实验系统:CentOS 6.6_x86_64
实验前提:
1)提前准备好编译环境,防火墙和selinux都关闭;
2)本配置共有两个测试节点,分别coro1和coro2,对应的IP地址分别为192.168.19.55和192.168.19.98;
3)集群服务为nginx服务;
实验软件:corosync-0:1.4.7-1.el6 pacemaker-0:1.1.12-4.el6 crmsh-2.1-1.6 drbd-8.4.4 php-5.6.11 mariadb-10.0.20 tengine-2.1.0
软件下载:http://pan.baidu.com/s/116gQq
注意事项:单主机操作会有标注,除此之外的是两台主机都要完成的
实验拓扑:
一、准备工作
1.所有节点的主机名称和对应的IP地址解析服务可以正常工作,且每个节点的主机名称需要跟“uname -n”命令的结果保持一致。因此,需要保证两个节点上的/etc/hosts文件均添加了下面的内容:
2.为了使得重新启动系统后仍能保持如上的主机名称,还分别需要在各节点执行类似如下的命令:
coro1上:
1 sed -i 's@\(HOSTNAME=\).*@\1coro1@g' /etc/sysconfig/network 2 hostname coro1
coro2上:
1 sed -i 's@\(HOSTNAME=\).*@\1coro2@g' /etc/sysconfig/network 2 hostname coro2
3.设定两个节点可以基于密钥进行ssh通信,并测试:
coro1上:
1 ssh-keygen -t rsa 2 ssh-copy-id -i ~/.ssh/id_rsa.pub root@coro2
coro2上:
ssh-keygen -t rsa ssh-copy-id -i ~/.ssh/id_rsa.pub root@coro1
4.两个节点时间必须同步:一般使用ntpdate进行操作,这里不再演示。
二、安装corosync和pacemaker并配置
1.安装epel源:
1 yum -y install epel*
2.安装corosync和pacemaker:
1 yum -y install corosync pacemaker
3.在coro1上编辑/etc/corosync.conf
1 cd /etc/corosync/ 2 cp corosync.conf.example corosync.conf 3 vim corosync.conf 4 ------------------------------------ 5 bindnetaddr: 192.168.19.0 //为你的网卡所在网络的网络地址 6 mcastaddr: 239.255.19.1 //组播地址 7 service { //添加此部分 8 ver: 0 9 name: pacemaker 10 # use_mgmtd: yes 11 }
4.在coro1上生成节点间通信时用到的认证密钥文件:
1 corosync-keygen
5.将corosync.conf和authkey复制至coro2上:
1 scp -p corosync.conf authkey coro2:/etc/corosync/
三、安装crmsh并启动corosync
1.安装crmsh,需要提前下好rpm包:
1 yum -y install crmsh-2.1-1.6.x86_64.rpm
2.启动corosync(在coro1上执行):
1 service corosync start
3.查看corosync引擎是否正常启动:
1 grep -e "Corosync Cluster Engine" -e "configuration file" /var/log/cluster/corosync.log
4.查看初始化成员节点通知是否正常发出:
1 grep TOTEM /var/log/cluster/corosync.log
5.检查启动过程中是否有错误产生。下面的错误信息表示packmaker不久之后将不再作为corosync的插件运行,因此,建议使用cman作为集群基础架构服务,此处可安全忽略:
1 grep ERROR: /var/log/cluster/corosync.log
6.查看pacemaker是否正常启动:
1 grep pcmk_startup /var/log/cluster/corosync.log
7.如果上面命令执行均没有问题,接着可以在coro1上执行如下命令启动coro2上的corosync:
1 ssh coro2 'service corosync start'
8.如果安装了crmsh,可使用如下命令查看集群节点的启动状态。从下面的信息可以看出两个节点都已经正常启动,并且集群已经处于正常工作状态:
1 crm status
四、编译安装drbd并配置
drbd共有两部分组成:内核模块和用户空间的管理工具。其中drbd内核模块代码已经整合进Linux内核2.6.33以后的版本中,因此,如果您的内核版本高于此版本的话,你只需要安装管理工具即可;否则,您需要同时安装内核模块和管理工具两个软件包,并且此两者的版本号一定要保持对应。
1.安装用户空间工具:
1 tar xf drbd-8.4.4.tar.gz 2 cd drbd-8.4.4 3 ./configure --prefix=/usr/local/drbd --with-km 4 make KDIR=/usr/src/kernels/2.6.32-504.30.3.el6.x86_64 //换成自己的内核版本 5 make install 6 mkdir -pv /usr/local/drbd/var/run/drbd 7 cp /usr/local/drbd/etc/rc.d/init.d/drbd /etc/rc.d/init.d 8 chkconfig --add drbd 9 chkconfig drbd off //一会作为corosync的资源,所以不要开机自动启动
2.安装drbd模块:
1 cd drbd 2 make clean 3 make KDIR=/usr/src/kernels/2.6.32-504.30.3.el6.x86_64 //换成自己的内核版本 4 cp drbd.ko /lib/modules/`uname -r`/kernel/lib 5 depmod
3.给主机分区,大小为5G。这里不再给出具体步骤,两台主机的/dev/sda7为一会要使用的分区:
