Puppet小试牛刀
ansible比较适合做“一次性”的工作,例如,系统部署、应用发布、打补丁等等,而且ansible一次性默认只能在5台主机上批量执行命令,规模再大点的就不太适用了,这时。候就轮到Puppet上场了。Puppet是由Ruby语言开发,执行高效,而且Puppet的foreman可以提供比较高大上的web界面,更方便使用和管理节点。Puppet历史悠久,使用率比ansible更高,且其功能和性能更加稳定,唯一不足的是Puppet的运行需要agent端也安装Puppet,下面列出常用自动化部署工具对比
Puppet基础
puppet的基础概念和架构
puppet的工作模型:
- 单机模型(standalone):手动应用清单;
- master/agent:由agent周期性地向Master请求清单并自动应用于本地;
单机模型:
配置文件:/etc/puppet/puppet.conf
主程序:/usr/bin/puppet
puppet程序:Usage: puppet <subcommand> [options] <action> [options]
puppet help <subcommand>' for help on a specific subcommand.
puppet help <subcommand> <action>' for help on a specific subcommand action.
puppet apply:Applies a standalone Puppet manifest to the local system.
puppet apply [-d|--debug] [-v|--verbose] [-e|--execute] [--noop] <file>
puppet describe:Prints help about Puppet resource types, providers, and metaparameters.
puppet describe [-h|--help] [-s|--short] [-p|--providers] [-l|--list] [-m|--meta] [type]
- -l:列出所有资源类型;
- -s:显示指定类型的简要帮助信息;
- -m:显示指定类型的元参数,一般与-s一同使用;
资源定义:向资源类型的属性赋值来实现,可称为资源类型实例化;定义了资源实例的文件即清单,manifest;
定义资源的语法:
type {'title':
attribute1 => value1,
atrribute2 => value2,
……
}
注意:type必须使用小写字符;title是一个字符串,在同一类型中必须惟一;
资源属性中的三个特殊属性:
- Namevar, 可简称为name;
- ensure:资源的目标状态;
- Provider:指明资源的管理接口;
Puppet的资源类型
group:Manage groups.
属性:
- name:组名;
- gid:GID;
- system:是否为系统组,true OR false;
- ensure:目标状态,present/absent;
- members:成员用户;
user:Manage users.
属性:
- name:用户名;
- uid: UID;
- gid:基本组ID;
- groups:附加组,不能包含基本组;
- comment:注释;
- expiry:过期时间 ;
- home:家目录;
- shell:默认shell类型;
- system:是否为系统用户 ;
- ensure:present/absent;
- password:加密后的密码串;
资源引用:Type['title'] 类型的首字母必须大写;
关系元参数:before/require
A before B: B依赖于A,定义在A资源中;
{
...
before => Type['B'],
...
}
B require A: B依赖于A,定义在B资源中;
{
...
require => Type['A'],
...
}
package:Manage packages.
属性:
- ensure:installed, present, latest, absent, any version string (implies present)
- name:包名;
- source:程序包来源,仅对不会自动下载相关程序包的provider有用,例如rpm或dpkg;
- provider:指明安装方式;
service:Manage running services.
属性:
- ensure:Whether a service should be running. Valid values are `stopped` (also called `false`), `running` (also called `true`).
- enable:Whether a service should be enabled to start at boot. Valid values are `true`, `false`, `manual`.
- name:
- path:The search path for finding init scripts. 如:/bin/:/sbin/:/usr/sbin/:/usr/bin: 脚本的搜索路径,默认为/etc/init.d/;
- hasrestart: (true)
- hasstatus:
- start:手动定义启动命令,如‘service start redis’;
- stop:
- status:
- restart:Specify a *restart* command manually. If left unspecified, the service will be stopped and then started. 通常用于定义reload操作;
file:Manages files, including their content, ownership, and permissions.
- ensure:Possible values are `present`, `absent`, `file`, `directory`, and `link`.
- path:文件路径;
- source:源文件;
- content:文件内容;
- target:符号链接的目标文件;
- owner:属主
- group:属组
- mode:权限;
- atime/ctime/mtime:时间戳;
资源有特殊属性:
名称变量(namevar):name可省略,此时将由title表示;
ensure:定义资源的目标状态;
元参数:metaparameters
依赖关系:before、require
通知关系:通知相关的其它资源进行“刷新”操作;notify/subscribe
A notify B:B依赖于A,且A发生改变后会通知B;
{
...
notify => Type['B'],
...
