现异常时会在这显示错误信息
stderr_lines：以行分割的格式输出错误信息
stdout：运行成功会在此处输出返回结果
stdout_lines：以行分割的格式输出结果
ok: [servera] => {
"su": {
"changed": true,
"cmd": "id su",
"delta": "0:00:00.002850",
"end": "2023-10-19 14:12:43.406662",
"failed": false,
"rc": 0,
"start": "2023-10-19 14:12:43.403812",
"stderr": "",
"stderr_lines": [],
"stdout": "uid=1000(su) gid=1000(su) groups=1000(su)",
"stdout_lines": [
"uid=1000(su) gid=1000(su) groups=1000(su)"
]
}
}4.在debug中使用when做条件判断
（1）通过rc的结果做判断
[root@main ~]# cat iduser.yaml
---
- name: is su exist
hosts: servera
tasks:
- name: test su
shell: id su
register: su
ignore_errors: yes
- name: echo it
debug:
msg: "用户存在"
when: su.rc==0 #当返回值为0时才输出msg（2）通过是否failed做判断
[root@main ~]# cat iduser.yaml
---
- name: is su exist
hosts: servera
tasks:
- name: test su
shell: id su
register: su
ignore_errors: yes
- name: echo it
debug:
msg: "用户存在"
#when: su.rc==0
when: su is not failed #当su变量的结果不失败时才输出
[root@main ~]# ansible-playbook iduser.yaml
PLAY [is su exist] ******************************************************************************************************************************
TASK [Gathering Facts] **************************************************************************************************************************
ok: [servera]
TASK [test su] **********************************************************************************************************************************
changed: [servera]
TASK [echo it] **********************************************************************************************************************************
ok: [servera] => {
"msg": "用户存在"
}
PLAY RECAP **************************************************************************************************************************************
servera : ok=3 changed=1 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
二.fact变量
setup用于获取受管节点的详细信息（硬盘、IP、cpu等信息），可以将信息作为变量在playbook中引用，setup依赖fact进行获取信息
EXAMPLES:
# Display facts from all hosts and store them indexed by I(hostname) at C(/tmp/facts).
# ansible all -m setup --tree /tmp/facts
# Display only facts regarding memory found by ansible on all hosts and output them.
# ansible all -m setup -a 'filter=ansible_*_mb'
# Display only facts returned by facter.
# ansible all -m setup -a 'filter=facter_*'
# Collect only facts returned by facter.
# ansible all -m setup -a 'gather_subset=!all,!any,facter'
- name: Collect only facts returned by facter
setup:
gather_subset:
- '!all'
- '!any'
- facter1.setup简单用法演示
（1）命令行通过filter进行过滤
[root@main ~]# ansible servera -m setup -a 'filter=ansible_*_ipv4'
servera | SUCCESS => {
"ansible_facts": {
"ansible_default_ipv4": {
"address": "192.168.2.131",
"alias": "ens33",
"broadcast": "192.168.2.255",
"gateway": "192.168.2.1",
"interface": "ens33",
"macaddress": "00:0c:29:bc:03:89",
"mtu": 1500,
"netmask": "255.255.255.0",
"network": "192.168.2.0",
"type": "ether"
},
"discovered_interpreter_python": "/usr/bin/python"
},
"changed": false
}（2）--tree将信息输出到指定目录
[root@main ~]# ansible servera -m setup -a 'filter=ansible_*_mb' --tree /root/facts
[root@main ~]# cat facts/servera
{"ansible_facts": {"ansible_memfree_mb": 5327, "ansible_memory_mb": {"nocache": {"free": 5510, "used": 338}, "real": {"free": 5327, "total": 5848, "used": 521}, "swap": {"cached": 0, "free": 2047, "total": 2047, "used": 0}}, "ansible_memtotal_mb": 5848, "ansible_swapfree_mb": 2047, "ansible_swaptotal_mb": 2047, "discovered_interpreter_python": "/usr/bin/python"}, "changed": false}2.手动设置fact
可以为某写主机手动定制fact，称其为本地fact，将管理节点定义好的fact文件传输给需要定制fact的节点，定制的fact默认存放在受管节点的/etc/ansible/facts.d目录下，
示例：为server节点自定义一个fact，使用这个fact启动servera上的httpd服务，此示例主要用到三个文件（cus.fact、afact.yaml、useafact.yaml），都放在同一目录下
[root@main ~]# cat cus.fact #在管理节点定义好fact文件
[su]
mypkg=httpd
myser=httpd
state=started
[root@main ~]# cat afact.yaml
#定义yaml文件，在受管节点创建/etc/ansible/facts.d目录，将fact文件拷贝至这个目录
---
- hosts: servera
vars:
remote_dir: /etc/ansible/facts.d
facts_file: cus.fact
tasks:
- name: create remote_dir in servera
file:
state: directory
recurse: yes
path: "{{ remote_dir }}"
- name: copy local cus.fact
copy:
src: "{{ facts_file }}"
dest: "{{ remote_dir }}"
[root@main ~]# ansible servera -m setup -a 'filter="ansible_local"'
#成功在servera上过滤出本地fact
servera | SUCCESS => {