4.配置drbd(在coro1上操作):
drbd的主配置文件为/etc/drbd.conf;为了管理的便捷性,目前通常会将些配置文件分成多个部分,且都保存至/etc/drbd.d/目录中,主配置文件中仅使用"include"指令将这些配置文件片断整合起来。通常/etc/drbd.d目录中的配置文件为global_common.conf和所有以.res结尾的文件。其中global_common.conf中主要定义global段和common段,而每一个.res的文件用于定义一个资源。
1 cd /usr/local/drbd/etc/drbd.d/ 2 vim global_common.conf 3 -------------------------------------- 4 global { 5 usage-count no; 6 # minor-count dialog-refresh disable-ip-verification 7 } 8 9 common { 10 protocol C; 11 12 handlers { 13 # These are EXAMPLE handlers only. 14 # They may have severe implications, 15 # like hard resetting the node under certain circumstances. 16 # Be careful when chosing your poison. 17 18 # pri-on-incon-degr "/usr/lib/drbd/notify-pri-on-incon-degr.sh; /usr/lib/drbd/notify-emergency-reboot.sh; echo b > /p 19 roc/sysrq-trigger ; reboot -f"; 20 # pri-lost-after-sb "/usr/lib/drbd/notify-pri-lost-after-sb.sh; /usr/lib/drbd/notify-emergency-reboot.sh; echo b > /p 21 roc/sysrq-trigger ; reboot -f"; 22 # local-io-error "/usr/lib/drbd/notify-io-error.sh; /usr/lib/drbd/notify-emergency-shutdown.sh; echo o > /proc/sysrq- 23 trigger ; halt -f"; 24 # fence-peer "/usr/lib/drbd/crm-fence-peer.sh"; 25 # split-brain "/usr/lib/drbd/notify-split-brain.sh root"; 26 # out-of-sync "/usr/lib/drbd/notify-out-of-sync.sh root"; 27 # before-resync-target "/usr/lib/drbd/snapshot-resync-target-lvm.sh -p 15 -- -c 16k"; 28 # after-resync-target /usr/lib/drbd/unsnapshot-resync-target-lvm.sh; 29 } 30 31 startup { 32 # wfc-timeout degr-wfc-timeout outdated-wfc-timeout wait-after-sb 33 } 34 35 options { 36 # cpu-mask on-no-data-accessible 37 } 38 39 disk { 40 #on-io-error detach; 41 # size max-bio-bvecs on-io-error fencing disk-barrier disk-flushes 42 # disk-drain md-flushes resync-rate resync-after al-extents 43 # c-plan-ahead c-delay-target c-fill-target c-max-rate 44 # c-min-rate disk-timeout 45 } 46 net { 47 #cram-hmac-alg "sha1"; 48 #shared-secret "mydrbd"; 49 # protocol timeout max-epoch-size max-buffers unplug-watermark 50 # connect-int ping-int sndbuf-size rcvbuf-size ko-count 51 # allow-two-primaries cram-hmac-alg shared-secret after-sb-0pri 52 # after-sb-1pri after-sb-2pri always-asbp rr-conflict 53 # ping-timeout data-integrity-alg tcp-cork on-congestion 54 # congestion-fill congestion-extents csums-alg verify-alg 55 # use-rle 56 } 57 }
5.定义一个资源,web是资源名,可以自己改:
1 vim /usr/local/drbd/etc/drbd.d/web.res 2 -------------------------------------- 3 resource web { 4 on coro1 { 5 device /dev/drbd0; 6 disk /dev/sda7; 7 address 192.168.19.55:7789; 8 meta-disk internal; 9 } 10 on coro2 { 11 device /dev/drbd0; 12 disk /dev/sda7; 13 address 192.168.19.98:7789; 14 meta-disk internal; 15 } 16 }
6.以上文件在两个节点上必须相同,因此,可以基于ssh将刚才配置的文件全部同步至coro2节点:
1 scp /usr/local/drbd/etc/drbd.d/* coro2:/usr/local/drbd/etc/drbd.d/
7.在两个节点上初始化已定义的资源并启动服务:
