江南才子 发表于 2021-7-3 21:44:31

LVS+keepalived+DRBD+heartbeat+mysql


  

  IP地址规划:
  LVS:
10.1.1.23 主LVS+keepalived
10.1.1.22 备LVS+keepalived  

  主mysql:
10.1.1.5DRBD+heartbeat
10.1.1.6DRBD+heartbeat
10.1.1.7VIP(heartbeat)
10.1.1.10 VIP(lvs)  

  从mysql+nginx:
10.1.1.102 Ng+php
10.1.1.102 Ng+php
VIP1:10.1.1.8 (nginx)
VIP2: 10.1.1.9 (mysql)  安装DRBD+mysql+heartbeat

  安装DRBD
  1.1.规划
10.1.1.5(主)/data
10.1.1.6(备)/data  

  2.安装(主备安装一致):
  2.1 安装依赖:
yum install kernel-devel kernel-headers  

  2.2.编译安装
tar -zxf drbd-8.3.16.tar.gz
./configure --prefix=/usr/local/drbd --with-km
make && make install  2.3 拷贝配置文件模板
mv /usr/local/drbd/etc/drbd.conf /usr/local/drbd/drbd.conf.old
cp /tmp/soft/drbd-8.3.16/scripts/drbd.conf.exampl /usr/local/drbd/etc/drbd.conf  

  2.4 设置链接
ln -s /usr/local/drbd/etc/drbd.conf /etc/drbd.conf
ln -s /usr/local/drbd/etc/rc.d/init.d/drbd /etc/init.d/drbd  3.配置文件(主备一致):

global {
    usage-count no;
}

common {
syncer { rate 200M; }
}
resource r0 {

protocol C;
handlers {
    pri-on-incon-degr "/usr/lib/drbd/notify-pri-on-incon-degr.sh; /usr/lib/drbd/notify-emergency-reboot.sh; echo b > /proc/sysrq-trigger ; reboot -f";
    pri-lost-after-sb "/usr/lib/drbd/notify-pri-lost-after-sb.sh; /usr/lib/drbd/notify-emergency-reboot.sh; echo b > /proc/sysrq-trigger ; reboot -f";
    local-io-error "/usr/lib/drbd/notify-io-error.sh; /usr/lib/drbd/notify-emergency-shutdown.sh; echo o > /proc/sysrq-trigger ; halt -f";
    fence-peer "/usr/lib/heartbeat/drbd-peer-outdater -t 5";
    pri-lost "/usr/lib/drbd/notify-pri-lost.sh; /usr/lib/drbd/notify-emergency-reboot.sh; echo b > /proc/sysrq-trigger ; reboot -f";
    split-brain "/usr/lib/drbd/notify-split-brain.sh root";
    out-of-sync "/usr/lib/drbd/notify-out-of-sync.sh root";
}

startup {
    degr-wfc-timeout 120;
    outdated-wfc-timeout 2;
}

disk {
    on-io-error   detach;
    fencing resource-only;
}

net {
    cram-hmac-alg "sha1";
    shared-secret "ckl893";
}

on ckl-bd1 {
    device   /dev/drbd0;
    disk       /dev/sda5;
    address    10.1.1.5:7788;
    flexible-meta-diskinternal;
}

on ckl-bd2 {
    device    /dev/drbd0;
    disk      /dev/sda5;
    address   10.1.1.6:7788;
    meta-disk internal;

}
}  
  4.创建数据块:
  (主备一致)
#dd if=/dev/zero bs=1M count=1 of=/dev/sda5;sync
#drbdadm create-md r0  

  5.启动DRBD服务
#umount /data
#/etc/init.d/drdb start  6.配置主服务,格式化DRBD(10.1.1.5):
#drbdsetup /dev/drbd1 primary –o
# mkfs.ext4 /dev/drbd0  

  7.主挂载DRBD,创建文件(10.1.1.5):
#mount /dev/drbd0 /mnt
#cd /mnt
#touch nihao.txt  

  8.测试同步:
  停掉主(10.1.1.5):
#/etc/init.d/drbd stop  

  从设置为主(10.1.1.6):
#drbdsetup /dev/drbd1 primary –o
# mount /dev/drbd0 /mnt#cd /mnt
# ll nihao.txt
-rw-r--r-- 1 root root 0 May7 10:42 nihao.txt  9.切换回来:
  10.1.1.6上执行:
#umount /mnt
#drbdadm secondary all  10.1.1.5 上执行:
#drbdadm primary all
# mount /dev/drbd0 /mnt  

