百万PV架构搭建详解

百万PV架构

●先了解一下什么是PV
PV（page view）即页面浏览量，通常是衡量一个网络新闻频道或网站甚至一条网络新闻的主要指标。网页浏览数是评价网站流量最常用的指标之一，简称为PV。监测网站PV的变化趋势和分析其变化原因是很多站长定期要做的工作。 Page Views中的Page一般是指普通的html网页，也包含php、jsp等动态产生的html内容。来自浏览器的一次html内容请求会被看作一个PV，逐渐累计成为PV总数。

●环境及组件介绍

操作系统	IP地址	角色	web组件
Centos7	192.168.70 .136	主服务器	keepalived、nginx、mysql、redis
Centos7	192.168.70.137	从服务器	keepalived、nginx、mysql、redis
Centos7	192.168.70 .134	web后端1	tomcat、项目
Centos7	192.168.70 .132	web后端2	tomcat、项目

●使用的软件包
nginx1.8.1：http://101.96.10.46/nginx.org/download/nginx-1.8.1.tar.gz
web服务包：https://pan.baidu.com/s/143ZRkqfUxJJIBzO_yz7gPg
密码：wsgd
mysql解压版https://pan.baidu.com/s/11b_ccrosT0IPdnXhRrU4yQ
密码：ruh5

一、主从服务器配置keepalived

1、安装

[root@localhost ~]# yum install keepalived -y

2、修改keepalive配置文件

[root@localhost ~]# vim /etc/keepalived/keepalived.conf

! Configuration File for keepalived

global_defs {
     notification_email {
         [email protected]
         [email protected]
         [email protected]
     }
     notification_email_from [email protected]
     smtp_server 192.168.70.131      #指向本机
     smtp_connect_timeout 30
     router_id NGINX_01                   #备机为NGINX_02
}

vrrp_instance VI_1 {
        state MASTER                        #备机为BACKUP
        interface ens33                       #网卡名称
        virtual_router_id 51                #备机52，不与主相同
        priority 100                               #优先级，备机优先级要低于主机
        advert_int 1
        authentication {
                auth_type PASS
                auth_pass 1111
        }
        virtual_ipaddress {
                192.168.70.100                #虚拟Ip地址
        }
}

==========以下是配置keepalive的顺带着启动nginx===========
个人看法，不太希望这样做，因为每两秒尝试启动nginx，nginx会两秒打一次error日志，提示端口被占用无法启动，这样反而造成了不必要的磁盘读写。

 Configuration File for keepalived

#定义NGINX启动脚本位置，每两秒检查一次
vrrp_script nginx {
                script "/opt/shell/nginx.sh"
                interval 2
}

#删除原本内容，添加route_id，备机为NGINX_02
global_defs {
                route_id NGINX_01
}

vrrp_instance VI_1 {
        state MASTER    #主机为MASTER
        interface ens33 #网卡名称
        virtual_router_id 51    #备机为52
        priority 100             #优先级，备机为99
        advert_int 1
        authentication {
                auth_type PASS
                auth_pass 1111
        }
#触发脚本
track_script {
                nginx
}

#虚拟IP
        virtual_ipaddress {
                192.168.70.100
        }
}

3、启动keepalived

[root@localhost ~]# systemctl start keepalived.service

4、查看虚拟ip

[root@localhost ~]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:5b:aa:ce brd ff:ff:ff:ff:ff:ff
    inet 192.168.70.131/24 brd 192.168.70.255 scope global dynamic ens33
       valid_lft 1403sec preferred_lft 1403sec
    inet 192.168.70.100/32 scope global ens33         #虚拟ip
       valid_lft forever preferred_lft forever
    inet6 fe80::49c4:1329:39cd:4427/64 scope link 
       valid_lft forever preferred_lft forever

注：备机也要启动keepalived，然后关闭主机keepalived，看虚拟Ip会不会漂移到备机。

二、安装nginx

上传nginx安装包至/opt目录下
1、安装环境包

[root@localhost opt]# yum -y install gcc gcc-c++ autoconf gd-devel automake zlib zlib-devel openssl openssl-devel pcre*

2、解压，编译nginx

[root@localhost opt]# tar zxf nginx-1.8.1.tar.gz
[root@localhost opt]# cd nginx-1.8.1/
[root@localhost nginx-1.8.1]# ./configure \
--prefix=/usr/local/nginx \
--user=nginx \
--group=nginx \
--with-http_ssl_module \
--with-http_gzip_static_module \
--with-http_image_filter_module \
--with-http_stub_status_modulev