1 drbdadm create-md web 2 service drbd start
8.查看启动状态,从下面的信息中可以看出此时两个节点均处于Secondary状态:
1 cat /proc/drbd
9.我们接下来需要将其中一个节点设置为Primary,在要设置为Primary的节点上执行如下命令,这里在coro1上操作:
1 drbdadm primary --force web 2 cat /proc/drbd
而后再次查看状态,可以发现数据同步过程已经开始。等数据同步完成以后再次查看状态,可以发现主次位置显示是不一样的。斜线前面的代表当前主机,后面的代表对方主机。
在coro1上查看:
在coro2上查看:
10.创建文件系统并挂载:
文件系统的挂载只能在Primary节点进行,因此,也只有在设置了主节点后才能对drbd设备进行格式化,这里在coro1上进行操作。
1 mke2fs -t ext4 -L DRBD /dev/drbd0 2 mkdir /mydata //创建挂载点 3 mount /dev/drbd0 /mydata/ 4 mkdir /mydata/www //创建网页目录 5 mkdir /mydata/data //创建数据库目录
11.切换Primary和Secondary节点:
对主Primary/Secondary模型的drbd服务来讲,在某个时刻只能有一个节点为Primary,因此要切换两个节点的角色,只能在先将原有的Primary节点设置为Secondary后,才能原来的Secondary节点设置为Primary。
在coro1上:
1 umount /mydata/ 2 drbdadm secondary web //变为secondary节点
在coro2上:
1 drbdadm primary web //变为primary节点 2 cat /proc/drbd 3 mkdir /mydata //创建挂载点 4 mount /dev/drbd0 /mydata/ 5 ls /mydata/
可以看到,coro2已经成为了Primary,之前在coro1上创建的目录也能够访问到。测试完成后,再次将coro1变为Primary并挂载。
五、准备LNMP环境
1.编译安装tengine:
1 yum -y install pcre-devel 2 useradd -r nginx //要保证两台主机的nginx用户id和组id完全一致 3 tar xf tengine-2.1.0.tar.gz 4 cd tengine-2.1.0 5 ./configure --prefix=/usr/local/nginx --sbin-path=/usr/local/nginx/sbin/nginx --conf-path=/etc/nginx/nginx.conf --error-log-path=/var/log/nginx/error.log --http-log-path=/var/log/nginx/access.log --pid-path=/var/run/nginx/nginx.pid --lock-path=/var/lock/nginx.lock --user=nginx --group=nginx --with-http_ssl_module --with-http_flv_module --with-http_stub_status_module --with-http_gzip_static_module --http-client-body-temp-path=/var/tmp/nginx/client/ --http-proxy-temp-path=/var/tmp/nginx/proxy/ --http-fastcgi-temp-path=/var/tmp/nginx/fcgi/ --http-uwsgi-temp-path=/var/tmp/nginx/uwsgi --http-scgi-temp-path=/var/tmp/nginx/scgi --with-pcre 6 make && make install
2.为tengine提供脚本:
1 vim /etc/init.d/nginx 2 ------------------------------------ 3 #!/bin/sh 4 # 5 # nginx - this script starts and stops the nginx daemon 6 # 7 # chkconfig: - 85 15 8 # description: Nginx is an HTTP(S) server, HTTP(S) reverse \ 9 # proxy and IMAP/POP3 proxy server 10 # processname: nginx 11 # config: /etc/nginx/nginx.conf 12 # config: /etc/sysconfig/nginx 13 # pidfile: /var/run/nginx.pid 14 15 # Source function library. 16 . /etc/rc.d/init.d/functions 17 18 # Source networking configuration. 19 . /etc/sysconfig/network 20 21 # Check that networking is up. 22 [ "$NETWORKING" = "no" ] && exit 0 23 24 nginx="/usr/local/nginx/sbin/nginx" 25 prog=$(basename $nginx) 26 27 NGINX_CONF_FILE="/etc/nginx/nginx.conf" 28 29 [ -f /etc/sysconfig/nginx ] && . /etc/sysconfig/nginx 30 31 lockfile=/var/lock/subsys/nginx 32 33 make_dirs() { 34 # make required directories 35 user=`nginx -V 2>&1 | grep "configure arguments:" | sed 's/[^*]*--user=\([^ ]*\).*/\1/g' -` 36 options=`$nginx -V 2>&1 | grep 'configure arguments:'` 37 for opt in $options; do 38 if [ `echo $opt | grep '.