首页 > 代码库 > puppet集群
puppet集群
实验目的：
由于现有的环境中，puppetmaster是单节点，客户端更新时出现了更新失败和时间较长等现象。
考虑将puppetmaster做成集群的模式，解决大量客户端更新延时和单节点故障问题。主要解决证书问题
环境：
puppetmaster两台：
Puppetmaster1：10.9.24.168
Puppetmaster2：10.9.24.184
Nginx:
Nginx: 10.9.24.183
Client:
Client: 10.9.3.153
yum install puppet-server rubygem-mongrel
配置：
在/etc/hosts写好对应关系
Puppetmaster1:
/etc/sysconfig/puppetmaster(启用mongrel模式)
PUPPETMASTER_PORTS=( 18140 18141 18142 18143 )
PUPPETMASTER_EXTRA_OPTS="—servertype=mongrel --ssl_client_header=HTTP_X_SSL_SUBJECT"
/etc/puppet/puppet.conf:
[main]
bindaddress = 0.0.0.0
（加上这句让puppetmaster监听地址为0.0.0.0，否则只以localhost监听）
netstat -antp
tcp 0 0 0.0.0.0:18140 0.0.0.0:* LISTEN 22822/ruby
tcp 0 0 0.0.0.0:18141 0.0.0.0:* LISTEN 22872/ruby
tcp 0 0 0.0.0.0:18142 0.0.0.0:* LISTEN 22922/ruby
tcp 0 0 0.0.0.0:18143 0.0.0.0:* LISTEN 22972/ruby
/etc/exports:
/var/lib/puppet 10.9.24.184(rw,sync)
(用nfs共享的模式，同步证书)
/etc/puppet//fileserver.conf:
[files]
path /var/lib/puppet/files
allow *
Puppetmaster2:
配置文件参照puppetmaster1
service puppetmaster start,启动puppetmaster
在/etc/fstab中加入如下：
10.9.24.168:/var/lib/puppet /var/lib/puppet nfs defaults 0 0
mount -a //挂载/var/lib/puppet
Nginx:
将puppetmaster1/var/lib/puppet下的文件拷贝到本机/var/lib/puppet下（用nfs挂载也可以）
cat /etc/nginx/nginx.conf:
user daemon daemon;
worker_processes 2;
error_log /var/log/nginx/error.log notice;
pid /var/run/nginx.pid;
events {
worker_connections 1024;
}
http {
default_type application/octet-stream;
log_format main ‘$remote_addr - $remote_user [$time_local] "$request" ‘
‘$status $body_bytes_sent "$http_referer" ‘
‘"$http_user_agent" "$http_x_forwarded_for"‘;
access_log /var/log/nginx/access.log main;
sendfile on;
tcp_nopush on;
tcp_nodelay on;
large_client_header_buffers 16 4k;
proxy_buffers 128 4k;
keepalive_timeout 65;
ssl on;
upstream puppet-production {
server 10.9.24.168:18140;
server 10.9.24.168:18141;
server 10.9.24.168:18142;
server 10.9.24.168:18143;
server 10.9.24.184:18140;
server 10.9.24.184:18141;
server 10.9.24.184:18142;
server 10.9.24.184:18143;
}
//nginx分发请求
server {
listen 8140;
ssl on;
ssl_session_timeout 5m;
ssl_certificate /var/lib/puppet/ssl/certs/xx.pem;
ssl_certificate_key /var/lib/puppet/ssl/private_keys/xx.pem;
ssl_client_certificate /var/lib/puppet/ssl/ca/ca_crt.pem;
ssl_ciphers SSLv2:-LOW:-EXPORT:RC4+RSA;
ssl_verify_client optional;
ssl_crl /var/lib/puppet/ssl/ca/ca_crl.pem;
（证书位置，先复制到本地）
access_log /var/log/host.access.log main;
location / {
proxy_pass http://puppet-production;
proxy_redirect off;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Client-Verify $ssl_client_verify;
proxy_set_header X-Client-DN $ssl_client_s_dn;
proxy_set_header X-SSL-Subject $ssl_client_s_dn;
proxy_set_header X-SSL-Issuer $ssl_client_i_dn;
proxy_read_timeout 65;
}
}
}
现在puppetmaster是用同步方式实现的，可以考虑用证书链的方式实现
Client:
我故意在puppetmaster1和puppetmaster2配置不同的文件：
puppetmaster1
cat /etc/puppet/manifests/main.pp:
node default {
file { "/tmp/temp1.txt": content => "Hello,from puppetmaster1\n"; }
file { "/tmp/test.file":
path => "/tmp/test.file",
source => "puppet://puppet/files/test.file",
owner => "root",
group => "root",
mode => 644,
}
}
puppetmaster2
node default {
file { "/tmp/temp1.txt": content => "Hello,from puppetmaster2\n"; }
file { "/tmp/test.file":
path => "/tmp/test.file",
source => "puppet://puppet/files/test.file",
owner => "root",
group => "root",
mode => 644,
}
}
Client:
puppetd -t:
info: Caching catalog for taffy
info: Applying configuration version ‘1301714542‘
notice: /Stage[main]//Node[default]/File[/tmp/test.file]/ensure: content changed ‘{md5}52688f6d76cbeccd058bdf6f412b4da0‘ to ‘{md5}52688f6d76cbeccd058bdf6f412b4da0‘
notice: /Stage[main]//Node[default]/File[/tmp/temp1.txt]/checksum: defined ‘checksum‘ as ‘{md5}1cf6c62d2c8ddde94e97aa9140861b0e‘
notice: /Stage[main]//Node[default]/File[/tmp/temp1.txt]/content: defined content as ‘unknown checksum‘
notice: Finished catalog run in 0.13 seconds
cat /tmp/temp1.txt:
Hello,from puppetmaster1 (从puppetmaster1取过来的)
puppetd -t: （再推一次）
info: Caching catalog for taffy
info: Applying configuration version ‘1301505457‘
--- /tmp/temp1.txt 2011-04-02 11:20:38.509009000 +0800
+++ /tmp/puppet-diffing20110402-1217-jqru9o-0 2011-04-02 11:30:10.258009000 +0800
@@ -1 +1 @@
-Hello,from puppetmaster1
+Hello,from puppetmaster2 （过来了，从puppetmaster2取出的）
info: /Stage[main]//Node[default]/File[/tmp/temp1.txt]: Filebucketed /tmp/temp1.txt to puppet with sum cab0e6a2c555b3f96f10ed2972708d34
notice: /Stage[main]//Node[default]/File[/tmp/temp1.txt]/content: content changed ‘{md5}cab0e6a2c555b3f96f10ed2972708d34‘ to ‘unknown checksum‘
notice: Finished catalog run in 0.11 seconds
puppet集群