3、安装

[root@localhost nginx-1.8.1]# make && make install

4、创建nginx用户

[root@localhost nginx-1.8.1]# useradd -M -s /sbin/nologin nginx

5、优化命令路径

[root@localhost nginx-1.8.1]# ln -s /usr/local/nginx/sbin/nginx /usr/sbin/

6、添加代理并简单优化

[root@localhost nginx-1.8.1]# cd /usr/local/nginx/conf/
[root@localhost conf]# vim nginx.conf

#用户nginx,单核
user  nginx;
worker_processes  1;

#每个核心连接数2048
events {
    worker_connections  2048;
}

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

#隐藏版本号
    server_tokens off;

#sendfile参数用于开启文件的高效传输模式。
    sendfile   on;
    tcp_nopush on;
    tcp_nodelay on;
    server_names_hash_bucket_size 128;
    server_names_hash_max_size 512;
    client_header_timeout 15s;
    client_body_timeout 15s;
    send_timeout 60s;

   keepalive_timeout  65;
#压缩模块
    gzip  on;
    gzip_buffers  4 64k;
    gzip_http_version 1.1;
    gzip_comp_level 2;
    gzip_min_length 1k;
    gzip_vary on;
    gzip_types text/plain text/javascript application/x-javascript text/css text/xml application/xml applicatin/xml+rss;

#反向代理
    upstream tomcat_pool {
                server 192.168.70.134:8080;
                server 192.168.70.132:8080;
                ip_hash;        #会话稳固，防止停留页面过久导致需要重新登录。
        }
        server {
                listen 80;
                server_name 192.168.70.100;     #虚拟Ip
                location / {
                        proxy_pass http://tomcat_pool;
                        proxy_set_header X-Real-IP $remote_addr;
                        expires 1d;
                }
        }

    server {
        listen       80;
        server_name  localhost;

        location / {
            root   html;
            index  index.html index.htm;
        }
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   html;
        }

    }

}

7、检查配置文件语法

[root@localhost conf]# nginx -t

nginx: the configuration file /usr/local/nginx/conf/nginx.conf syntax is ok
nginx: configuration file /usr/local/nginx/conf/nginx.conf test is successful

8、启动nginx

[root@localhost conf]# nginx

三、安装tomcat

上传jdk和tomcat到/opt目录下
1、解压jdk和tomcat

[root@localhost opt]# tar zxf jdk-8u144-linux-x64.tar.gz
[root@localhost opt]# tar zxf apache-tomcat-8.5.23.tar.gz

2、更名jdk为java

[root@localhost opt]# mv jdk1.8.0_144/ java

3、添加环境变量

[root@localhost opt]# vim ~/.bashrc

#末行添加如下三行
export JAVA_HOME=/opt/java
export CLASSPATH=$CLASSPATH:$JAVA_HOME/lib:$JAVA_HOME/jre/lib
export PATH=$JAVA_HOME/bin:$JAVA_HOME/jre/bin:$PATH:$HOME/bin

4、刷新环境变量，查看java版本

[root@localhost opt]# source ~/.bashrc
[root@localhost opt]# java -version

java version "1.8.0_144"
Java(TM) SE Runtime Environment (build 1.8.0_144-b01)
Java HotSpot(TM) 64-Bit Server VM (build 25.144-b01, mixed mode)

5、更名tomcat为tomcat8

[root@localhost opt]# mv apache-tomcat-8.5.23 tomcat8

6、修改默认首页进行测试，两台tomcat都需要操作

[root@localhost opt]# cd tomcat8/webapps/ROOT/
[root@localhost ROOT]# vim index.jsp

#删除原有的内容，添加如下本机ip，用于测试
#134主机添加如下
<h1> this is 134 server <h1>

#132主机添加如下
<h1> this is 132 server <h1>

7、建立命令软链接，启动tomcat

[root@localhost ROOT]# ln -s /opt/tomcat8/bin/startup.sh /usr/bin/tomcatup
[root@localhost ROOT]# ln -s /opt/tomcat8/bin/shutdown.sh /usr/bin/tomcatdown
[root@localhost ROOT]# tomcatup #启动tomcat

Using CATALINA_BASE:   /opt/tomcat8
Using CATALINA_HOME:   /opt/tomcat8
Using CATALINA_TMPDIR: /opt/tomcat8/temp
Using JRE_HOME:        /opt/java
Using CLASSPATH:       /opt/tomcat8/bin/bootstrap.jar:/opt/tomcat8/bin/tomcat-juli.jar
Tomcat started.

