mysql MHA高可用测试
【环境介绍】
系统环境：Red Hat Enterprise Linux 7 + 5.7.18 + MHA version 0.57
【测试步骤：自动切换】
当前数据库状态：
|
系统 |
IP |
主机名 |
备注 |
版本 |
|
xx系统 |
192.168.142.111 |
mysqlmha1 |
主库 |
5.7.18 -log MySQL Community Server (GPL) |
|
192.168.142.112 |
mysqlmha2 |
备库(预主库) |
||
|
192.168.142.113 |
mysqlmha3 |
备库&MHA MGM |
||
|
192.168.142.111 |
mysqlmha1 |
VIP |
切换后数据库状态：
|
系统 |
IP |
主机名 |
备注 |
版本 |
|
xx系统 |
192.168.142.111 |
mysqlmha1 |
备库（修复后） |
5.7.18 -log MySQL Community Server (GPL) |
|
192.168.142.112 |
mysqlmha2 |
主库 |
||
|
192.168.142.113 |
mysqlmha3 |
备库&MHA MGM |
||
|
192.168.142.112 |
mysqlmha1 |
VIP |
确认当前数据库状态情况，每次检查或者操作前都需要确认当前数据库状态：
检查节点间的ssh互信状态是否正常，如果有报错，确认用户，用户互信及密码
$masterha_check_ssh --conf=/etc/masterha/app1.cnf
检查mysql主从复制是否正常，如果有报错，确认脚本文件权限是否准确，配置文件信息是否准确
$masterha_check_repl --conf=/etc/masterha/app1.cnf
在mysqlmha3管理节点启动MHA自动切换脚本，此脚本支持一次切换后则自动停止自己的监控进程：
部署masterha_manager.sh脚本
#!/bin/sh
then
echo "the user is not mha,please check !!!"
exit
fi
nohup /usr/bin/masterha_manager --conf=/etc/masterha/app1.cnf --ignore_last_failover < /dev/null > /var/log/masterha/app1/manager.log 2>&1 &
}
function stop(){
/usr/bin/masterha_stop --conf=/etc/masterha/app1.cnf
}
function status(){
/usr/bin/masterha_check_status --conf=/etc/masterha/app1.cnf
}
case "$1" in
start)
start
;;
stop)
stop
;;
status)
status
;;
*)
echo "Usage: $0 start|stop"
esac
[root@mysqlmha3 shell]#
[mha@mysqlmha3 shell]$ sh masterha_manager.sh start
[1]+ 完成 sh masterha_manager.sh start
[mha@mysqlmha3 shell]$
[mha@mysqlmha3 shell]$ ps -ef |grep masterha_manager
mha 3154 1 2 23:34 pts/2 00:00:00 perl /usr/bin/masterha_manager --conf=/etc/masterha/app1.cnf --ignore_last_failover
mha 3169 2977 0 23:34 pts/2 00:00:00 grep --color=auto masterha_manager
[mha@mysqlmha3 shell]$
在mysqlmha1节点上模拟数据DOWN掉：
[root@mysqlmha1 ~]# ps -ef|grep mysqld|awk '{print "kill -9 "$2}'|sh
sh: 第 2 行:kill: (3602) - 没有那个进程
[1]+ 已杀死 mysqld --defaults-file=/etc/mymha.cnf
[root@mysqlmha1 ~]#
查看mysqlmha3节点查看自动切换日志：
[root@mysqlmha3 app1]# cat manager.log
Wed May 16 23:34:34 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.
Wed May 16 23:34:34 2018 - [info] Reading application default configuration from /etc/masterha/app1.cnf..
Wed May 16 23:34:34 2018 - [info] Reading server configuration from /etc/masterha/app1.cnf..
Wed May 16 23:45:49 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.
Wed May 16 23:45:49 2018 - [info] Reading application default configuration from /etc/masterha/app1.cnf..
Wed May 16 23:45:49 2018 - [info] Reading server configuration from /etc/masterha/app1.cnf..
34:35 2018 - [info] 192.168.142.112(192.168.142.112:3306)
Wed May 16 23:34:35 2018 - [info] 192.168.142.113(192.168.142.113:3306)
Wed May 16 23:34:35 2018 - [info] Alive Slaves:
Wed May 16 23:34:35 2018 - [info] 192.168.142.112(192.168.142.112:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:34:35 2018 - [info] GTID ON
Wed May 16 23:34:35 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:34:35 2018 - [info] Primary candidate for the new Master (candidate_master is set)
Wed May 16 23:34:35 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:34:35 2018 - [info] GTID ON
Wed May 16 23:34:35 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:34:35 2018 - [info] Not candidate for the new Master (no_master is set)
Wed May 16 23:34:35 2018 - [info] Current Alive Master: 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:34:35 2018 - [info] Checking slave configurations..
Wed May 16 23:34:35 2018 - [info] read_only=1 is not set on slave 192.168.142.112(192.168.142.112:3306).
Wed May 16 23:34:35 2018 - [info] read_only=1 is not set on slave 192.168.142.113(192.168.142.113:3306).
