自动化运维工具之Puppet master/agent模型、站点清单和puppet多环境设定

今天我来了解下puppet的master/agent模型以及站点清单的相关话题；

在说puppet的master/agent模型前，我们要先来回顾下master/agent的工作过程

如上图所示，puppet的master/agent模型的工作过程；首先agent向master发送自己的主机名和facts信息，然后对应的master收到信息后，会根据对应的主机名来查找和agent相关的配置，然后把对应的配置（主要是类和模块中的资源清单内容）编译好，master把编译好的catalog发送给agent；agent收到master发送到catalog后，首先要在本机查找相关资源的状态，如果对应资源的状态和编译好的catalog中的状态一样，则不改变其状态或者跳过；如果对应的资源状态和catalog中的状态不一样，此时agent就会应用catalog，把对应资源强制更改为和catalog中的状态一样；最后agent把应用报告发送给master；此时agent就变成了我们在资源清单中定义的系统状态；这里需要注意一点catalog是一个二进制文件，不能直接查看，这个二进制文件是通过master上定义的模块，或类等资源清单中的内容编译而成；

master和agent的通信

在puppet的master/agent模型中，master和agent通信是以https协议通信；使用https通信就意味着要有证书验证，有证书就会有ca；在puppet的master/agent模型中，它内置了ca，意思就是我们不需要再手动搭建ca；对应master的证书，私钥文件以及ca的证书私钥文件，puppet master都会自动生成；对于agent的私钥和证书签署文件也会由puppet agent自动生成；并且在第一次启动agent时，它默认会把生成的证书签发文件发送给master，等待master签发证书；在master上签发证书，这一步需要人工手动干预；证书签发好以后，master和agent才可以正常通信；默认情况下agent是每30分钟到master上拉取一次相关配置，这也就意味着我们在master端上的资源清单内容最多不超过30分钟就能在对应的agent上得到应用；

站点清单

在上述master/agent模型的工作过程中，agent向master发送主机名和facts，然后对应master把对应主机名的配置进行编译然后发送给对应agent；这里master根据什么来查找对应主机名的对应配置呢？通常情况下如果我们在master端上不定义站点清单，对应的agent就无法获取到对应的配置；也就说站点清单是用来定义哪些agent上，应该跑哪些资源或模块；这个逻辑和我们在使用ansible一样，定义好角色以后，还需要额外的playbook来定义那些主机上应该应用那些角色；对于puppet也是一样，我们定义好模块以后，至于哪些主机应用哪些模块这个是需要我们定义站点清单；

puppet的master/agent模型的搭建

部署master端

配置主机名解析

1
2
3
4
5
6
7
[root@master ~]# cat /etc/hosts
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
192.168.16.151 master.test.org master
192.168.16.152 node01.test.org node01
192.168.16.153 node02.test.org node02
[root@master ~]#
　　提示：除了各服务器之间的主机名解析，我们还需要注意确保各服务器时间同步，关闭selinux，确保iptables是关闭状态；

安装puppet-server包

1
[root@master ~]# yum install -y puppet-server
　　提示：在master端安装puppet-server这个包默认会把puppet当作依赖包一并安装；

启动服务

1
2
3
4
5
6
7
8
9
[root@master ~]# systemctl start puppetmaster.service
[root@master ~]# ss -tnl
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 :22 :
LISTEN 0 100 127.0.0.1:25 :
LISTEN 0 128 :8140 :
LISTEN 0 128 [::]:22 [::]:
LISTEN 0 100 [::1]:25 [::]:
[root@master ~]#
　　提示：puppet master默认会监听在tcp的8140端口，用于接收agent来master获取配置；所以请确保master端的8140端口能够正常处于监听状态；

部署agent端

安装puppet

1
[root@node01 ~]# yum install -y puppet
　　编辑配置文件，配置puppet server的地址

提示：编辑/etc/puppet/puppet.conf文件，将server=puppetmaster主机的主机名配置上，如上；如果使用主机名的形式，请确保对应agent能够正常解析；