"ansible_facts": {
"ansible_local": {
"cus": {
"su": {
"mypkg": "httpd",
"myser": "httpd",
"state": "started"
}
}
},
"discovered_interpreter_python": "/usr/bin/python"
},
"changed": false
}
[root@main ~]# cat useafact.yaml #为servera定义yaml文件引用其下的fact进行启动httpd
---
- hosts: servera
tasks:
- name: using servera local fact to start httpd
service:
name: "{{ ansible_facts.ansible_local.cus.su.myser }}"
#引用方式较长（ansible的facts.本地的.fact文件.fact内的字段名）
state: "{{ ansible_facts.ansible_local.cus.su.state }}"
[root@main ~]# ansible-playbook useafact.yaml
PLAY [servera] **********************************************************************************************************************************
TASK [Gathering Facts] **************************************************************************************************************************
ok: [servera]
TASK [using servera local fact to start httpd] **************************************************************************************************
changed: [servera]
PLAY RECAP **************************************************************************************************************************************
servera : ok=2 changed=1 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
[root@main ~]# ansible servera -m shell -a 'systemctl status httpd | grep Active'
#启动成功
servera | CHANGED | rc=0 >>
Active: active (running) since Thu 2023-10-19 15:15:09 CST; 5s ago3.使用set_fact模块定义变量
set_fact用于自定义facts，从而通过template或作为变量在playbook中被引用，set_fact定义的变量只能在此playbook中使用有效
EXAMPLES:
# Example setting host facts using key=value pairs, note that this always creates strings or booleans
- set_fact: one_fact="something" other_fact="{{ local_var }}"
# Example setting host facts using complex arguments
- set_fact:
one_fact: something
other_fact: "{{ local_var * 2 }}"
another_fact: "{{ some_registered_var.results | map(attribute='ansible_facts.some_fact') | list }}"
# Example setting facts so that they will be persisted in the fact cache
- set_fact:
one_fact: something
other_fact: "{{ local_var * 2 }}"
cacheable: yes
# As of Ansible 1.8, Ansible will convert boolean strings ('true', 'false', 'yes', 'no')
# to proper boolean values when using the key=value syntax, however it is still
# recommended that booleans be set using the complex argument style:
- set_fact:
one_fact: yes
other_fact: no示例：通过set_fact计算进程使用内存的情况，这个计算结果也可以在playbook中引用
[root@main ~]# ansible servera -m setup -a 'filter="ansible_memtotal_mb"'
servera | SUCCESS => {
"ansible_facts": {
"ansible_memtotal_mb": 5848, #先过滤一下看参数是否存在
"discovered_interpreter_python": "/usr/bin/python"
},
"changed": false
}
[root@main ~]# cat initfree.yaml
---
- hosts: servera
tasks:
- name: cal pool size
set_fact: #定义一个变量为这个计算结果
pool_size: "{{ ansible_memtotal_mb / 2 | int}}"
- debug:
var: pool_size #输出这个变量
[root@main ~]# ansible-playbook initfree.yaml
PLAY [servera] **********************************************************************************************************************************
TASK [Gathering Facts] **************************************************************************************************************************
ok: [servera]
TASK [cal pool size] ****************************************************************************************************************************
ok: [servera]
TASK [debug] ************************************************************************************************************************************
ok: [servera] => {
"pool_size": "2924.0"
}
PLAY RECAP **************************************************************************************************************************************
servera : ok=3 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 4.手动采集变量
运行playbook时，ansible会先ssh到受管节点去采集fact，如果收集信息过多过大会影响执行速度和效率，可以选择关闭采集或先关闭采集完成任务再重新采集
#显式采集行为
TASK [Gathering Facts] **************************************************************************************************************************
ok: [servera]示例：上述情况下，我们可以使用gather_facts显式关闭fact采集，然后执行完任务再重新采集facts
[root@main ~]# cat nogather.yaml
---
- hosts: servera
gather_facts: false
tasks:
- name: debug this
debug: msg="hello"
- name: wait for 10
wait_for:
timeout: 6
- name: regather facts
setup:
gather_subset: all #参考EXAMPLES
[root@main ~]# ansible-playbook nogather.yaml
PLAY [servera] **********************************************************************************************************************************
TASK [debug this] *******************************************************************************************************************************
ok: [servera] => {
"msg": "hello"
}
TASK [wait for 10] ******************************************************************************************************************************
ok: [servera]
TASK [regather facts] ***************************************************************************************************************************
ok: [servera]
PLAY RECAP **************************************************************************************************************************************
servera : ok=3 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 三.fact缓存
在playbook中引入fact时可以设置fact缓存，目前以是json、redis、memcached三种方式，以下是需要修改的ansible.cfg的defaults部分参数
- gathering
是否开启fact支持三个选项：smart（默认采集facts，facts已存在时不采集，即缓存），implicit（默认采集facts，可以使用"gather_facts: false"来禁止采集），explicit（默认不采集，可以使用"gather_facts: true"来假期采集）
- fact_caching_timeout
缓存时间，秒为单位
- fact_caching
缓存方式，jsonfile，redis，memcached
- fact_caching_connection
指定fact缓存的json文件位置，若没有会自动创建
1.json方式
此方式下，ansible会将fact写入控制主机的文件中
#在ansible.cfg配置文件的defaults模块加这些参数
gathering=smart
fact_caching_timeout=86400
fact_caching=jsonfile
fact_caching_connection=/root/ansible_fact_cache示例：
执行一个会采集fact的playbook，查看是否将fact缓存到指定的位置
[root@main ~]# cat ansible.cfg
[defaults]
......
gathering=smart
fact_caching_timeout=86400
fact_caching=jsonfile
fact_caching_connection=/root/ansible_fact_cache
......
[root@main ~]# ansible-playbook myhttpd.yaml
PLAY [stop servera httpd] ***********************************************************************************************************************
TASK [Gathering Facts] **************************************************************************************************************************
ok: [servera]
TASK [stop it] **********************************************************************************************************************************
ok: [servera]
PLAY [install serverb mod_ssl] ******************************************************************************************************************
TASK [Gathering Facts] **************************************************************************************************************************
ok: [serverb]
TASK [install it] *******************************************************************************************************************************
ok: [serverb]
PLAY RECAP **************************************************************************************************************************************
servera : ok=2 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
serverb : ok=2 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
[root@main ~]# ll /root/ansible_fact_cache/ #查看
total 56
-rw-r--r-- 1 root root 25052 Oct 19 18:00 servera
-rw-r--r-- 1 root root 25071 Oct 19 18:00 serverb2.redis方式
需要安装redis服务且保持运行，还需要安装python支持redis的相关包，更改ansible.cfg
fact_caching=redis #指定redis方式
fact_caching_connection=127.0.0.1:6379:0 #指定redis服务设备的IP和端口，使用0号数据库
[root@main ~]# yum install -y redis #安装redis
[root@main ~]# systemctl start redis
[root@main ~]# ps -ef | grep redis
redis 4278 1 0 17:50 ? 00:00:00 /usr/bin/redis-server 127.0.0.1:6379
root 4286 1485 0 17:51 pts/0 00:00:00 grep --color=auto redis
[root@main ~]# systemctl enable redis
Created symlink from /etc/systemd/system/multi-user.target.wants/redis.service to /usr/lib/systemd/system/redis.service.
[root@main ~]# python --version
Python 2.7.5
[root@main ~]# yum list | grep python2-redis
python2-redis.noarch 2.10.6-2.el7 epel
[root@main ~]# yum install -y python2-redis.noarch #安装对应python版本对应的redis支持包
[root@main ~]# ansible-playbook myhttpd.yaml #再执行一次
[root@main ~]# redis-cli #进入查看
127.0.0.1:6379> keys *
1) "ansible_factsservera"
2) "ansible_factsserverb"
3) "ansible_cache_keys"
127.0.0.1:6379> type "ansible_cache_keys"
zset
127.0.0.1:6379> type "ansible_factsservera" #前两个就是采集的facts
string
127.0.0.1:6379> 3.memcached方式基本已弃用