*-temp-path'` ]; then 39 value=`echo $opt | cut -d "=" -f 2` 40 if [ ! -d "$value" ]; then 41 # echo "creating" $value 42 mkdir -p $value && chown -R $user $value 43 fi 44 fi 45 done 46 } 47 48 start() { 49 [ -x $nginx ] || exit 5 50 [ -f $NGINX_CONF_FILE ] || exit 6 51 make_dirs 52 echo -n $"Starting $prog: " 53 daemon $nginx -c $NGINX_CONF_FILE 54 retval=$? 55 echo 56 [ $retval -eq 0 ] && touch $lockfile 57 return $retval 58 } 59 60 stop() { 61 echo -n $"Stopping $prog: " 62 killproc $prog -QUIT 63 retval=$? 64 echo 65 [ $retval -eq 0 ] && rm -f $lockfile 66 return $retval 67 } 68 69 restart() { 70 configtest || return $? 71 stop 72 sleep 1 73 start 74 } 75 76 reload() { 77 configtest || return $? 78 echo -n $"Reloading $prog: " 79 killproc $nginx -HUP 80 RETVAL=$? 81 echo 82 } 83 84 force_reload() { 85 restart 86 } 87 88 configtest() { 89 $nginx -t -c $NGINX_CONF_FILE 90 } 91 92 rh_status() { 93 status $prog 94 } 95 96 rh_status_q() { 97 rh_status >/dev/null 2>&1 98 } 99 100 case "$1" in 101 start) 102 rh_status_q && exit 0 103 $1 104 ;; 105 stop) 106 rh_status_q || exit 0 107 $1 108 ;; 109 restart|configtest) 110 $1 111 ;; 112 reload) 113 rh_status_q || exit 7 114 $1 115 ;; 116 force-reload) 117 force_reload 118 ;; 119 status) 120 rh_status 121 ;; 122 condrestart|try-restart) 123 rh_status_q || exit 0 124 ;; 125 *) 126 echo $"Usage: $0 {start|stop|status|restart|condrestart|try-restart|reload|force-reload|configtest}" 127 exit 2 128 esac 129 ------------------------------------ 130 chmod +x /etc/init.d/nginx 131 chkconfig --add nginx 132 chkconfig nginx off //一会作为corosync的资源,所以不要开机自动启动
3.安装mysql,man和path路径输出这里不再演示:
1 useradd -r mysql //保证两台主机的mysql用户id和组id完全一致 2 tar xf mariadb-10.0.20-linux-x86_64.tar.gz -C /usr/local/ 3 cd /usr/local/ 4 ln -sv mariadb-10.0.20-linux-x86_64 mysql 5 cd mysql/ 6 chown -R root:mysql . 7 ln -sv /usr/local/mysql/include /usr/include/mysql //输出mysql的头文件至系统头文件路径 8 echo '/usr/local/mysql/lib' > /etc/ld.so.conf.d/mysql.conf //输出mysql的库文件给系统库查找路径 9 ldconfig
本试验中coro1为drbd的Primary,所以下面的操作在coro1上进行,提前将/dev/brbd0挂载至/mydata:
1 chown -R mysql:mysql /mydata/data/ 2 /usr/local/mysql/scripts/mysql_install_db --user=mysql --datadir=/mydata/data/ //初始化mysql 3 cp /usr/local/mysql/support-files/my-large.cnf /etc/my.cnf //提供配置文件 4 vim /etc/my.cnf 5 ------------------------------------ 6 datadir = /mydata/data //增加此项 7 ------------------------------------ 8 cp /usr/local/mysql/support-files/mysql.server /etc/init.d/mysqld //提供服务脚本 9 scp -p /etc/init.d/mysqld coro2:/etc/init.d/ 10 scp -p /etc/my.cnf coro2:/etc/ 11 chkconfig --add mysqld 12 chkconfig mysqld off //一会作为corosync的资源,所以不要开机自动启动 13 service mysqld start 14 /usr/local/mysql/bin/mysql //连接mysql 15 ------------------------------------ 16 CREATE DATABASE abcd; //创建abcd数据库 17 SHOW DATABASES; //显示数据库列表