启动服务

1
2
3
4
5
6
7
8
9
10
11
12
[root@node01 ~]# systemctl start puppet
[root@node01 ~]# ss -tnl
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 :22 :
LISTEN 0 100 127.0.0.1:25 :
LISTEN 0 128 [::]:22 [::]:
LISTEN 0 100 [::1]:25 [::]😗
[root@node01 ~]# ps aux |grep puppet
root 1653 0.0 0.0 115404 1436 ? Ss 15:27 0:00 /bin/sh /usr/bin/start-puppet-agent agent --no-daemonize
root 1654 26.7 2.2 317640 41888 ? Sl 15:27 0:02 /usr/bin/ruby /usr/bin/puppet agent --no-daemonize
root 1680 0.0 0.0 112808 968 pts/0 S+ 15:28 0:00 grep --color=auto puppet
[root@node01 ~]#
　　提示：puppet agent虽然也是一个守护进程工作，但它主要不是对外提供服务，它之所以要工作为守护进程，是因为它要周期性的到master端拉取和自己相关的配置；所以它不会监听任何端口；

在master端查看未签发证书列表

1
2
3
4
[root@master ~]# puppet cert list
“node01.test.org” (SHA256) AD:01:59:E7:6C:97:E7:5E:67:09:B9:52:94:0D:37:89:82:8B:EE:49:BB:4D:FC:E1:51:64:BE:EF:71:47:15:11
“node02.test.org” (SHA256) E1:EC:5B:0C:BF:B7:4C:B9:4F:10:A9:12:34:8B:7A:36:E3:A1:D4:EC:DD:DD:DC:F4:05:48:0B:85:B5:70:AC:28
[root@master ~]#
　　提示：puppet默认会自己维护一个ca，我们只需要用对应的子命令进行管理即可；以上命令表示列出未签发的证书列表（即收到的agent证书签发文件列表）；从上面的信息可以看到现在node01和node02都没有签发证书；

在master端签发证书

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
[root@master ~]# puppet cert list
“node01.test.org” (SHA256) AD:01:59:E7:6C:97:E7:5E:67:09:B9:52:94:0D:37:89:82:8B:EE:49:BB:4D:FC:E1:51:64:BE:EF:71:47:15:11
“node02.test.org” (SHA256) E1:EC:5B:0C:BF:B7:4C:B9:4F:10:A9:12:34:8B:7A:36:E3:A1:D4:EC:DD:DD:DC:F4:05:48:0B:85:B5:70:AC:28
[root@master ~]# puppet cert sign node01.test.org
Notice: Signed certificate request for node01.test.org
Notice: Removing file Puppet::SSL::CertificateRequest node01.test.org at ‘/var/lib/puppet/ssl/ca/requests/node01.test.org.pem’
[root@master ~]# puppet cert list
“node02.test.org” (SHA256) E1:EC:5B:0C:BF:B7:4C:B9:4F:10:A9:12:34:8B:7A:36:E3:A1:D4:EC:DD:DD:DC:F4:05:48:0B:85:B5:70:AC:28
[root@master ~]# puppet cert sign --all
Notice: Signed certificate request for node02.test.org
Notice: Removing file Puppet::SSL::CertificateRequest node02.test.org at ‘/var/lib/puppet/ssl/ca/requests/node02.test.org.pem’
[root@master ~]# puppet cert list
[root@master ~]# puppet cert list --all

