7、nginx的upstream及fastcgi模块应用

ngx_http_proxy_module, ngx_http_upstream_module
 
ngx_http_proxy_module：实现反向代理及缓存功能
proxy_pass http://{SERVER_IP|UPSTREAM_NAME}/uri
 
proxy_cache_path path [levels=levels] keys_zone=name:size [inactive=time] [max_size=size];
proxy_cache zone_name;          //缓存名称
proxy_cache_valid [code] time;  //指明不同响应码内容的缓存时长
proxy_cache_method              //只对哪些方法获取的资源进行缓存，一般使用get
proxy_cache_use_stale error timeout ...   //是否使用过期的缓存进行响应
proxy_cache_min_uses             //某资源至少响应请求多少次后才被缓存下来
 
proxy_cache_bypass string: 设置在何种情形下nginx将不从cache取数据的；比如邮件
例如：$cookie_nocache、$arg_nocache 、$http_authorization
proxy_set_header
ngx_http_upstream_module:定义服务器组    http://nginx.org/en/docs/http/ngx_http_upstream_module.html
可以调用服务器组反向代理有：proxy_pass, fastcgi_pass, uwsgi_pass, 
upstream name {
server address [parameters];
ip_hash;    //基于源地址绑定，比ip_hash更精确的是session绑定机制是cookie记录，但LVS是没办法识别cookie的，因为cookie是应用层数据
}

nginx(2)

1、SNAT模式的大量Client

基于sticky实现session绑定：对应于工作在应用层的程序才拥有的功能

(1)cookie：基于cookie的绑定

upstream backend {
    server backend1.example.com;
    server backend2.example.com;

    sticky cookie srv_id expires=1h domain=.example.com path=/;
}

(2)route：基于路由的绑定

map $cookie_jsessionid $route_cookie {
    ~.+\.(?P<route>\w+)$ $route;
}

map $request_uri $route_uri {
    ~jsessionid=.+\.(?P<route>\w+)$ $route;
}

upstream backend {
    server backend1.example.com route=a;
    server backend2.example.com route=b;

    sticky route $route_cookie $route_uri;
}

(3)learn () ：基于学习的绑定

upstream backend {
   server backend1.example.com:8080;
   server backend2.example.com:8081;

   sticky learn
          create=$upstream_cookie_examplecookie
          lookup=$cookie_examplecookie
          zone=client_sessions:1m;
}

2、least_conn: 调度方法，最少连接(相当于LVS中的wlc算法)；

3、keeplive: 激活代理服务器nginx和后端服务器之间的持久连接功能的，专用在upstream，

Syntax:	keepalive connections;
Default:	—
Context:	upstream

例子：keepalive一般用在上游服务器是非http服务器时才使用，因为keepalive会损坏http的并发性能，如果后端是专用存储时(比如基于键值存储的memcache)，基于keepalive指令可以使链接处于一段时间的长连接，提高查询性能。

upstream memcached_backend {
    server 127.0.0.1:11211;
    server 10.0.0.2:11211;

    keepalive 32;
}

health_check:只能同在location
使用health_check时建议关闭访问日志，因为后端服务器不能检测health_check的检测是正常请求还是只是用来检测的

自定义响应首部：

add_header X-Via $server_addr;  自定义响应首部通知客户端是由谁响应的

add_header X-Cache $upstream_cache_status;   告诉客户端响应是否命中

首先编辑代理服务器的server配置文件

# vim /etc/nginx/conf.d/default.conf

server {
    listen       ;
    server_name  localhost;
    add_header X-Via $server_addr;
    add_header X-Cache $upstream_cache_status;

    #charset koi8-r;
    #access_log  /var/log/nginx/host.access.log  main;

    location / {
        root   /usr/share/nginx/html;
        index  index.html index.htm;
    }   

    location /forum/ {
        proxy_cache mycache;
        proxy_cache_valid  1d;
        proxy_cache_valid   10m;
        proxy_cache_valid any 1m;
        proxy_cache_use_stale error timeout http_500 http_502 http_503 http_504;
        proxy_pass http://upservers/;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        add_header X-Via $server_addr;    //启用添加自定义头部
        add_header X-Cache $upstream_cache_status;
    }   