}
B subscribe A:B依赖于A,且B监控A资源的变化产生的事件;
{
...
subscribe => Type['A'],
...
}
示例1:
file{'test.txt':
path => '/tmp/test.txt',
ensure => file,
source => '/etc/fstab',
}
file{'test.symlink':
path => '/tmp/test.symlink',
ensure => link,
target => '/tmp/test.txt',
require => File['test.txt'],
}
file{'test.dir':
path => '/tmp/test.dir',
ensure => directory,
source => '/etc/yum.repos.d/',
recurse => true,
}
示例2:
service{'httpd':
ensure => running,
enable => true,
restart => 'systemctl restart httpd.service',
# subscribe => File['httpd.conf'],
}
package{'httpd':
ensure => installed,
}
file{'httpd.conf':
path => '/etc/httpd/conf/httpd.conf',
source => '/root/manifests/httpd.conf',
ensure => file,
notify => Service['httpd'],
}
Package['httpd'] -> File['httpd.conf'] -> Service['httpd']
exec:相当于ansible的command和shell.
- **command** (*namevar*):要运行的命令;
- cwd:The directory from which to run the command.
- **creates**:文件路径,仅此路径表示的文件不存在时,command方才执行;
- user/group:运行命令的用户身份;
- path:The search path used for command execution. Commands must be fully qualified if no path is specified.
- onlyif:此属性指定一个命令,此命令正常(退出码为0)运行时,当前command才会运行;
- unless:此属性指定一个命令,此命令非正常(退出码为非0)运行时,当前command才会运行;
- refresh:重新执行当前command的替代命令;
- refreshonly:仅接收到订阅的资源的通知时方才运行;
cron:指定周期性任务计划
- command:要执行的任务;
- ensure:present/absent;
- hour:
- minute:
- monthday:
- month:
- weekday:
- user:以哪个用户的身份运行命令
- target:添加为哪个用户的任务
- name:cron job的名称;
示例:
cron{'timesync':
command => '/usr/sbin/ntpdate 172.18.0.1 &> /dev/null',
ensure => present,
minute => '*/3',
user => 'root',
}
notify:Sends an arbitrary message to the agent run-time log.
属性:
- message:信息内容
- name:信息名称;
核心类型:
- group: 组
- user:用户
- packge:程序包
- service:服务
- file:文件
- exec:执行自定义命令,要求幂等
- cron:周期性任务计划
- notify:通知
可以为资源定义tag,格式:
type{'title':
...
tag => 'TAG1',
}
type{'title':
...
tag => ['TAG1','TAG2',...],
}
手动调用:
puppet apply --tags TAG1,TAG2,... FILE.PP
Puppet的变量、语法和类
puppet variable:$variable_name=value
数据类型:
- 字符型:引号可有可无;但单引号为强引用,双引号为弱引用;
- 数值型:默认均识别为字符串,仅在数值上下文才以数值对待;
- 数组:[]中以逗号分隔元素列表;
- 布尔型值:true, false;
- hash:{}中以逗号分隔k/v数据列表; 键为字符型,值为任意puppet支持的类型;{ 'mon' => 'Monday', 'tue' => 'Tuesday', };
- undef:未定义 ;
正则表达式:
(?<ENABLED OPTION>:<PATTERN>)
(?-<DISABLED OPTION>:<PATTERN>)
OPTIONS:
i:忽略字符大小写;
m:把.当换行符;
x:忽略<PATTERN>中的空白字符
(?i-mx:PATTERN)
puppet的变量种类:
facts:直接运行facter即可查看,由facter提供;top scope;
内建变量:
master端变量,$servername, $serverip, $serverversion
agent端变量 ,$clientcert, $clientversion, $environment
parser变量,$module_name
用户自定义变量:
变量有作用域,称为Scope;
- top scope: $::var_name
- node scope
- class scope
puppet流程控制语句:
if语句:
if CONDITION {
...
} else {
...