• “master.test.org” (SHA256) 0C:CC:20:EE:F5:FC:73:21:0B:15:73:EF:A5:0B:3A:F8:01:DB:F7:07:7C:DB:78:87:80:87:FC:F2:BF:E7:2F:30 (alt names: “DNS:master.test.org”, “DNS:puppet”, “DNS:puppet.test.org”)
• “node01.test.org” (SHA256) 34:BE:E1:1E:26:15:56:56:C3:A0:0D:FB:7F:01:B1:80:35:EC:1D:07:26:C7:05:CA:6E:19:8C:75:9A:A4:67:4E
• “node02.test.org” (SHA256) E7:B6:B0:FD:04:61:A8:87:D9:E5:DA:51:8B:1D:E0:AD:11:F0:A2:65:43:6D:C4:8D:54:C8:75:8B:DF:CC:51:93
  [root@master ~]#
  　　提示：签发证书需要用cert sign+对应要签发的证书的主机名；如果不想一个一个指定来签发，也可以使用–all选项来签发所有未签发的证书；到此puppet的master/agent模型就搭建好了；后续我们只需要在master端上定义模块和站点清单即可；

示例：在master上创建redis模块，主要功能是安装启动redis，并让其配置为主从复制模式；

在master端上创建模块目录结构

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
[root@master ~]# mkdir -pv /etc/puppet/modules/redis/{manifests,files,templates,lib,spec,tests}
mkdir: created directory ‘/etc/puppet/modules/redis’
mkdir: created directory ‘/etc/puppet/modules/redis/manifests’
mkdir: created directory ‘/etc/puppet/modules/redis/files’
mkdir: created directory ‘/etc/puppet/modules/redis/templates’
mkdir: created directory ‘/etc/puppet/modules/redis/lib’
mkdir: created directory ‘/etc/puppet/modules/redis/spec’
mkdir: created directory ‘/etc/puppet/modules/redis/tests’
[root@master ~]# tree /etc/puppet/modules/redis/
/etc/puppet/modules/redis/
├── files
├── lib
├── manifests
├── spec
├── templates
└── tests

6 directories, 0 files
[root@master ~]#
　　在/etc/puppet/modules/redis/manifests/目录下创建资源清单

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
[root@master ~]# cat /etc/puppet/modules/redis/manifests/init.pp
class redis{
package{“redis”:
ensure => installed,
}
service{“redis”:
ensure => running,
enable => true,
hasrestart => true,
restart => ‘service redis restart’,
}
}
[root@master ~]# cat /etc/puppet/modules/redis/manifests/master.pp
class redis::master($masterport{=}^{\prime }6379^{\prime },$masterpass=‘admin’) inherits redis {
file{"/etc/redis.conf":
ensure => file,
content => template(‘redis/redis-master.conf.erb’),
owner => ‘redis’,
group => ‘root’,
mode => ‘0644’,
}
Service[“redis”]{
subscribe => File["/etc/redis.conf"],
restart => ‘systemctl restart redis’
}
}
[root@master ~]# cat /etc/puppet/modules/redis/manifests/slave.pp
class redis::slave($masterip,$masterport=‘6379’,$masterpass=‘admin’) inherits redis {
file{"/etc/redis.conf":
ensure => file,
content => template(‘redis/redis-slave.conf.erb’),
owner => ‘redis’,
group => ‘root’,
mode => ‘0644’,
}
Service[“redis”]{
subscribe => File["/etc/redis.conf"],
restart => ‘systemctl restart redis’
}
}
[root@master ~]#
　　在templates目录创建对应的模版文件

redis-master.conf.erb文件内容

View Code
　　redis-slave.conf.erb文件内容

View Code
　　redis模块目录和文件存放情况，以及模版文件中需要传递参数

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
[root@master ~]# tree /etc/puppet/modules/redis/
/etc/puppet/modules/redis/
├── files
├── lib
├── manifests
│ ├── init.pp
│ ├── master.pp
│ └── slave.pp
├── spec
├── templates
│ ├── redis-master.conf.erb
│ └── redis-slave.conf.erb
└── tests