  安装mysql
  1.mysql安装规划:
10.1.1.5(主):/mnt DRBD挂载的分区
10.1.1.6(备)  

  2.首先启动主DRBD(10.1.1.5)挂载DRBD到/mnt
安装mysql到/mnt 下  

  3.备(10.1.1.6)切换到主,挂载DRBD 到/mnt
安装mysql到/mnt 下  

  4.切换回
10.1.1.5主
10.1.1.6备  

  安装heartbeat
  规划:10.1.1.5 (主)
10.1.1.6 (备)
10.1.1.7 (heartbeat VIP)  主备安装一致:
  1.安装heartbeat 下载需要的源:
wget ftp://mirror.switch.ch/pool/1/mirror/scientificlinux/6rolling/x86_64/os/Packages/epel-release-6-5.noarch.rpm  2.安装heartbeat:
yum -y install heartbeat*  

  3.配置heartbeat
  3.1.拷贝模板配置文件:
cp /usr/share/doc/heartbeat-3.0.4/ha.cf /etc/ha.d/
cp /usr/share/doc/heartbeat-3.0.4/authkeys /etc/ha.d/
cp /usr/share/doc/heartbeat-3.0.4/haresources /etc/ha.d/  4.heartbeat主配置文件(主备一致):
cat /etc/ha.d/ha.cf
logfile /var/log/ha-log
logfacility   local0
keepalive 2
deadtime 30
warntime 10
initdead 120
udpport 694
baud    19200
bcast   eth0
auto_failback on
watchdog /dev/watchdog
node    ckl-bd1
node    ckl-bd2
ping 10.1.1.1
respawn hacluster /usr/lib64/heartbeat/ipfail
apiauth ipfail gid=haclient uid=hacluster  5.heartbeat认证文件(主备一致):
cat /etc/ha.d/authkeys
auth 1
1 sha1 ckl893!  6.资源文件
  10.1.1.5(主):
# cat /etc/ha.d/haresources
ckl-bd1 drbddisk::r0 Filesystem::/dev/drbd0::/mnt mysql IPaddr::10.1.1.7/24/eth0:1
10.1.1.6(备):
# cat /etc/ha.d/haresources
ckl-bd2 drbddisk::r0 Filesystem::/dev/drbd0::/mnt mysql IPaddr::10.1.1.7/24/eth0:1  

  drbdisk脚本需要添加到/etc/ha.d/resource.d 下
  7.测试heartbeat+mysql+DRBD
  主,停掉mysql,停掉heartbeat
  从监测到后,挂载DRBD,启动mysql
  

  配置mysql主从同步
10.1.1.107 主
10.1.1.102 从
10.1.1.103 从  配置简单略过
  

  安装LVS
  1.安装lvs
10.1.1.23
10.1.1.22  

  执行同样操作:
  1.1.安装需要包:
yum -y install popt popt-devel popt-static libnl libnl-devel  1.2.lvs在2.4之后内核自带,查看是否已经存在:
# cat /boot/config-2.6.32-573.el6.x86_64 | grep -i -A 15 "ipvs"
# IPVS transport protocol load balancing support
#
CONFIG_IP_VS_PROTO_TCP=y
CONFIG_IP_VS_PROTO_UDP=y
CONFIG_IP_VS_PROTO_AH_ESP=y
CONFIG_IP_VS_PROTO_ESP=y
CONFIG_IP_VS_PROTO_AH=y
CONFIG_IP_VS_PROTO_SCTP=y

#
# IPVS scheduler
#
CONFIG_IP_VS_RR=m
CONFIG_IP_VS_WRR=m
CONFIG_IP_VS_LC=m
CONFIG_IP_VS_WLC=m
CONFIG_IP_VS_LBLC=m
CONFIG_IP_VS_LBLCR=m
CONFIG_IP_VS_DH=m
CONFIG_IP_VS_SH=m
CONFIG_IP_VS_SED=m
CONFIG_IP_VS_NQ=m

