nginx+tomcat

nginx + tomcat 集群实现负载均衡
2012-09-21 14:15 1338人阅读 评论(1) 收藏 举报
nginxtomcat集群server负载均衡file
1、安装nginx
     所需的prce库
     wget ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/pcre-8.10.tar.gz
     
    tar zxvf pcre-8.10.tar.gz
    cd pcre-8.10/
    ./configure
    make && make install
    cd ../

   安装nginx
    wget  http://nginx.org/download/nginx-0.8.54.tar.gz  
    wget  http://nginx.org/download/nginx-1.0.2.tar.gz

   tar zxvf nginx-1.0.2.tar.gz
   cd nginx-1.0.2/
   ./configure --user=www --group=www --prefix=/usr/local/nginx --with-http_stub_status_module --with-http_ssl_module
   make && make install
   cd ../ 
创建Nginx日志目录 
mkdir -p /home/nginx/logs
chmod +w /home/nginx/logs
chown -R www:www /home/nginx/logs 
创建Nginx配置文件 
①、在/usr/local/nginx/conf/目录中创建nginx.conf文件:
       rm -f /usr/local/nginx/conf/nginx.conf
       vi /usr/local/nginx/conf/nginx.conf 
输入以下内容: 
user www www;
worker_processes 8;
error_log /home/nginx/logs/nginx_error.log  crit;
pid /usr/local/nginx/nginx.pid;
worker_rlimit_nofile 65535;


events
{
    use epoll;
    worker_connections 65535;
}

http
{
    include mime.types;
    default_type application/octet-stream;
    #charset gb2312;

    server_names_hash_bucket_size 128;
    client_header_buffer_size 32k;
    large_client_header_buffers 4 32k;
    client_max_body_size 8m;
    sendfile on;
    tcp_nopush on;

    keepalive_timeout 60;

    tcp_nodelay on;

    fastcgi_connect_timeout 300;
    fastcgi_send_timeout 300;
    fastcgi_read_timeout 300;
    fastcgi_buffer_size 64k;
    fastcgi_buffers 4 64k;
    fastcgi_busy_buffers_size 128k;
    fastcgi_temp_file_write_size 128k;

    gzip on;
    gzip_min_length 1k;
    gzip_buffers 4 16k;
    gzip_http_version 1.0;
    gzip_comp_level 2;
    gzip_types text/plain application/x-javascript text/css application/xml;
    gzip_vary on;

    #limit_zone crawler$binary_remote_addr10m;

  #测试
 server
  {
   listen      80;
    server_name  www.test.com;
    index index.htm index.html index.jsp;
   root  /home/htdocs/web/ROOT/;

    location ~ .*$
    {
      index index.jsp;
      proxy_pass  http://www.test.com:8080;
    }
    access_log logs/sp.imichat.com.log combined;
    error_page  404 = /404.html;
  }


}

 ②、在/usr/local/nginx/conf/目录中创建fcgi.conf文件:
            vi /usr/local/nginx/conf/fcgi.conf 
    输入以下内容: 
    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;


# PHP only, required if PHP was built with --enable-force-cgi-redirect
fastcgi_param  REDIRECT_STATUS    200;
启动Nginx 
ulimit -SHn 65535
/usr/local/nginx/sbin/nginx 
配置开机自动启动Nginx 
    vi /etc/rc.local 
在末尾增加以下内容: 
ulimit -SHn 65535 
/usr/local/nginx/sbin/nginx 
优化Linux内核参数 
vi /etc/sysctl.conf 
在末尾增加以下内容: 
    # Add
net.ipv4.tcp_max_syn_backlog = 65536
net.core.netdev_max_backlog =  32768
net.core.somaxconn = 32768


net.core.wmem_default = 8388608
net.core.rmem_default = 8388608
net.core.rmem_max = 16777216
net.core.wmem_max = 16777216


net.ipv4.tcp_timestamps = 0
net.ipv4.tcp_synack_retries = 2
net.ipv4.tcp_syn_retries = 2


net.ipv4.tcp_tw_recycle = 1
#net.ipv4.tcp_tw_len = 1
net.ipv4.tcp_tw_reuse = 1


net.ipv4.tcp_mem = 94500000 915000000 927000000
net.ipv4.tcp_max_orphans = 3276800