8、web网页用虚拟ip访问

192.168.70.100
显示后台我们设置的tomcat页面即可，然后关闭一台tomcat看页面会不会变

四、mysql主从复制

上传mysql压缩包到/opt目录下
1、解压mysql

[root@localhost opt]# tar zxf mysql-5.7.21-linux-glibc2.12-x86_64.tgz
[root@localhost opt]# mv mysql-5.7.21-linux-glibc2.12-x86_64 mysql

2、修改配置文件

[root@localhost opt]# vim /etc/my.cnf

[client]
port = 3306
socket = /tmp/mysql.sock

[mysqld]
character-set-server = utf8mb4
skip_name_resolve = 1
user = mysql
port = 3306
server-id = 1           #注意，备机的mysql，server-id不能与主机相同
socket = /tmp/mysql.sock
basedir = /opt/mysql/
datadir = /opt/mysql/data
pid-file = /opt/mysql/data/mysql.pid
log_bin=/opt/mysql/data/mysql-bin
log-error = /opt/mysql/data/log-error.log
innodb_data_home_dir = /opt/mysql/data
slow-query-log-file=/opt/mysql/data/slow.log
relay-log-index = /opt/mysql/data/relaylog
relay-log-info-file = /opt/mysql/data/relaylog
relay-log = /opt/mysql/data/relaylog
open_files_limit = 10240
table_open_cache = 2048
back_log = 300
max_connections = 10000
max_connect_errors = 20
explicit_defaults_for_timestamp = 1
max_allowed_packet = 64M
thread_cache_size = 300
query_cache_size = 256M
query_cache_limit = 2M
query_cache_min_res_unit = 2k
default-storage-engine = InnoDB
thread_stack = 512K
transaction_isolation = READ-COMMITTED
tmp_table_size = 256M
max_heap_table_size = 256M
key_buffer_size = 2G
sort_buffer_size = 2M
join_buffer_size = 6M
read_buffer_size = 4M
read_rnd_buffer_size = 16M
bulk_insert_buffer_size = 64M
myisam_sort_buffer_size = 128M
myisam_max_sort_file_size = 15G
myisam_repair_threads = 1
interactive_timeout = 1800
wait_timeout = 28800
innodb_data_file_path = ibdata1:120M;ibdata2:200M;ibdata3:200M:autoextend
innodb_buffer_pool_size = 1G
innodb_thread_concurrency = 0
innodb_flush_log_at_trx_commit = 2
innodb_log_buffer_size = 16M
innodb_log_file_size = 512M
innodb_log_files_in_group = 3
innodb_max_dirty_pages_pct = 90
innodb_lock_wait_timeout = 120
innodb_purge_threads = 0
slow_query_log = 1
long_query_time = 3
replicate-ignore-db = mysql
replicate-ignore-db = test
replicate-ignore-db = information_schema
#slave-skip-errors = 1032,1062,1026,1114,1146,1048,1396

sql_mode=STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION
[mysqldump]
quick
max_allowed_packet = 64M

3、初始化mysql，设置密码

[root@localhost opt]# ln -s /opt/mysql/bin/* /usr/bin/
[root@localhost opt]# mysql/bin/mysqld --initialize-insecure #初始化
[root@localhost opt]# mysqld_safe & #启动
[root@localhost opt]# mysqladmin -uroot password #设置密码
[root@localhost opt]# mysql -uroot -p #登陆查看

4、设置mysql主从复制

#以下在主库上操作
mysql> grant replication slave on *.* to 'replication'@'192.168.70.%' identified by '123123' with grant  option;     #授权replication用户权限

mysql> flush privileges;           #刷新权限

mysql> show master status;   #查看二进制日志位置
+------------------+----------+--------------+------------------+-------------------+
| File             | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |
+------------------+----------+--------------+------------------+-------------------+
| mysql-bin.000002 |      885 |              |                  |                   |
+------------------+----------+--------------+------------------+-------------------+

#以下在从库上操作
mysql>  change master to master_host='192.168.70.136',master_user='replication',master_password='123123',,master_log_file='mysql-bin.000002',master_log_pos=885;

mysql> start slave;

mysql> show slave status\G;
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: 192.168.70.136
                                    ..........................
            Slave_IO_Running: Yes          #要看到这两项为yes，主从复制成
            Slave_SQL_Running: Yes

5、主库导入sql语句

[root@localhost opt]# mysql -uroot -p < slsaledb-2014-4-10.sql

6、项目的mysql用户授权

mysql> grant all on . to 'root'@'%' identified by '123123';
mysql> flush privileges;