6 directories, 5 files
[root@master ~]# grep -Ei “^port|requirepass|masterauth|slaveof” /etc/puppet/modules/redis/templates/redis-master.conf.erb
port <%= @masterport %>
requirepass <%= @masterpass %>
[root@master ~]# grep -Ei “^port|requirepass|masterauth|slaveof” /etc/puppet/modules/redis/templates/redis-slave.conf.erb
port 6379
slaveof <%= @masterip %> <%= @masterport %>
masterauth <%= @masterpass %>
[root@master ~]#
　　提示：到此redis模块就准备好了；

定义站点清单文件

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[root@master ~]# cat /etc/puppet/manifests/site.pp
node ‘node01.test.org’{
class{“redis::master”:
masterport => “6379”,
masterpass => “admin123.com”
}
}
node ‘node02.test.org’{
class{“redis::slave”:
masterip => ‘192.168.16.152’,
masterport => ‘6379’,
masterpass => ‘admin123.com’
}
}
[root@master ~]#
　　提示：站点清单必须是在/etc/puppet/manifests/目录下，名字必须为site.pp，其内容必须由关键字node来定义对应一个主机或一类主机要应用的资源；站点清单也可以使用类的继承和我们在定义类时的使用差不多；

在node01上获取自己相关的配置

提示：可以看到在node01上手动拉取配置能够获取到对应到配置；

取消–noop 选项跑一遍，看看对应到redis是否启动？配置文件是否上我们指定到就配置？

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
[root@node01 ~]# puppet agent -v --no-daemonize
Notice: Starting Puppet client version 3.6.2
Info: Retrieving pluginfacts
Info: Retrieving plugin
Info: Caching catalog for node01.test.org
Warning: The package type’s allow_virtual parameter will be changing its default value from false to true in a future release. If you do not want to allow virtual packages, please explicitly set allow_virtual to false.
(at /usr/share/ruby/vendor_ruby/puppet/type.rb:816:in `set_default’)
Info: Applying configuration version ‘1607328529’
Info: /Stage[main]/Redis::Master/File[/etc/redis.conf]: Filebucketed /etc/redis.conf to puppet with sum d43fae161baaf47638d0bb571785974d
Notice: /Stage[main]/Redis::Master/File[/etc/redis.conf]/content: content changed ‘{md5}d43fae161baaf47638d0bb571785974d’ to ‘{md5}32a39b519eb7a9f70a4bdf4705a7afc0’
Notice: /Stage[main]/Redis::Master/File[/etc/redis.conf]/mode: mode changed ‘0640’ to ‘0644’
Info: /Stage[main]/Redis::Master/File[/etc/redis.conf]: Scheduling refresh of Service[redis]
Info: /Stage[main]/Redis::Master/File[/etc/redis.conf]: Scheduling refresh of Service[redis]
Notice: /Stage[main]/Redis/Service[redis]/ensure: ensure changed ‘stopped’ to ‘running’
Info: /Stage[main]/Redis/Service[redis]: Unscheduling refresh on Service[redis]
Notice: Finished catalog run in 0.32 seconds
^CNotice: Caught INT; calling stop
[root@node01 ~]# ss -tnl
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 :22 :
LISTEN 0 100 127.0.0.1:25 :
LISTEN 0 128 :6379 :
LISTEN 0 128 [::]:22 [::]:
LISTEN 0 100 [::1]:25 [::]:
[root@node01 ~]# grep -Ei “^port|bind|requirepass” /etc/redis.conf
bind 0.0.0.0
port 6379
requirepass admin123.com
[root@node01 ~]#
　　提示：可以看到node01上的reids已经正常启动，并且配置文件中的内容也是我们传递参数的内容；