#net.ipv4.tcp_fin_timeout = 30
#net.ipv4.tcp_keepalive_time = 120
net.ipv4.ip_local_port_range = 1024  65535

使配置立即生效: 
   /sbin/sysctl -p 

在不停止Nginx服务的情况下变更Nginx配置 
修改/usr/local/nginx/conf/nginx.conf配置文件后,请执行以下命令检查配置文件是否正确:
/usr/local/nginx/sbin/nginx -t 

如果屏幕显示以下两行信息,说明配置文件正确:
  the configuration file /usr/local/nginx/conf/nginx.conf syntax is ok
  the configuration file /usr/local/nginx/conf/nginx.conf was tested successfully 
重启nginx: 
      /usr/local/nginx/sbin/nginx -s reload
编写每天定时切割Nginx日志的脚本 
       创建脚本/usr/local/nginx/sbin/cut_nginx_log.sh
            vi /usr/local/nginx/sbin/cut_nginx_log.sh
输入以下内容: 
#!/bin/bash
# This script run at 00:00
# The Nginx logs path
logs_root_path="/home/nginx/logs/"
logs_path=${logs_root_path}$(date -d "yesterday" +"%Y")/$(date -d "yesterday" +"%m")/
mkdir -p ${logs_path}
#日志文件名 ====================================================
logs_name="weblogs"
logs_file=${logs_root_path}${logs_name}.log
cut_logs_file=${logs_path}${logs_name}_$(date -d "yesterday" +"%Y%m%d").log
mv ${logs_file} ${cut_logs_file}
tar czf ${cut_logs_file}.tar.gz ${cut_logs_file}
rm -f ${cut_logs_file}
#重启nginx
/usr/local/nginx/sbin/nginx -s reload

对这个脚本赋执行权限
# chmod +x /usr/local/nginx/sbin/cut_nginx_log.sh 
设置crontab,每天凌晨00:00切割nginx访问日志 
crontab -e 
输入以下内容: 
    00 00 * * * /bin/bash  /usr/local/nginx/sbin/cut_nginx_log.sh 

2、配置nginx负载均衡
    编辑nginx.conf
        vim /usr/local/nginx/conf/nginx.conf
    在server上面增加
     upstream   www.test.com {
              server   www.test.com:8888;
              server   www.test.com:8080;
        #ip_hash;   注:如果加入了ip_hash, 这种只能适用于当某个用户连接上了一台服务器后,他登陆之后所做的一切操作都只会在那一台服务器不会跳转到另外的服务器,如果那台服务器over掉了,则会自动退出,连接到另一服务器
    }
www.test.com 是自己的服务器,最好是用域名
upstream后面的名字要与proxy_pass 这个名字一样,建议upstream后面的名字,proxy_pass 后面的名字,server_name 后面的名字一致
示例:
  upstream   www.test.com {
              server   www.test.com:8888;
              server   www.test.com:8080;
    }

  #测试
 server
  {
   listen      80;
    server_name  www.test.com;
    index index.htm index.html index.jsp;
   root  /home/htdocs/web/ROOT/;

    location ~ .*$
    {
      index index.jsp;
      proxy_pass  http://www.test.com; 注:做了负载均衡,这个后面就不能再加端口
    }
    access_log logs/sp.imichat.com.log combined;
    error_page  404 = /404.html;
  }


注:如果负载均衡不成功,请检查是否没有经过nginx,80端口直接被转发到了tomcat服务的端口上

3、配置tomcat集群
    优化tomcat 最大并发数,编辑server.xml
             <Connector port="8080" maxHttpHeaderSize="8192"

               maxThreads="2048" minSpareThreads="100" maxSpareThreads="200"

               enableLookups="false" redirectPort="8443" acceptCount="500"

               connectionTimeout="20000" disableUploadTimeout="true" />


            <Engine name="Catalina" defaultHost="localhost" jvmRoute="worker1">

        <Cluster className="org.apache.catalina.ha.tcp.SimpleTcpCluster" tcpListenAddress="127.0.0.1" />
    修改web.xml
        增加标签:<distributable/>
            直接加在</web-app> 之前就可以了,这个是加入tomcat 的session 复制的,做tomcat 集群必须需要这一步,否则用户的session 就无法正常使用.

你可能感兴趣的:(nginx+tomcat)