#
# IPVS application helper
#
CONFIG_IP_VS_FTP=m
CONFIG_IP_VS_PE_SIP=m

#
# IP: Netfilter Configuration
#
CONFIG_NF_DEFRAG_IPV4=m
CONFIG_NF_CONNTRACK_IPV4=m
# CONFIG_NF_CONNTRACK_PROC_COMPAT is not set
CONFIG_IP_NF_QUEUE=m
CONFIG_IP_NF_IPTABLES=m
CONFIG_IP_NF_MATCH_ADDRTYPE=m
CONFIG_IP_NF_MATCH_AH=m
CONFIG_IP_NF_MATCH_ECN=m  1.3.说明ipvs:

IPVS transport protocol:ipvs支持的协议,tcp、udp、ah、esp、sctp
ipvs scheduler:ipvs调度模式,RR、WRR、LC、WLC、LBLC、LBLCR、DH、SH、SED、NQ  1.4.安装lvsadmin(lvs管理工具):
tar -zxvf ipvsadm-1.26.tar.gz
cd ipvsadm-1.26
make && make install  

  2.安装keepalived
  2.1.安装需要包:
yum -y install openssl-devel popt-devel libnl* libpopt* popt-static
ln -s /usr/src/kernels/2.6.32-279.el6.x86_64/ /usr/src/linux  2.2.安装lvs
tar -zxvf keepalived-1.2.4.tar.gz
cd keepalived-1.2.4
./configure --prefix=/usr --exec-prefix=/usr --bindir=/usr/bin --sbindir=/usr/sbin --sysconfdir=/etc --datadir=/usr/share --includedir=/usr/include --libdir=/usr/lib64 --libexecdir=/usr/libexec --localstatedir=/var --mandir=/usr/share/man --infodir=/usr/share/info --sharedstatedir=/usr/com --with-kernel-dir=/usr/src/linux
make && make install  3.配置keepalived
  3.1.在10.1.1.23 上配置:
# vi /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
   notification_email {
      kangly@cosyjoy.com
   }
   notification_email_from Alexandre.Cassen@firewall.loc
   smtp_server 10.1.1.23
   smtp_connect_timeout 30
   router_id LVS_MASTER
}
vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 77
    priority 100
    advert_int 1
    authentication {
      auth_type PASS
      auth_pass ckl893
    }
    virtual_ipaddress {
       10.1.1.8/24   #nginx VIP
       10.1.1.9/24   #mysql slave VIP
       10.1.1.10/24#mysql 主VIP
    }
}

virtual_server 10.1.1.8 80 {
    delay_loop 6
    lb_algo rr
    lb_kind DR
    persistence_timeout 50
    protocol TCP

    real_server 10.1.1.102 80 {
      weight 1
      TCP_CHECK {
            connect_timeout 10
            nb_get_retry 3
            delay_before_retry 3
      }
    }

    real_server 10.1.1.103 80 {
      weight 1
      TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
      }
    }
}

virtual_server 10.1.1.9 3306 {
    delay_loop 30
    lb_algo rr
    lb_kind DR
    persistence_timeout 120
    protocol TCP

    real_server 10.1.1.102 3306 {
      weight 1
                MISC_CHECK {
                misc_path "/etc/keepalived/check_slave.pl 10.1.1.102"
                misc_dynamic
      }
    }

    real_server 10.1.1.103 3306 {
      weight 1
                MISC_CHECK {
                misc_path "/etc/keepalived/check_slave.pl 10.1.1.103"
                misc_dynamic
      }
    }
}  3.2.配置备lvs
  在10.1.1.22 上:
  主从配置文件不一致
  state BACKUP状态为备份
  priority 90优先级小于主
  