在node02上手动拉取配置并运行，看看对应redis是否正常运行？配置文件下是否是我们指定传递的参数的配置文件？

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
[root@node02 ~]# puppet agent -v --no-daemonize
Notice: Starting Puppet client version 3.6.2
Info: Retrieving pluginfacts
Info: Retrieving plugin
Info: Caching catalog for node02.test.org
Warning: The package type’s allow_virtual parameter will be changing its default value from false to true in a future release. If you do not want to allow virtual packages, please explicitly set allow_virtual to false.
(at /usr/share/ruby/vendor_ruby/puppet/type.rb:816:in `set_default’)
Info: Applying configuration version ‘1607328529’
Notice: /Stage[main]/Redis/Package[redis]/ensure: created
Info: FileBucket got a duplicate file {md5}d98629fded012cd2a25b9db0599a9251
Info: /Stage[main]/Redis::Slave/File[/etc/redis.conf]: Filebucketed /etc/redis.conf to puppet with sum d98629fded012cd2a25b9db0599a9251
Notice: /Stage[main]/Redis::Slave/File[/etc/redis.conf]/content: content changed ‘{md5}d98629fded012cd2a25b9db0599a9251’ to ‘{md5}d1f0efeaee785f0d26eb2cd82acaf1f9’
Notice: /Stage[main]/Redis::Slave/File[/etc/redis.conf]/mode: mode changed ‘0640’ to ‘0644’
Info: /Stage[main]/Redis::Slave/File[/etc/redis.conf]: Scheduling refresh of Service[redis]
Info: /Stage[main]/Redis::Slave/File[/etc/redis.conf]: Scheduling refresh of Service[redis]
Notice: /Stage[main]/Redis/Service[redis]/ensure: ensure changed ‘stopped’ to ‘running’
Info: /Stage[main]/Redis/Service[redis]: Unscheduling refresh on Service[redis]
Notice: Finished catalog run in 10.62 seconds
^CNotice: Caught INT; calling stop
[root@node02 ~]# ss -tnl
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 :22 :
LISTEN 0 100 127.0.0.1:25 :
LISTEN 0 128 :6379 :
LISTEN 0 128 [::]:22 [::]:
LISTEN 0 100 [::1]:25 [::]:
[root@node02 ~]# grep -Ei “^port|requirepass|slaveof|masterauth” /etc/redis.conf
port 6379
slaveof 192.168.16.152 6379
masterauth admin123.com
[root@node02 ~]#
　　提示：可以看到node02上的redis已经正常跑起来了，并且配置文件中的对应参数也是为指定传入的参数；

验证：在node01上登陆redis，创建一个key，看看对应node02上的redis是否正常将node01上创建的key同步到node02上的redis上？

1
2
3
4
5
6
7
8
9
10
11
12
[root@node01 ~]# redis-cli
127.0.0.1:6379> AUTH admin123.com
OK
127.0.0.1:6379> set test test
OK
127.0.0.1:6379> get test
“test”
127.0.0.1:6379> exit
[root@node01 ~]# redis-cli -h node02.test.org -a admin123.com
node02.test.org:6379> get test
“test”
node02.test.org:6379>
　　提示：可以看到在node01上的redis上写入一个key在node02上能够看到对应key的值，说明redis的主从复制正常；以上就是puppet master/agent模型上安装启动redis主从的简单示例；

puppet多环境设定

所谓多环境指一个agent通过指定不同的环境名称，可以在master上获取不同环境的配置清单，从而实现在不同的环境中，对应的agent拉取不同的配置到本地应用；默认不配置多环境，在agent来master拉取配置都是拉取的生产环境的配置（production）；

示例：在master端上配置多个环境，对应节点使用传递不同环境来获取不同环境的配置

提示：在master端端配置文件中新整以上master配置段，并指明environmentpath位置；这里需要注意，更改了配置文件需要重启服务才可生效；

查看confdir的位置

1
2
3
[root@master ~]# puppet config print |grep confdir
confdir = /etc/puppet
[root@master ~]#
　　在/etc/puppet目录下创建environments目录，并在其下创建对应环境名的目录以及子目录