停掉mysql,让drbd主节点切换至coro2并挂载,在coro2上连接mysql,切换主次步骤请参考前面:
1 chkconfig --add mysqld 2 chkconfig mysqld off 3 service mysqld start 4 /usr/local/mysql/bin/mysql 5 ----------------------------------- 6 SHOW DATABASES;
mysql测试成功,coro2上也可以看到coro1创建的数据。停掉mysql,让主节点再次切换到coro1上并挂载/dev/drbd0。
4.编译安装php:
1 yum -y install libxml2-devel bzip2-devel libcurl-devel libmcrypt-devel 2 tar xf php-5.6.11.tar.bz2 3 cd php-5.6.11 4 ./configure --prefix=/usr/local/php --with-mysql=/usr/local/mysql --with-openssl --enable-fpm --enable-sockets --enable-sysvshm --with-mysqli=/usr/local/mysql/bin/mysql_config --enable-mbstring --with-freetype-dir --with-jpeg-dir --with-png-dir --with-zlib-dir --with-libxml-dir=/usr --enable-xml --with-mhash --with-mcrypt --with-config-file-path=/etc --with-config-file-scan-dir=/etc/php.d --with-bz2 --with-curl --with-pdo-mysql=/usr/local/mysql 5 make && make install
5.为php和php-fpm提供配置文件:
1 cp php.ini-production /etc/php.ini //提供php配置文件 2 cp sapi/fpm/init.d.php-fpm /etc/init.d/php-fpm //提供脚本 3 chmod +x /etc/init.d/php-fpm 4 chkconfig --add php-fpm 5 chkconfig php-fpm off //一会作为corosync的资源,所以不要开机自动启动 6 cp /usr/local/php/etc/php-fpm.conf.default /usr/local/php/etc/php-fpm.conf //提供php-fpm配置文件 7 vim /usr/local/php/etc/php-fpm.conf 8 --------------------------------------------------- 9 pm.max_children = 150 10 pm.start_servers = 8 11 pm.min_spare_servers = 5 12 pm.max_spare_servers = 10 13 pid = /usr/local/php/var/run/php-fpm.pid 14 user = nginx 15 group = nginx 16 listen = 192.168.19.150:9000 //监听在VIP的9000端口
六、LNMP与DRBD结合
1.现在试验中coro1为drbd的Primary节点,所以以下操作在coro1上进行:
1 chown -R nginx:nginx /mydata/www/ 2 vim /etc/nginx/nginx.conf 3 ---------------------------------------- 4 location / { 5 root /mydata/www; 6 index index.php index.html index.htm; 7 } 8 location ~ \.php$ { 9 root /mydata/www; 10 fastcgi_pass 192.168.19.150:9000; 11 fastcgi_index index.php; 12 fastcgi_param SCRIPT_FILENAME /scripts$fastcgi_script_name; 13 include fastcgi_params; 14 } 15 ---------------------------------------- 16 vim /etc/nginx/fastcgi_params //将此文件内容改为如下几行 17 ---------------------------------------- 18 fastcgi_param GATEWAY_INTERFACE CGI/1.1; 19 fastcgi_param SERVER_SOFTWARE nginx; 20 fastcgi_param QUERY_STRING $query_string; 21 fastcgi_param REQUEST_METHOD $request_method; 22 fastcgi_param CONTENT_TYPE $content_type; 23 fastcgi_param CONTENT_LENGTH $content_length; 24 fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name; 25 fastcgi_param SCRIPT_NAME $fastcgi_script_name; 26 fastcgi_param REQUEST_URI $request_uri; 27 fastcgi_param DOCUMENT_URI $document_uri; 28 fastcgi_param DOCUMENT_ROOT $document_root; 29 fastcgi_param SERVER_PROTOCOL $server_protocol; 30 fastcgi_param REMOTE_ADDR $remote_addr; 31 fastcgi_param REMOTE_PORT $remote_port; 32 fastcgi_param SERVER_ADDR $server_addr; 33 fastcgi_param SERVER_PORT $server_port; 34 fastcgi_param SERVER_NAME $server_name; 35 ---------------------------------------- 36 scp /etc/nginx/nginx.conf coro2:/etc/nginx/ 37 scp /etc/nginx/fastcgi_params coro2:/etc/nginx/ 38 ifconfig eth0:0 192.168.19.150/24 up //手动启动VIP 39 service nginx start 40 service mysqld start 41 service php-fpm start