五、redis缓存服务器配置

1、下载epel源和redis

[root@localhost opt]# yum install -y epel-release
[root@localhost opt]# yum install redis -y

2、修改配置文件

[root@localhost opt]# vim /etc/redis.conf

主从服务器
  61行 bind 0.0.0.0
从服务器添加
266行 slaveof 192.169.70.136 6379

3、主从服务器启动redis

[root@localhost opt]# systemctl start redis

4、验证主从功能

[root@localhost opt]# redis-cli -h 192.168.70.136 -p 6379

#主服务器写入
192.168.70.136:6379> set name lisi
OK
192.168.70.136:6379> get name 
"lisi"
#从服务器
192.168.70.137:6379> get name     #能获取值说明主从同步成功
"lisi"

5、搭建主从切换
注：主服务器上操作

[root@localhost opt]# vim /etc/redis-sentinel.conf

protected-mode no#是否开启保护模式
sentinel monitor mymaster 192.168.70.137 6379 1 #1表示1台从
sentinel down-after-milliseconds mymaster 3000  #切换时间为3000毫秒

#主从都启动群集
[root@localhost opt]# systemctl start redis-sentinel.service

#查看群集状态

[root@localhost opt]# redis-cli -h 192.168.70.136 -p 26379 info Sentinel

# Sentinel
sentinel_masters:1
sentinel_tilt:0
sentinel_running_scripts:0
sentinel_scripts_queue_length:0
sentinel_simulate_failure_flags:0   #可以看到下面行有从137
master0:name=mymaster,status=odown,address=192.168.70.137:6379,slaves=0,sentinels=2

六、部署项目

#两台tomcat主机上操作
1、上传项目包，解压缩到指定目录

[root@localhost opt]# tar zxf SLSaleSystem.tar.gz -C /opt/tomcat8/webapps/

2、修改tomcat配置文件，确定项目位置

[root@localhost opt]# cd /opt/tomcat8/conf/
[root@localhost conf]# vim server.xml

      <Host name="localhost"  appBase="webapps"
            unpackWARs="true" autoDeploy="true">
        #接近尾行位置添加如下
<Context path="" docBase="SLSaleSystem" reloadable="true" debug="0"></Context>

3、修改项目文件，指定连接数据库和redis缓存服务

[root@localhost conf]# cd /opt/tomcat8/webapps/SLSaleSystem/WEB-INF/classes/

#修改为虚拟ip
url=jdbc\:mysql\://192.168.70.100\:3306/slsaledb?useUnicode\=true&characterEncoding\=UTF-8
#数据库授权的用户
uname=root
#密码
password=123123

#修改redis缓存服务器ip地址

[root@localhost classes]# vim applicationContext-mybatis.xml

    <!--redis 配置 开始-->
        <bean id="jedisPoolConfig" class="redis.clients.jedis.JedisPoolConfig">
            <property  name="maxActive"  value="90" />
            <property  name="maxIdle"   value="5" />
            <property  name="maxWait"  value="1000" />
            <property  name="testOnBorrow"  value="true" />
        </bean>
        <bean id="jedisPool" class="redis.clients.jedis.JedisPool" destroy-method="destroy" >
            <constructor-arg ref="jedisPoolConfig"/>
            <constructor-arg value="192.168.70.100"/>      #虚拟ip
            <constructor-arg value="6379"/>
        </bean>
        <bean id="redisAPI" class="org.slsale.common.RedisAPI">
            <property name="jedisPool" ref="jedisPool"/>
        </bean>

4、配置完重新启动tomcat

[root@localhost classes]# tomcatdown
[root@localhost classes]# tomcatup

5、网页中访问
192.168.70.100
#默认账号admin
#默认密码123456

6、查看redis缓存服务

[root@localhost opt]# redis-cli -h 192.168.70.136 -p 6379

192.168.70.136:6379> info
keyspace_hits:2    #找到这两行，这个是命中次数
keyspace_misses:0

注：点击进网页内容出现感叹属于正常，缓存服务器需要一段时间缓存内容。
redis缓存服务器主从切换的问题，主宕机，从会顶替上来，但再次启动主的时候，从不会让出当前master的角色，除非手动切换，或从宕机。
这个项目是很久很久的一个项目，拿来学习理解下百万pv架构还行，学习为主，另外想要达到百万pv，硬件性能是必不可少的，软件层面上的优化也是必须的。
本文主要介绍百万pv的架构及其中一些服务之间的关系，谨以介绍学习，实际操作以自己公司环境为主。