Module ngx_http_fastcgi_module模块介绍

http://nginx.org/en/docs/http/ngx_http_fastcgi_module.html

proxy_pass http://{SERVER_IP|UPSTREAM_NAME}/uri   //proxy_pass指令是把用户的请求反向代理至上有的web服务器，所以这里的协议是http协议，如果上有的协议不是http协议而是fastcgi协议，就必须使用fastcgi模块来实现。

fastcgi模块的意义是：

nginx作为代理服务器，当客户端发起对nginx代理服务器的php类型资源的请求，比如是index.php，此时nginx代理服务器只是一个静态内容服务器，所以nginx就会把index.php的文件内容拿过来包装成响应报文后直接返回给客户端，但此时响应报文是源代码，客户端是看不懂的。

所以应该让nginx自持php应用代码的运行，但是nginx自己不支持，所以需要做一个php的服务器。此后nginx在接收到报文时就可以根据请求资源的后缀名进行判断，一旦是php，nginx就会使用特定协议将请求报文反向代理至php的server上，php服务器将会执行index.php上的代码并将执行后的结果返回值nginx代理服务器上，然后nginx就会把响应报文进行封装并返回给客户端。这又是一种反向代理，但是使用的协议是fastcgi协议而不是http协议。

如何让nginx支持LNMP：

php是无法作为nginx的模块的，所以php只能工作于fpm模式

# yum list all *fpm*   //查找是否有fpm模块

Loaded plugins: fastestmirror
Repodata is over 2 weeks old. Install yum-cron? Or run: yum makecache fast
Determining fastest mirrors
* base: mirrors.nwsuaf.edu.cn
* extras: mirrors.neusoft.edu.cn
* updates: mirrors.neusoft.edu.cn
Available Packages
php-fpm.x86_64 5.4.16-45.el7 base

# yum info php-fpm  //查看此模块的详细信息

# yum install php-fpm   //安装php-fpm模块

# rpm -ql php-fpm    //查看php-fpm安装信息

/etc/logrotate.d/php-fpm
/etc/php-fpm.conf    //主配置文件
/etc/php-fpm.d
/etc/php-fpm.d/www.conf   //这里可以查看php监听的地址，如果php和nginx不在同一主机的话，就需要修改php监听的地址了
/etc/sysconfig/php-fpm    //服务脚本
/run/php-fpm
/usr/lib/systemd/system/php-fpm.service
/usr/lib/tmpfiles.d/php-fpm.conf
/usr/sbin/php-fpm   //可执行程序
/usr/share/doc/php-fpm-5.4.16
/usr/share/doc/php-fpm-5.4.16/fpm_LICENSE
/usr/share/doc/php-fpm-5.4.16/php-fpm.conf.default
/usr/share/fpm
/usr/share/fpm/status.html
/usr/share/man/man8/php-fpm.8.gz
/var/log/php-fpm

# systemctl start php-fpm   //启动

# ss -tunl  //查看端口

下面让nginx将用户的请求基于fastcgi协议发送给php

# vim /etc/nginx/conf.d/default.conf   //编辑此文件，启用php配置段

 location ~ \.php$ {
        root           html;    //这里的页面要和上面nginx的web页面保持一致，修改如下
         root   /usr/share/nginx/html;
        fastcgi_pass   127.0.0.1:;
        fastcgi_index  index.php;
        fastcgi_param  SCRIPT_FILENAME  /scripts$fastcgi_script_name;  //向后端传递的参数
                       //SCRIPT_FILENAME是变量名指令，用户请求的文件名(即变量名$fastcgi_script_name)，
                         通过uri(~ \.php$)的方式传递给后端的服务器(127.0.0.1:9000) 

        include fastcgi_params;   //包含另外一个文件
}  

# cd /usr/share/nginx/html/   //对主页面进行修改
# vim index.php

<?php
phpinfo();
?>

 # vim /etc/nginx/conf.d/default.conf   

location / {
root /usr/share/nginx/html;
index index.php index.html index.htm;   //并在根这里添加index.php

