【HA】Heartbeat高可用集群配置
主：hh.huangmingming.cn 192.168.1.31 从：yo.huangmingming.cn 192.168.1.250主和从hosts文件配置：
192.168.1.31 hh hh.huangmingming.cn
192.168.1.250 yo yo.huangmingming.cn
一、安装epel扩展源
# wget http://mirrors.sohu.com/fedora-epel/6/i386/epel-release-6-8.noarch.rpm
# yum list |grep heartbeat
heartbeat.x86_64 3.0.4-2.el6 epel
heartbeat-devel.i686 3.0.4-2.el6 epel
heartbeat-devel.x86_64 3.0.4-2.el6 epel
heartbeat-libs.i686 3.0.4-2.el6 epel
heartbeat-libs.x86_64 3.0.4-2.el6 epel
二、安装heartbeat、libnet（主从都需要安装）
# yum -y install heartbeat
# yum -y install libnet
# yum -y install nginx（安装nginx做测试）
三、配置heartbeat（在主上配置）
1、拷贝配置文件authkeys、ha.cf、haresources到目录/etc/ha.d/下
# cd /usr/share/doc/heartbeat-3.0.4/
# ls
apphbd.cfauthkeysAUTHORSChangeLogCOPYINGCOPYING.LGPLha.cfharesourcesREADME
# cp authkeys ha.cf haresources /etc/ha.d/
2、配置authkeys文件
# cd /etc/ha.d/
# vim authkeys
auth 3
#1 crc
#2 sha1 HI!
3 md5 Hello!
# chmod 600 authkeys （authkeys文件的权限要注意改成600）
3、配置haresources文件
# vim haresources
hh.huangmingming.cn 192.168.1.13/24/eth0:0 nginx（指定集群IP）
4、配置ha.cf文件
# vim ha.cf
debugfile /var/log/ha-debug
logfile /var/log/ha-log
logfacility local0
keepalive 2
deadtime 30
warntime 10
initdead 60
udpport 694
ucast eth0 192.168.1.250
auto_failback on
node hh.huangmingming.cn
node yo.huangmingming.cn
ping 192.168.1.1
respawn hacluster /usr/lib64/heartbeat/ipfail
5、把这三个文件发送到yo（从机器上）
# scp authkeys haresources ha.cf yo:/etc/ha.d/
6、启动heartbeat服务，先主后从
# /etc/init.d/heartbeat start
Starting High-Availability services: INFO:Resource is stopped
Done.
四、启动产生的错误
1、<--respawn hacluster /usr/lib64/heartbeat/ipfail （注意当前系统是64位还是32位的否则会产生以下错误）--/>
# /etc/init.d/heartbeat start
heartbeat: udpport setting must precede media statementsheartbeat: 2016/01/01_05:04:26 ERROR: Client child command is not executable
heartbeat: 2016/01/01_05:04:26 ERROR: Heartbeat not started: configuration error.
heartbeat: 2016/01/01_05:04:26 ERROR: Configuration error, heartbeat not started.
2、<--以下两个错误都是因为主机名配置不对造成的，在heartbeat配置中指定主机名的地方要与当前主机名对应，否则启不来--/>
# /etc/init.d/heartbeat start
heartbeat: udpport setting must precede media statementsheartbeat: baudrate setting must precede media statementsheartbeat: 2015/10/29_21:31:14 info: Pacemaker support: false
heartbeat: 2015/10/29_21:31:14 ERROR: Current node not in configuration!
heartbeat: 2015/10/29_21:31:14 info: By default, cluster nodes are named by `uname -n` and must be declared with a 'node' directive in the ha.cf file.
heartbeat: 2015/10/29_21:31:14 info: See also: http://linux-ha.org/wiki/Ha.cf#node_directive
heartbeat: 2015/10/29_21:31:14 WARN: Logging daemon is disabled --enabling logging daemon is recommended
heartbeat: 2015/10/29_21:31:14 ERROR: Configuration error, heartbeat not started.
# /etc/init.d/heartbeat start
Starting High-Availability services: INFO:Resource is stopped
Heartbeat failure . Failed.
heartbeat: udpport setting must precede media statementsheartbeat: baudrate setting must precede media statementsheartbeat: 2015/10/29_21:56:29 info: Pacemaker support: false
heartbeat: 2015/10/29_21:56:29 WARN: Logging daemon is disabled --enabling logging daemon is recommended
heartbeat: 2015/10/29_21:56:29 info: **************************
heartbeat: 2015/10/29_21:56:29 info: Configuration validated. Starting heartbeat 3.0.4
heartbeat: 2015/10/29_21:56:29 ERROR: Bad nodename in /etc/ha.d//haresources
heartbeat: 2015/10/29_21:56:29 ERROR: Configuration error, heartbeat not started.
五、故障模拟测试，如主down掉，看会不会切换到从继续提供服务
1、正常情况下
# /etc/init.d/heartbeat start
Starting High-Availability services: INFO:Resource is stopped
Done.