}
CONDITION的给定方式:
(1) 变量
(2) 比较表达式
(3) 有返回值的函数
case语句:
case CONTROL_EXPRESSION {
case1: { ... }
case2: { ... }
case3: { ... }
...
default: { ... }
}
CONTROL_EXPRESSION:
(1) 变量
(2) 表达式
(3) 有返回值的函数
各case的给定方式:
(1) 直接字串;
(2) 变量
(3) 有返回值的函数
(4) 正则表达式模式;
(5) default
例:
case $osfamily {
"RedHat": { $webserver='httpd' }
/(?i-mx:debian)/: { $webserver='apache2' }
default: { $webserver='httpd' }
}
package{"$webserver":
ensure => installed,
before => [ File['httpd.conf'], Service['httpd'] ],
}
file{'httpd.conf':
path => '/etc/httpd/conf/httpd.conf',
source => '/root/manifests/httpd.conf',
ensure => file,
}
service{'httpd':
ensure => running,
enable => true,
restart => 'systemctl restart httpd.service',
subscribe => File['httpd.conf'],
}
selector语句:
CONTROL_VARIABLE ? {
case1 => value1,
case2 => value2,
...
default => valueN,
}
CONTROL_VARIABLE的给定方法:
(1) 变量
(2) 有返回值的函数
各case的给定方式:
(1) 直接字串;
(2) 变量
(3) 有返回值的函数
(4) 正则表达式模式;
(5) default
注意:不能使用列表格式;但可以是其它的selecor;
例1:
class nginx {
package{'nginx':
ensure => installed,
}
service{'nginx':
ensure => running,
enable => true,
restart => '/usr/sbin/nginx -s reload',
}
}
class nginx::web inherits nginx {
Service['nginx'] {
subscribe => File['ngx-web.conf'],
}
file{'ngx-web.conf':
path => '/etc/nginx/conf.d/ngx-web.conf',
ensure => file,
source => '/root/manifests/ngx-web.conf',
}
}
class nginx::proxy inherits nginx {
Service['nginx'] {
subscribe => File['ngx-proxy.conf'],
}
file{'ngx-proxy.conf':
path => '/etc/nginx/conf.d/ngx-proxy.conf',
ensure => file,
source => '/root/manifests/ngx-proxy.conf',
}
}
例2:
$webserver = $osfamily ? {
"Redhat" => 'httpd',
/(?i-mx:debian)/ => 'apache2',
default => 'httpd',
}
package{"$webserver":
ensure => installed,
before => [ File['httpd.conf'], Service['httpd'] ],
}
file{'httpd.conf':
path => '/etc/httpd/conf/httpd.conf',
source => '/root/manifests/httpd.conf',
ensure => file,
}
service{'httpd':
ensure => running,
enable => true,
restart => 'systemctl restart httpd.service',
subscribe => File['httpd.conf'],
}
puppet的类:
类:puppet中命名的代码模块,常用于定义一组通用目标的资源,可在puppet全局调用;类可以被继承,也可以包含子类;
语法格式:
class NAME {
...puppet code...
}
class NAME(parameter1, parameter2) {
...puppet code...
}
类代码只有声明后才会执行,调用方式:
(1) include CLASS_NAME1, CLASS_NAME2, ...
(2) class{'CLASS_NAME':
attribute => value,
}
示例1:
class apache2 {
$webpkg = $operatingsystem ? {
/(?i-mx:(centos|redhat|fedora))/ => 'httpd',
/(?i-mx:(ubuntu|debian))/ => 'apache2',
default => 'httpd',
}
package{"$webpkg":
ensure => installed,
}
file{'/etc/httpd/conf/httpd.conf':
ensure => file,
owner => root,
group => root,
source => '/tmp/httpd.conf',
require => Package["$webpkg"],
notify => Service['httpd'],
}
service{'httpd':
ensure => running,
enable => true,
}
}
include apache2
示例2:
class dbserver($pkg,$srv) {
package{"$pkg":
ensure => latest,
}
service{"$srv":
ensure => running,
}
}
if $operatingsystem == "CentOS" or $operatingsystem == "RedHat" {
case $operatingsystemmajrelease {
'7': { $pkgname = 'mariadb-server' $srvname='mariadb' }
default: { $pkgname = 'mysql-server' $srvname='mysqld' }
}
}
class{'dbserver':
pkg => "$pkgname",
srv => "$srvname",
}
类继承的方式:
class SUB_CLASS_NAME inherits PARENT_CLASS_NAME {
...puppet code...