要对/etc/nginx/fastcgi_params做修改

/etc/nginx   //在这个文件目录下有一个fastcgi_params配置文件

fastcgi_param  QUERY_STRING       $query_string;
fastcgi_param  REQUEST_METHOD     $request_method;
fastcgi_param  CONTENT_TYPE       $content_type;
fastcgi_param  CONTENT_LENGTH     $content_length;

fastcgi_param  SCRIPT_NAME        $fastcgi_script_name;
fastcgi_param  REQUEST_URI        $request_uri;
fastcgi_param  DOCUMENT_URI       $document_uri;
fastcgi_param  DOCUMENT_ROOT      $document_root;
fastcgi_param  SERVER_PROTOCOL    $server_protocol;
fastcgi_param  REQUEST_SCHEME     $scheme;
fastcgi_param  HTTPS              $https if_not_empty;

fastcgi_param  GATEWAY_INTERFACE  CGI/1.1;
fastcgi_param  SERVER_SOFTWARE    nginx/$nginx_version;

fastcgi_param  REMOTE_ADDR        $remote_addr;
fastcgi_param  REMOTE_PORT        $remote_port;
fastcgi_param  SERVER_ADDR        $server_addr;
fastcgi_param  SERVER_PORT        $server_port;
fastcgi_param  SERVER_NAME        $server_name;

# PHP only, required if PHP was built with --enable-force-cgi-redirect
fastcgi_param  REDIRECT_STATUS    ;

编辑/etc/nginx/fastcgi_params，将其内容更改为如下内容：

fastcgi_param  GATEWAY_INTERFACE  CGI/1.1;

fastcgi_param  SERVER_SOFTWARE    nginx;

fastcgi_param  QUERY_STRING       $query_string;

fastcgi_param  REQUEST_METHOD     $request_method;

fastcgi_param  CONTENT_TYPE       $content_type;

fastcgi_param  CONTENT_LENGTH     $content_length;

fastcgi_param  SCRIPT_FILENAME    $document_root$fastcgi_script_name;

fastcgi_param  SCRIPT_NAME        $fastcgi_script_name;

fastcgi_param  REQUEST_URI        $request_uri;

fastcgi_param  DOCUMENT_URI       $document_uri;

fastcgi_param  DOCUMENT_ROOT      $document_root;

fastcgi_param  SERVER_PROTOCOL    $server_protocol;

fastcgi_param  REMOTE_ADDR        $remote_addr;

fastcgi_param  REMOTE_PORT        $remote_port;

fastcgi_param  SERVER_ADDR        $server_addr;

fastcgi_param  SERVER_PORT        $server_port;

fastcgi_param  SERVER_NAME        $server_name;

此时nginx已经可以和php结合工作了，下面就要解决php和mysql/mariadb结合

# yum install php-mysql

# rpm -ql php-mysql

/etc/php.d/mysql.ini    
/etc/php.d/mysqli.ini
/etc/php.d/pdo_mysql.ini
/usr/lib64/php/modules/mysql.so
/usr/lib64/php/modules/mysqli.so
/usr/lib64/php/modules/pdo_mysql.so

# systemctl reload php-fpm

# yum install mariadb-server   //安装mariadb

安装好mariadb后，验证php和mariadb是否可以结合工作

# vim /usr/share/nginx/html/index.php

<?php
    $conn = mysql_connect('127.0.0.1','root','');
    if ($conn)
        echo succ；
    else
       echo fail；

    mysql_close();
?>

# systemctl stop mariadb   //关闭mariadb后在测试

LNAMP ：以nginx做代理服务器进行负载均衡，后端的服务器平台是多个LAMP，在这种场景中nginx和后端的LAMP交互时是两个web服务器之间的交互(http协议)，LAMP中的php可以以                       apache的模块进行工作。

动态内容和静态内容分离：php服务器和MySQL服务器与后端负载均衡的服务器不在同一个服务器上

LNMP, fastcgi_cache   fastcgi协议的缓存服务器

Syntax:	fastcgi_cache_path path [levels=levels] [use_temp_path=on|off]keys_zone=name:size [inactive=time] [max_size=size] [manager_files=number]                   [manager_sleep=time] [manager_threshold=time][loader_files=number] [loader_sleep=time] [loader_threshold=time] [purger=on|off] [purger_files=number] [purger_sleep=time] [purger_threshold=time];
Default:	—
Context:	http   //需要定义在http上下文

fastcgi_cache_path /data/nginx/cache levels=1:2 keys_zone=one:10m;
以上是定义位置以及定义方法，定义完成后需要使用fastvgi_cache进行调用

Syntax:	fastcgi_cache zone | off;
Default:	fastcgi_cache off;
Context:	http, server, location   使用上下文