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
[root@master ~]# mkdir -pv /etc/puppet/environments/{testing,development,production}/{manifests,modules}
mkdir: created directory ‘/etc/puppet/environments’
mkdir: created directory ‘/etc/puppet/environments/testing’
mkdir: created directory ‘/etc/puppet/environments/testing/manifests’
mkdir: created directory ‘/etc/puppet/environments/testing/modules’
mkdir: created directory ‘/etc/puppet/environments/development’
mkdir: created directory ‘/etc/puppet/environments/development/manifests’
mkdir: created directory ‘/etc/puppet/environments/development/modules’
mkdir: created directory ‘/etc/puppet/environments/production’
mkdir: created directory ‘/etc/puppet/environments/production/manifests’
mkdir: created directory ‘/etc/puppet/environments/production/modules’
[root@master ~]# tree /etc/puppet/environments/
/etc/puppet/environments/
├── development
│ ├── manifests
│ └── modules
├── production
│ ├── manifests
│ └── modules
└── testing
├── manifests
└── modules

9 directories, 0 files
[root@master ~]#
　　提示：在每个环境目录下必须创建manifests和modules目录，manifests目录用于存放站点清单，modules用于存放模块；

在testing环境下创建安装memcached，并指定监听在11211端口的模块和站点清单，配置development环境下的memcached监听在11212端口，配置prediction环境的memcached监听在11213端口

创建memcached模块目录结构

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
[root@master ~]# mkdir -pv /etc/puppet/environments/testing/modules/memcached/{manifests,templates,files,lib,spec,tests}
mkdir: created directory ‘/etc/puppet/environments/testing/modules/memcached’
mkdir: created directory ‘/etc/puppet/environments/testing/modules/memcached/manifests’
mkdir: created directory ‘/etc/puppet/environments/testing/modules/memcached/templates’
mkdir: created directory ‘/etc/puppet/environments/testing/modules/memcached/files’
mkdir: created directory ‘/etc/puppet/environments/testing/modules/memcached/lib’
mkdir: created directory ‘/etc/puppet/environments/testing/modules/memcached/spec’
mkdir: created directory ‘/etc/puppet/environments/testing/modules/memcached/tests’
[root@master ~]# tree /etc/puppet/environments/testing/modules/memcached/
/etc/puppet/environments/testing/modules/memcached/
├── files
├── lib
├── manifests
├── spec
├── templates
└── tests

6 directories, 0 files
[root@master ~]#
　　在testing环境下的manifests目录下创建init.pp文件

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
[root@master ~]# cat /etc/puppet/environments/testing/modules/memcached/manifests/init.pp
class memcached{
package{“memcached”:
ensure => installed,
} ->
file{"/etc/sysconfig/memcached":
ensure => file,
source => ‘puppet:///modules/memcached/memcached’,
owner => ‘root’,
group => ‘root’,
mode => ‘0644’,
} ~>
service{“memcached”:
ensure => running,
enable => true
}
}
[root@master ~]#
　　在testing环境下的files目录提供memcached配置文件

1
2
3
4
5
6
7
[root@master ~]# cat /etc/puppet/environments/testing/modules/memcached/files/memcached
PORT=“11211”
USER=“memcached”
MAXCONN=“1024”
CACHESIZE=“64”
OPTIONS=""
[root@master ~]#
　　在testing环境下的manifests目录下创建站点文件site.pp

1
2
3
4
5
[root@master ~]# cat /etc/puppet/environments/testing/manifests/site.pp
node ‘node02.test.org’{
include memcached
}
[root@master ~]#
　　提示：到此testing环境的memcached模块的配置和站点清单就配置好了；

配置development环境下的memcached模块和站点清单

复制testing环境下的模块到development

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[root@master ~]# cp -a /etc/puppet/environments/testing/modules/memcached/ /etc/puppet/environments/development/modules/
[root@master ~]# tree /etc/puppet/environments/development/modules/
/etc/puppet/environments/development/modules/
└── memcached
├── files
│ └── memcached
├── lib
├── manifests
│ └── init.pp
├── spec
├── templates
└── tests