  4.本地添加VIP脚本:
  10.1.1.5
  10.1.1.6
# cat /etc/init.d/lvsrs
#!/bin/bash
#description : start realserver
VIP=10.1.1.10
. /etc/rc.d/init.d/functions
case "$1" in
    start)
      echo " start LVSofREALServer"
/sbin/ifconfig lo:0 $VIP broadcast $VIP netmask 255.255.255.255 up
      echo "1" >/proc/sys/net/ipv4/conf/lo/arp_ignore
      echo "2" >/proc/sys/net/ipv4/conf/lo/arp_announce
      echo "1" >/proc/sys/net/ipv4/conf/all/arp_ignore
      echo "2" >/proc/sys/net/ipv4/conf/all/arp_announce
      ;;
    stop)
      /sbin/ifconfig lo:0 down
      echo "close LVS REALserver"
      echo "0" >/proc/sys/net/ipv4/conf/lo/arp_ignore
      echo "0" >/proc/sys/net/ipv4/conf/lo/arp_announce
      echo "0" >/proc/sys/net/ipv4/conf/all/arp_ignore
      echo "0" >/proc/sys/net/ipv4/conf/all/arp_announce
      ;;
    *)
      echo "Usage: $0 {start|stop}"
      exit 1
esac  5.本地添加ip脚本:
  10.1.1.102
  10.1.1.103
#!/bin/bash
#description : start realserver
VIP1=10.1.1.8
VIP2=10.1.1.9
. /etc/rc.d/init.d/functions
case "$1" in
    start)
      echo " start LVSofREALServer"
/sbin/ifconfig lo:0 $VIP1 broadcast $VIP1 netmask 255.255.255.255 up
/sbin/ifconfig lo:1 $VIP2 broadcast $VIP2 netmask 255.255.255.255 up
      echo "1" >/proc/sys/net/ipv4/conf/lo/arp_ignore
      echo "2" >/proc/sys/net/ipv4/conf/lo/arp_announce
      echo "1" >/proc/sys/net/ipv4/conf/all/arp_ignore
      echo "2" >/proc/sys/net/ipv4/conf/all/arp_announce
      ;;
    stop)
      /sbin/ifconfig lo:0 down
      echo "close LVS REALserver"
      echo "0" >/proc/sys/net/ipv4/conf/lo/arp_ignore
      echo "0" >/proc/sys/net/ipv4/conf/lo/arp_announce
      echo "0" >/proc/sys/net/ipv4/conf/all/arp_ignore
      echo "0" >/proc/sys/net/ipv4/conf/all/arp_announce
      ;;
    *)
      echo "Usage: $0 {start|stop}"
      exit 1
esac  arp相关参数说明:

arp_annonuce(是否将本机地址网络通过arp通告给其它网络):
    0:通告本机所有地址接口的网络。
    1:尽量只通告本接口地址的网络。
    2:只通告连接本地借口地址网络。

arg_ignore(是否响应其它的arp请求):
    0:在接受到请求后,将本地所有地址响应出去。
    1:请求报文从哪个接口进入,就必须使用此接口配置的地址才能响应。  

  6.mysql 同步检测脚本:
  注意事项:执行用户必须有show slave status 权限
# cat check_slave.pl
#!/usr/bin/perl -w
use DBI;
use DBD::mysql;

$SBM = 120;
$db = "cookbook";
$host = $ARGV;
$port = 3306;
$user = "root";
$pw = 'yiqu@123'; #这里是单引号,因为有@


$query = "show slave status";
$dbh = DBI->connect("DBI:mysql:$db:$host:$port", $user, $pw, { RaiseError => 0,PrintError => 0 });

if (!defined($dbh)) {
    exit 1;
}

$sqlQuery = $dbh->prepare($query);
$sqlQuery->execute;

$Slave_IO_Running ="";
$Slave_SQL_Running = "";
$Seconds_Behind_Master = "";

while (my $ref = $sqlQuery->fetchrow_hashref()) {
    $Slave_IO_Running = $ref->{'Slave_IO_Running'};
    $Slave_SQL_Running = $ref->{'Slave_SQL_Running'};
    $Seconds_Behind_Master = $ref->{'Seconds_Behind_Master'};
}

$sqlQuery->finish;
$dbh->disconnect();

if ( $Slave_IO_Running eq "No" || $Slave_SQL_Running eq "No" ) {
    exit 1;
} else {
    if ( $Seconds_Behind_Master gt $SBM ) {
      exit 1;
    } else {
      exit 0;
    }
}  7.启动LVS
# /etc/init.d/keepalived start  查看LVS路由表信息:
#ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port         Forward Weight ActiveConn InActConn
TCP10.1.1.8:80 rr persistent 50
-> 10.1.1.102:80                Route   1      0          0         
-> 10.1.1.103:80                Route   1      0          0         
TCP10.1.1.9:3306 rr persistent 120
-> 10.1.1.102:3306            Route   124    0          0         
-> 10.1.1.103:3306            Route   124    0          0         
TCP10.1.1.10:3306 rr persistent 120
-> 10.1.1.7:3306                Route   1      0          0  