示例：LNMP通过fastcgi_cahce加速响应

# vim /etc/nginx/nginx.conf //首先编辑nginx的主配置文件http上下文

fastcgi_cache_path /cache/fastcgi/ levels=1:1 keys_zone=fcgicache:10m inactive=3m max_size=1g;

# vim /etc/nginx/conf.d/default.conf   //在对应的location中进行调用

   location ~ \.php$ {    //对应的php的location中进行调用
        fastcgi_cache fcgicache;
        root           /usr/share/nginx/html;
        fastcgi_pass   127.0.0.1:;
        fastcgi_index  index.php;
        fastcgi_param  SCRIPT_FILENAME  /scripts$fastcgi_script_name;
        include        fastcgi_params;
    }   

# systemctl reload nginx
# mkdir /cache/fastcgi -pv

# ll /cache
    total 0
    drwx------ 2 nginx root 6 Nov 29 05:02 fastcgi
    drwxr-xr-x 4 nginx nginx 24 Nov 28 23:41 nginx

#vim /usr/share/nginx/html/test.php

<?php
  phpinfo();
?>

 
再次刷新时发现X-Cache仍然是MISS，即无法缓存
#vim /etc/nginx/conf.d/default.conf   //在php配置段中指明缓存响应码，如果不指明，就不缓存

fastcgi_cache_valid 200 10m;
fastcgi_cache_valid 302 3m;
fastcgi_cache_valid any 1m;

下面利用ab命令对开启缓存和不开启缓存进行性能对比

1、首先关闭fastcgi_cache

# /etc/nginx/conf.d/default.conf

#fastcgi_cache fcgicache;
fastcgi_cache off;   //暂时关闭fastcgi_cache

# systemctl reload nginx

# yum install httpd-tools

# ab -n 10000 -c 100 http://192.168.184.142/test.php   -n是请求数量，-c是并发量

This is ApacheBench, Version 2.3 <$Revision:  $>
Copyright  Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking 192.168.184.142 (be patient)
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Finished  requests

Server Software:        nginx/1.14.
Server Hostname:        192.168.184.142
Server Port:            

Document Path:          /test.php
Document Length:         bytes

Concurrency Level:
Time taken for tests:   4.891 seconds
Complete requests:
Failed requests:
   (Connect: , Receive: , Length: , Exceptions: )
Write errors:
Total transferred:       bytes
HTML transferred:        bytes
Requests per second:    2044.69 [#/sec] (mean)
Time per request:       48.907 [ms] (mean)
Time per request:       0.489 [ms] (mean, across all concurrent requests)
Transfer rate:          86863.18 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:               0.5
Processing:          13.7
Waiting:             13.6
Total:               13.6          

Percentage of the requests served within a certain time (ms)
  %
  %
  %
  %
  %
  %
  %
  %
 %     (longest request)

2、开启fastcgi_cache

# vim /etc/nginx/conf.d/default.conf

fastcgi_cache fcgicache;
#fastcgi_cache off;

#systemctl reload nginx

# ab -n 10000 -c 100 http://192.168.184.142/test.php

This is ApacheBench, Version 2.3 <$Revision:  $>
Copyright  Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking 192.168.184.142 (be patient)
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Completed  requests
Finished  requests

Server Software:        nginx/1.14.
Server Hostname:        192.168.184.142
Server Port:            

Document Path:          /test.php
Document Length:         bytes

Concurrency Level:
Time taken for tests:   0.527 seconds
Complete requests:
Failed requests:
Write errors:
Total transferred:       bytes
HTML transferred:        bytes
Requests per second:    18986.36 [#/sec] (mean)   //与上面的对比差距是非常大的
Time per request:       5.267 [ms] (mean)
Time per request:       0.053 [ms] (mean, across all concurrent requests)
Transfer rate:          806845.98 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:               0.3
Processing:            0.9
Waiting:               0.9
Total:                 0.9             

Percentage of the requests served within a certain time (ms)
  %
  %
  %
  %
  %
  %
  %
  %
 %       (longest request)

练习：

(1) root为同一路径；

(2) root为不同的路径；

location \.php$ {

root /web/app/wp;

}

location / {

root /web/htdocs;

}

如何解决root路径不同的问题？

(3) fpm server为另一主机；

location \.php$ {

fastcgi_pass fastcgi://192.168.184.142:9000;

}

location / {

root /web/htdocs;

}

总结：

cache:

proxy_cache

fastcgi_cache