在/mydata/www下放入网页文件,此处以wordpress为例。用浏览器打开http://192.168.19.150,如果没问题,则会出现安装页面,按照提示安装即可。我这里手动创建了wordpress数据库,并给root用户添加了密码,安装时就直接使用root。
1 /usr/local/mysql/bin/mysql 2 ----------------------------------------- 3 CREATE DATABASE wordpress; 4 UPDATE mysql.user SET Password=PASSWORD(123456) WHERE User='root'; 5 FLUSH PRIVILEGES;
2.切换drbd的Primary节点为coro2,并测试:
在coro1上操作:
1 service nginx stop 2 service mysqld stop 3 service php-fpm stop 4 umount /mydata/ 5 drbdadm secondary web
6 ifconfig eth0:0 down
在coro2上操作:
1 drbdadm primary web 2 mount /dev/drbd0 /mydata/ 3 ifconfig eth0:0 192.168.19.150/24 up 4 service nginx start 5 service mysqld start 6 service php-fpm start
手动切换完毕之后,同样打开http://192.168.19.150,进行任何操作均没问题。至此,drbd+LNMP搭建完毕。下一步,就要将corosync和pacemaker整合进来了。现在检查两台主机的VIP、nginx、php-fpm、mysql、drbd均为关闭状态,且不能开机自动启动。
七、使用crmsh为corosync添加资源
1.创建drbd配置文件的软连接(两台机器均操作):
1 ln -sv /usr/local/drbd/etc/drbd.conf /etc/drbd.conf 2 ln -sv /usr/local/drbd/etc/drbd.d /etc/drbd.d
2.修改php-fpm启动脚本为如下内容,如不修改,则corosync无法识别(两台机器均操作):
1 #!/bin/sh 2 #======================================================================================= 3 # php-fpm - this script start and stop the php-fpm daemon 4 # 5 # chkconfig 35 on 6 # description: php-fpm is a FastCGI web server. 7 # processname: php-fpm 8 # config: /usr/local/php/etc/php-fpm.conf 9 # pidfile: /var/run/php-fpm.pid 10 #======================================================================================= 11 PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin 12 13 BINFILE="/usr/local/php/sbin/php-fpm" 14 CFGFILE="/usr/local/php/etc/php-fpm.conf" 15 PIDFILE="/var/run/php-fpm.pid" 16 LOCKFILE="/var/lock/php-fpm.lock" 17 18 RETVAL=0 19 20 start() { 21 [[ -x $BINFILE ]] || exit 5 22 [[ -f $CFGFILE ]] || exit 6 23 24 if [[ `ps aux | grep php-fpm: | grep -v grep | wc -l` -gt 0 ]]; then 25 echo "The php-fpm is already running." 26 return 1 27 fi 28 29 $BINFILE -t >/dev/null 2>&1 30 31 if [[ $? -ne 0 ]]; then 32 echo "The php-fpm configure has error." 33 return 1 34 fi 35 36 echo -n "Starting php-fpm......" 37 $BINFILE -y $CFGFILE -g ${PIDFILE} 38 RETVAL=$? 39 echo 40 [[ $RETVAL -eq 0 ]] && touch $LOCKFILE 41 42 return $RETVAL 43 } 44 45 stop() { 46 if [[ `ps aux | grep php-fpm: | grep -v grep | wc -l` -eq 0 ]]; then 47 echo "The php-fpm is not running." 48 return 1 49 fi 50 51 echo -n "Shutting down php-fpm......" 52 53 if [[ -f $PIDFILE ]]; then 54 kill -QUIT `cat ${PIDFILE}` 55 else 56 kill -QUIT `ps aux | grep php-fpm | awk '/master/{print $2}'` 57 fi 58 59 RETVAL=$? 