  8.测试web
  停掉:10.1.1.102 nginx
# /etc/init.d/nginx stop  查看LVS路由表信息:
# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port         Forward Weight ActiveConn InActConn
TCP10.1.1.8:80 rr persistent 50
-> 10.1.1.103:80                Route   1      0          0         
TCP10.1.1.9:3306 rr persistent 120
-> 10.1.1.102:3306            Route   124    0          0         
-> 10.1.1.103:3306            Route   124    0          0         
TCP10.1.1.10:3306 rr persistent 120
-> 10.1.1.7:3306                Route   1      0          0  路由表删除了10.1.1.102 的信息;
  启动10.1.1.102 nginx
# /etc/init.d/nginx start  

  查看LVS路由表信息:
# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port         Forward Weight ActiveConn InActConn
TCP10.1.1.8:80 rr persistent 50
-> 10.1.1.102:80                Route   1      0          0         
-> 10.1.1.103:80                Route   1      0          0         
TCP10.1.1.9:3306 rr persistent 120
-> 10.1.1.102:3306            Route   124    0          0         
-> 10.1.1.103:3306            Route   124    0          0         
TCP10.1.1.10:3306 rr persistent 120
-> 10.1.1.7:3306                Route   1      0          0  9.测试mysql读
  停止10.1.1.102 mysql
  # /etc/init.d/mysqld stop
  查看路由表信息:
# ipvsadm -Ln      
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port         Forward Weight ActiveConn InActConn
TCP10.1.1.8:80 rr persistent 50
-> 10.1.1.102:80                Route   1      0          0         
-> 10.1.1.103:80                Route   1      0          0         
TCP10.1.1.9:3306 rr persistent 120
-> 10.1.1.103:3306            Route   124    0          0         
TCP10.1.1.10:3306 rr persistent 120
-> 10.1.1.7:3306                Route   1      0          0  已经剔除了10.1.1.102 mysql
  

  启动10.1.1.102 mysql
# /etc/init.d/mysqld start  查看路由表信息:
# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port         Forward Weight ActiveConn InActConn
TCP10.1.1.8:80 rr persistent 50
-> 10.1.1.102:80                Route   1      0          0         
-> 10.1.1.103:80                Route   1      0          0         
TCP10.1.1.9:3306 rr persistent 120
-> 10.1.1.103:3306            Route   124    0          0         
TCP10.1.1.10:3306 rr persistent 120
-> 10.1.1.7:3306                Route   1      0          0  测试检测同步脚本:
  停掉10.1.1.102 同步:
>slave stop;  

  查看路由表信息:
# ipvsadm -Ln      
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port         Forward Weight ActiveConn InActConn
TCP10.1.1.8:80 rr persistent 50
-> 10.1.1.102:80                Route   1      0          0         
-> 10.1.1.103:80                Route   1      0          0         
TCP10.1.1.9:3306 rr persistent 120
-> 10.1.1.103:3306            Route   124    0          0         
TCP10.1.1.10:3306 rr persistent 120
-> 10.1.1.7:3306                Route   1      0          0  启动同步
>slave start;  查看路由表信息:
# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port         Forward Weight ActiveConn InActConn
TCP10.1.1.8:80 rr persistent 50
-> 10.1.1.102:80                Route   1      0          0         
-> 10.1.1.103:80                Route   1      0          0         
TCP10.1.1.9:3306 rr persistent 120
-> 10.1.1.103:3306            Route   124    0          0         
TCP10.1.1.10:3306 rr persistent 120
-> 10.1.1.7:3306                Route   1      0          0  错误信息:
# tail -f /var/log/messages
May6 11:00:03 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 10.1.1.8
May6 11:00:10 localhost Keepalived_vrrp: receive an invalid ip number count associated with VRID!
May6 11:00:10 localhost Keepalived_vrrp: bogus VRRP packet received on eth0 !!!
May6 11:00:10 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Dropping received VRRP packet...
May6 11:00:11 localhost Keepalived_vrrp: receive an invalid ip number count associated with VRID!  

  解决办法:修改virtual_router_id 51 为另一个值,如:virtual_router_id 77
change master to master_host='10.1.1.7',master_port=3306,master_user='repl',master_password='slavepass',master_log_file='mysql-bin.000017',master_log_pos=881;
perl Can't locate DBI.pm
yum -y install perl perl-DBIperl-DBD-MySQL  


  
页: [1]
查看完整版本: LVS+keepalived+DRBD+heartbeat+mysql