7 directories, 2 files
[root@master ~]#
　　编辑files目录下的memcached配置文件，修改监听端口为11212

1
2
3
4
5
6
7
[root@master ~]# cat /etc/puppet/environments/development/modules/memcached/files/memcached
PORT=“11212”
USER=“memcached”
MAXCONN=“1024”
CACHESIZE=“64”
OPTIONS=""
[root@master ~]#
　　提示：testing和development环境下的memcached除了配置文件不同，站点清单一样；

复制testing环境下的站点清单到development环境下

1
2
3
4
5
6
[root@master ~]# cp /etc/puppet/environments/testing/manifests/site.pp /etc/puppet/environments/development/manifests/
[root@master ~]# cat /etc/puppet/environments/development/manifests/site.pp
node ‘node02.test.org’{
include memcached
}
[root@master ~]#
　　提示：到此development环境下到memcached模块和站点清单文件就配置好了；

配置production环境下的memcached模块和站点清单

复制testing环境下到memcached模块

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[root@master ~]# cp -a /etc/puppet/environments/testing/modules/memcached/ /etc/puppet/environments/production/modules/
[root@master ~]# tree /etc/puppet/environments/production/modules/
/etc/puppet/environments/production/modules/
└── memcached
├── files
│ └── memcached
├── lib
├── manifests
│ └── init.pp
├── spec
├── templates
└── tests

7 directories, 2 files
[root@master ~]#
　　编辑files目录下的memcached配置文件，修改监听端口为11213

1
2
3
4
5
6
7
[root@master ~]# cat /etc/puppet/environments/production/modules/memcached/files/memcached
PORT=“11213”
USER=“memcached”
MAXCONN=“1024”
CACHESIZE=“64”
OPTIONS=""
[root@master ~]#
　　复制站点文件到production环境下

1
2
3
4
5
6
[root@master ~]# cp /etc/puppet/environments/testing/manifests/site.pp /etc/puppet/environments/production/manifests/
[root@master ~]# cat /etc/puppet/environments/production/manifests/site.pp
node ‘node02.test.org’{
include memcached
}
[root@master ~]#
　　提示：到此三个环境到memcached模块和站点清单都准备好了；

在node02上使用不同的环境来启动监听不同的端口的memcached

使用testing环境

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
[root@node02 ~]# puppet agent -v --no-daemonize --environment=testing
Notice: Starting Puppet client version 3.6.2
Info: Retrieving pluginfacts
Info: Retrieving plugin
Info: Caching catalog for node02.test.org
Warning: The package type’s allow_virtual parameter will be changing its default value from false to true in a future release. If you do not want to allow virtual packages, please explicitly set allow_virtual to false.
(at /usr/share/ruby/vendor_ruby/puppet/type.rb:816:in `set_default’)
Info: Applying configuration version ‘1607334372’
Notice: /Stage[main]/Memcached/Service[memcached]/ensure: ensure changed ‘stopped’ to ‘running’
Info: /Stage[main]/Memcached/Service[memcached]: Unscheduling refresh on Service[memcached]
Notice: Finished catalog run in 0.29 seconds
^CNotice: Caught INT; calling stop
[root@node02 ~]# ss -tnl
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 :22 :
LISTEN 0 100 127.0.0.1:25 :
LISTEN 0 128 :11211 :
LISTEN 0 128 :6379 :
LISTEN 0 128 [::]:22 [::]:
LISTEN 0 100 [::1]:25 [::]:
LISTEN 0 128 [::]:11211 [::]:
[root@node02 ~]#
　　提示：可以看到在node02上使用–envrionment选项指定为testing环境，对应启动的memcached的端口为11211；