60 echo 61 [[ $RETVAL -eq 0 ]] && rm -f $LOCKFILE $PIDFILE 62 63 return $RETVAL 64 } 65 66 restart() { 67 stop 68 sleep 1 69 70 while true 71 do 72 if [[ `ps aux | grep php-fpm: | grep -v grep | wc -l` -eq 0 ]]; then 73 start 74 break 75 fi 76 sleep 1 77 done 78 79 RETVAL=$? 80 echo 81 82 return $RETVAL 83 } 84 85 reload() { 86 if [[ `ps aux | grep php-fpm: | grep -v grep | wc -l` -eq 0 ]]; then 87 echo "The php-fpm is not running." 88 return 1 89 fi 90 91 echo -n $"Reloading php-fpm......" 92 93 if [[ -f $PIDFILE ]]; then 94 kill -USR2 `cat ${PIDFILE}` 95 else 96 kill -USR2 `ps aux | grep php-fpm | awk '/master/{print $2}'` 97 fi 98 99 RETVAL=$? 100 echo 101 102 return $RETVAL 103 } 104 105 case "$1" in 106 start) 107 start 108 ;; 109 110 stop) 111 stop 112 ;; 113 114 restart) 115 restart 116 ;; 117 118 reload) 119 reload 120 ;; 121 122 *) 123 echo "Usage: service php-fpm {start|stop|restart|reload}" 124 RETVAL=1 125 esac 126 127 exit $RETVAL
下面运行crm命令,进入crm命令行模式,非特殊注明情况下操作均在此模式下进行,在一台机器上操作即可。
3.关闭corosync的stonith检测和仲裁:
1 configure //进入配置菜单 2 property stonith-enabled=false //关闭stonith设备 3 property no-quorum-policy=ignore //关闭仲裁设备 4 verify //语法检测 5 commit //提交
4.添加资源(configure菜单下):
1 primitive mydrbd ocf:drbd params drbd_resource=web op monitor role=Master interval=20s timeout=30s op monitor role=Slave interval=30s timeout=20s op start timeout=240s op stop timeout=240s //定义drbd资源 2 primitive myfile Filesystem params device="/dev/drbd0" directory="/mydata" fstype="ext4" op monitor interval=30s timeout=60s op start timeout=60s op stop timeout=60s //定义文件系统资源 3 primitive myip IPaddr params ip=192.168.19.150 op monitor interval=30s timeout=60s //定义VIP资源 4 primitive mynginx lsb:nginx op monitor interval=30s timeout=20s //定义nginx资源 5 primitive mysql lsb:mysqld op monitor interval=30s timeout=60s //定义mysqld资源 6 primitive myphp lsb:php-fpm //定义php-fpm资源 7 group myserver myfile myip mynginx myphp mysql //定义资源组 8 verify //语法检测,如没问题,先不要提交
5.添加各种约束:
1 ms ms_mydrbd mydrbd meta master-max=1 master-node-max=1 clone-max=2 clone-node-max=1 notify=True //drbd主从资源配置 2 colocation myfile_with_mydrbd_master inf: myfile ms_mydrbd:Master //定义排列约束,myfile要和drbd的Master节点在一起 3 colocation mysql_with_myfile inf: mysql myfile //mysql要和myfile在一起 4 order myfile_after_ms_mydrbd_master Mandatory: ms_mydrbd:promote myfile:start //定义顺序约束,drbd主节点被提升,myfile才能启动 5 order myip_after_myfile Mandatory: myfile myip 6 order mynginx_after_myip Mandatory: myip mynginx 7 verify 8 commit
6.提交之后,可以看到,所有资源已经运行在coro2,打开http://192.168.19.150进行操作也没问题:
7.下面操作让coro2成为备用节点。可以看到,资源全部转移到coro1上了,打开http://192.168.19.150进行各种操作仍然OK:
1 cd .. 2 node standby coro2 //coro2成为备用
----------------------------- 等待5秒 3 node online coro2 //重新上线 4 status //查看状态
至此,实验全部结束。因为时间原因,很多细节有待完善,欢迎大家提供宝贵意见,谢谢! QQ:82800452