}
示例1:
class nginx {
package{'nginx':
ensure => installed,
}
service{'nginx':
ensure => running,
enable => true,
restart => '/usr/sbin/nginx -s reload',
}
}
class nginx::web inherits nginx {
Service['nginx'] {
subscribe => File['ngx-web.conf'],
}
file{'ngx-web.conf':
path => '/etc/nginx/conf.d/ngx-web.conf',
ensure => file,
source => '/root/manifests/ngx-web.conf',
}
}
class nginx::proxy inherits nginx {
Service['nginx'] {
subscribe => File['ngx-proxy.conf'],
}
file{'ngx-proxy.conf':
path => '/etc/nginx/conf.d/ngx-proxy.conf',
ensure => file,
source => '/root/manifests/ngx-proxy.conf',
}
}
示例2:
class redis {
package{'redis':
ensure => latest,
} ->
service{'redis':
ensure => running,
enable => true,
}
}
class redis::master inherits redis {
file{'/etc/redis.conf':
ensure => file,
source => '/root/manifests/redis-master.conf',
owner => redis,
group => root,
}
Service['redis'] {
subscribe => File['/etc/redis.conf'],
}
}
class redis::slave inherits redis {
file{'/etc/redis.conf':
ensure => file,
source => '/root/manifests/redis-slave.conf',
owner => redis,
group => root,
}
Service['redis'] {
subscribe => File['/etc/redis.conf'],
}
}
在子类中为父类的资源新增属性或覆盖指定的属性的值:
Type['title'] {
attribute1 => value,
...
}
在子类中为父类的资源的某属性增加新值:
Type['title'] {
attribute1 +> value,
...
}
puppet模板
erb:模板语言,embedded ruby;
puppet兼容的erb语法:https://docs.puppet.com/puppet/latest/reference/lang_template_erb.html
file{'title':
ensure => file,
path =>
content => template('/PATH/TO/ERB_FILE'),
}
文本文件中内嵌变量替换机制:<%= @VARIABLE_NAME %>
示例:
class nginx {
package{'nginx':
ensure => installed,
}
service{'nginx':
ensure => running,
enable => true,
require => Package['nginx'],
}
}
class nginx::web inherits nginx {
file{'ngx-web.conf':
path => '/etc/nginx/conf.d/ngx-web.conf',
ensure => file,
require => Package['nginx'],
source => '/root/manifests/nginx/ngx-web.conf',
}
file{'nginx.conf':
path => '/etc/nginx/nginx.conf',
ensure => file,
content => template('/root/manifests/nginx.conf.erb'),
require => Package['nginx'],
}
Service['nginx'] {
subscribe => [ File['ngx-web.conf'], File['nginx.conf'] ],
}
}
include nginx::web
puppet模块:
模块就是一个按约定的、预定义的层级结构存放了多个文件或子目录的目录,目录里的这些文件或子目录必须遵循一定格式的命名规范;
puppet会在配置的路径下查找所需要的模块;
模块名只能以小写字母开头,可以包含小写字母、数字和下划线;但不能使用”main"和"settings“;
- manifests/init.pp:必须一个类定义,类名称必须与模块名称相同;
- files/:静态文件;
- puppet URL:
- puppet:///modules/MODULE_NAME/FILE_NAME
- templates/:
- tempate('MOD_NAME/TEMPLATE_FILE_NAME')
- lib/:插件目录,常用于存储自定义的facts以及自定义类型;
- spec/:类似于tests目录,存储lib/目录下插件的使用帮助和范例;
- tests/:当前模块的使用帮助或使用范例文件;
注意:
1、puppet 3.8及以后的版本中,资源清单文件的文件名要与文件子类名保持一致,例如某子类名为“base_class::child_class”,其文件名应该为child_class.pp;
2、无需在资源清单文件中使用import语句;
3、manifests目录下可存在多个清单文件,每个清单文件包含一个类,其文件名同类名;
puppet config命令:
获取或设定puppet配置参数:puppet config print [argument]
puppet查找模块文件的路径:modulepath
mariadb模块中的清单文件示例:
class mariadb($datadir='/var/lib/mysql') {
package{'mariadb-server':
ensure => installed,
}
file{"$datadir":
ensure => directory,
owner => mysql,
group => mysql,
require => [ Package['mariadb-server'], Exec['createdir'], ],
}
exec{'createdir':
command => "mkdir -pv $datadir",
require => Package['mariadb-server'],
path => '/bin:/sbin:/usr/bin:/usr/sbin',
creates => “$datadir",
}
file{'my.cnf':
path => '/etc/my.cnf',
content => template('mariadb/my.cnf.erb'),
require => Package['mariadb-server'],
notify => Service['mariadb'],
}
service{'mariadb':
ensure => running,
enable => true,
require => [ Exec['createdir'], File["$datadir"], ],
}
}
Puppet官网下载:https://puppet.com/products/open-source-projects
阿里云镜像下载:https://yum.puppet.com/