使用development环境

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
[root@node02 ~]# puppet agent -v --no-daemonize --environment=development
Notice: Starting Puppet client version 3.6.2
Info: Retrieving pluginfacts
Info: Retrieving plugin
Info: Caching catalog for node02.test.org
Warning: The package type’s allow_virtual parameter will be changing its default value from false to true in a future release. If you do not want to allow virtual packages, please explicitly set allow_virtual to false.
(at /usr/share/ruby/vendor_ruby/puppet/type.rb:816:in `set_default’)
Info: Applying configuration version ‘1607334662’
Info: /Stage[main]/Memcached/File[/etc/sysconfig/memcached]: Filebucketed /etc/sysconfig/memcached to puppet with sum 05503957e3796fbe6fddd756a7a102a0
Notice: /Stage[main]/Memcached/File[/etc/sysconfig/memcached]/content: content changed ‘{md5}05503957e3796fbe6fddd756a7a102a0’ to ‘{md5}b69eb8ec579bb28f4140f7debf17f281’
Info: /Stage[main]/Memcached/File[/etc/sysconfig/memcached]: Scheduling refresh of Service[memcached]
Notice: /Stage[main]/Memcached/Service[memcached]: Triggered ‘refresh’ from 1 events
Notice: Finished catalog run in 0.34 seconds
^CNotice: Caught INT; calling stop
[root@node02 ~]# ss -tnl
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 :22 :
LISTEN 0 100 127.0.0.1:25 :
LISTEN 0 128 :6379 :
LISTEN 0 128 :11212 :
LISTEN 0 128 [::]:22 [::]:
LISTEN 0 100 [::1]:25 [::]:
LISTEN 0 128 [::]:11212 [::]:
[root@node02 ~]#
　　提示：可以看到使用development环境，对应启动的memcached的端口就是11212;

使用production环境

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
[root@node02 ~]# puppet agent -v --no-daemonize --environment=production
Notice: Starting Puppet client version 3.6.2
Info: Retrieving pluginfacts
Info: Retrieving plugin
Info: Caching catalog for node02.test.org
Warning: The package type’s allow_virtual parameter will be changing its default value from false to true in a future release. If you do not want to allow virtual packages, please explicitly set allow_virtual to false.
(at /usr/share/ruby/vendor_ruby/puppet/type.rb:816:in `set_default’)
Info: Applying configuration version ‘1607334761’
Info: /Stage[main]/Memcached/File[/etc/sysconfig/memcached]: Filebucketed /etc/sysconfig/memcached to puppet with sum b69eb8ec579bb28f4140f7debf17f281
Notice: /Stage[main]/Memcached/File[/etc/sysconfig/memcached]/content: content changed ‘{md5}b69eb8ec579bb28f4140f7debf17f281’ to ‘{md5}f7cdb226870b0164bbdb8671eb11e433’
Info: /Stage[main]/Memcached/File[/etc/sysconfig/memcached]: Scheduling refresh of Service[memcached]
Notice: /Stage[main]/Memcached/Service[memcached]: Triggered ‘refresh’ from 1 events
Notice: Finished catalog run in 0.33 seconds
^CNotice: Caught INT; calling stop
[root@node02 ~]# ss -tnl
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 :22 :
LISTEN 0 100 127.0.0.1:25 :
LISTEN 0 128 :6379 :
LISTEN 0 128 :11213 :
LISTEN 0 128 [::]:22 [::]:
LISTEN 0 100 [::1]:25 [::]:
LISTEN 0 128 [::]:11213 [::]:
[root@node02 ~]#
　　提示：可以看到memcached监听地址就变为了11213;如以上测试没有任何问题，接下来就可以确定对应agent的环境，配置agent的环境；

配置agent端的环境

提示：在agent的配置文件中使用environment指定对应的环境名称，保存退出重启puppetagent服务，现在这台agent就会周期性的到master端拉取testing环境的相关配置；

以上就是puppet的master/agent模型以及多环境相关配置的使用和测试，能帮到大家。做这些简单的事情，也可以让代理游戏播种下优质的基因，最终长成参天大树，成为行业的翘楚。；