# ifconfig
eth0 Link encap:EthernetHWaddr 00:0C:29:97:EE:BF
inet addr:192.168.1.31Bcast:192.168.1.255Mask:255.255.255.0
inet6 addr: fe80::20c:29ff:fe97:eebf/64 Scope:Link
UP BROADCAST RUNNING MULTICASTMTU:1500Metric:1
RX packets:37359 errors:0 dropped:0 overruns:0 frame:0
TX packets:22139 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:36402808 (34.7 MiB)TX bytes:2377053 (2.2 MiB)
eth0:0 Link encap:EthernetHWaddr 00:0C:29:97:EE:BF
inet addr:192.168.1.13Bcast:192.168.1.255Mask:255.255.255.0
UP BROADCAST RUNNING MULTICASTMTU:1500Metric:1
eth0:1 Link encap:EthernetHWaddr 00:0C:29:97:EE:BF
inet addr:192.168.1.144Bcast:192.168.1.255Mask:255.255.255.0
UP BROADCAST RUNNING MULTICASTMTU:1500Metric:1
lo Link encap:Local Loopback
inet addr:127.0.0.1Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNINGMTU:16436Metric:1
RX packets:28 errors:0 dropped:0 overruns:0 frame:0
TX packets:28 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:2727 (2.6 KiB)TX bytes:2727 (2.6 KiB)
# ps aux |grep nginx （主）
root 313670.00.1964881728 ? Ss 21:58 0:00 nginx: master process /usr/sbin/nginx -c /etc/nginx/nginx.conf
nginx 313680.00.2968762500 ? S 21:58 0:00 nginx: worker process
root 316430.00.0 103252 824 pts/0 S+ 22:16 0:00 grep nginx
# netstat -tnlp |grep nginx （从）
# ps aux |grep nginx
root 52170.00.0 103256 828 pts/0 S+ 06:11 0:00 grep nginx
2、创建测试页面，并在正常情况下测试访问，此时由主提供服务
# echo "AAAAAAAAAAAAAAAAhh192.168.1.31" >/usr/share/nginx/html/index.html
# echo "AAAAAAAAAAAAAAAAyo192.168.1.250" >/usr/share/nginx/html/index.html
3、主服务节点down掉之后（测试）
# iptables -A INPUT -p icmp -j DROP
# tail /var/log/ha-log （查看日志信息）
ResourceManager(default):2015/10/29_22:24:20 info: Running /etc/ha.d/resource.d/IPaddr 192.168.1.13/24/eth0:0 start
IPaddr(IPaddr_192.168.1.13):2015/10/29_22:24:20 INFO: Adding inet address 192.168.1.13/24 with broadcast address 192.168.1.255 to device eth0 (with label eth0:0)
IPaddr(IPaddr_192.168.1.13):2015/10/29_22:24:20 INFO: Bringing device eth0 up
IPaddr(IPaddr_192.168.1.13):2015/10/29_22:24:20 INFO: /usr/libexec/heartbeat/send_arp -i 200 -r 5 -p /var/run/resource-agents/send_arp-192.168.1.13 eth0 192.168.1.13 auto not_used not_used
/usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_192.168.1.13):2015/10/29_22:24:20 INFO:Success
ResourceManager(default):2015/10/29_22:24:20 info: Running /etc/init.d/nginxstart
Oct 29 22:24:20 hh.huangmingming.cn ipfail: : info: NS: We are still alive!
Oct 29 22:24:20 hh.huangmingming.cn ipfail: : info: Link Status update: Link yo.huangmingming.cn/eth0 now has status dead
Oct 29 22:24:22 hh.huangmingming.cn ipfail: : info: Asking other side for ping node count.
Oct 29 22:24:22 hh.huangmingming.cn ipfail: : info: Checking remote count of ping nodes.
4、在从（yo）上查看有没有自动启动nginx，并在客户端访问，此时由从提供服务
# ps aux |grep nginx
root 55340.00.1964961972 ? Ss 06:20 0:00 nginx: master process /usr/sbin/nginx -c /etc/nginx/nginx.conf
nginx 55350.00.2968842960 ? S 06:20 0:00 nginx: worker process
root 55460.00.0 103256 828 pts/0 S+ 06:24 0:00 grep nginx
# ifconfig
eth0 Link encap:EthernetHWaddr 00:0C:29:8B:40:4A
inet addr:192.168.1.250Bcast:192.168.1.255Mask:255.255.255.0
inet6 addr: fe80::20c:29ff:fe8b:404a/64 Scope:Link
UP BROADCAST RUNNING MULTICASTMTU:1500Metric:1
RX packets:36106 errors:0 dropped:0 overruns:0 frame:0
TX packets:21435 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:39751462 (37.9 MiB)TX bytes:2077943 (1.9 MiB)
eth0:0 Link encap:EthernetHWaddr 00:0C:29:8B:40:4A
inet addr:192.168.1.13Bcast:192.168.1.255Mask:255.255.255.0
UP BROADCAST RUNNING MULTICASTMTU:1500Metric:1
lo Link encap:Local Loopback
inet addr:127.0.0.1Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNINGMTU:16436Metric:1
RX packets:31 errors:0 dropped:0 overruns:0 frame:0
TX packets:31 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:3005 (2.9 KiB)TX bytes:3005 (2.9 KiB)
删除防火墙再测试
# iptables -nvL
# iptables -D INPUT -p icmp -j DROP
页:
[1]