Wed May 16 23:34:35 2018 - [info] Checking replication filtering settings..
Wed May 16 23:34:35 2018 - [info] binlog_do_db= , binlog_ignore_db=
Wed May 16 23:34:35 2018 - [info] Replication filtering check ok.
Wed May 16 23:34:35 2018 - [info] GTID (with auto-pos) is supported. Skipping all SSH and Node package checking.
Wed May 16 23:34:35 2018 - [info] Checking SSH publickey authentication settings on the current master..
Wed May 16 23:34:35 2018 - [info] HealthCheck: SSH to 192.168.142.111 is reachable.
Wed May 16 23:34:35 2018 - [info]
192.168.142.111(192.168.142.111:3306) (current master) ###当前数据库信息
+--192.168.142.112(192.168.142.112:3306)
+--192.168.142.113(192.168.142.113:3306)
Wed May 16 23:34:35 2018 - [info] Checking master_ip_failover_script status:
Wed May 16 23:34:35 2018 - [info] /usr/bin/master_ip_failover --command=status --ssh_user=mha --orig_master_host=192.168.142.111 --orig_master_ip=192.168.142.111 --orig_master_port=3306
IN SCRIPT TEST====sudo /sbin/ifconfig eno16777736:2 down==sudo /sbin/ifconfig eno16777736:2 192.168.142.114 netmask 255.255.255.0;/sbin/arping -I eno16777736 -c 3 -s 192.168.142.114 192.168.142.2 >/dev/null 2>&1===
Checking the Status of the script.. OK
Wed May 16 23:34:38 2018 - [info] OK.
Wed May 16 23:34:38 2018 - [warning] shutdown_script is not defined.
Wed May 16 23:34:38 2018 - [info] Set master ping interval 5 seconds.
Wed May 16 23:34:38 2018 - [info] Set secondary check script: /usr/bin/masterha_secondary_check -s 192.168.142.112 -s 192.168.142.113 --user=root --master_host=mysqlmha1 --master_ip=192.168.142.111 --master_port=3306
Wed May 16 23:34:38 2018 - [info] Starting ping health check on 192.168.142.111(192.168.142.111:3306)..
Wed May 16 23:34:38 2018 - [info] Ping(SELECT) succeeded, waiting until MySQL doesn't respond..
Wed May 16 23:43:53 2018 - [warning] Got timeout on MySQL Ping(SELECT) child process and killed it! at /usr/share/perl5/vendor_perl/MHA/HealthCheck.pm line 431.
Wed May 16 23:43:53 2018 - [info] Executing secondary network check script: /usr/bin/masterha_secondary_check -s 192.168.142.112 -s 192.168.142.113 --user=root --master_host=mysqlmha1 --master_ip=192.168.142.111 --master_port=3306 --user=mha --master_host=192.168.142.111 --master_ip=192.168.142.111 --master_port=3306 --master_user=mha --master_password=Mha_ahm%0118 --ping_type=SELECT
Wed May 16 23:43:53 2018 - [info] Executing SSH check script: exit 0
Wed May 16 23:43:53 2018 - [info] Ping(SELECT) succeeded, waiting until MySQL doesn't respond..
Wed May 16 23:43:53 2018 - [info] HealthCheck: SSH to 192.168.142.111 is reachable.
Master is reachable from 192.168.142.112!
Wed May 16 23:43:54 2018 - [warning] Master is reachable from at least one of other monitoring servers. Failover should not happen.
Wed May 16 23:45:33 2018 - [warning] Got error on MySQL select ping: 2006 (MySQL server has gone away)
Wed May 16 23:45:33 2018 - [info] Executing SSH check script: exit 0
Wed May 16 23:45:33 2018 - [info] Executing secondary network check script: /usr/bin/masterha_secondary_check -s 192.168.142.112 -s 192.168.142.113 --user=root --master_host=mysqlmha1 --master_ip=192.168.142.111 --master_port=3306 --user=mha --master_host=192.168.142.111 --master_ip=192.168.142.111 --master_port=3306 --master_user=mha --master_password=Mha_ahm%0118 --ping_type=SELECT
Wed May 16 23:45:33 2018 - [info] HealthCheck: SSH to 192.168.142.111 is reachable.
Monitoring server 192.168.142.112 is reachable, Master is not reachable from 192.168.142.112. OK.
Monitoring server 192.168.142.113 is reachable, Master is not reachable from 192.168.142.113. OK.
Wed May 16 23:45:34 2018 - [info] Master is not reachable from all other monitoring servers. Failover should start.
Wed May 16 23:45:38 2018 - [warning] Got error on MySQL connect: 2003 (Can't connect to MySQL server on '192.168.142.111' (111))
Wed May 16 23:45:38 2018 - [warning] Connection failed 2 time(s)..
Wed May 16 23:45:43 2018 - [warning] Got error on MySQL connect: 2003 (Can't connect to MySQL server on '192.168.142.111' (111))
Wed May 16 23:45:43 2018 - [warning] Connection failed 3 time(s)..
Wed May 16 23:45:48 2018 - [warning] Got error on MySQL connect: 2003 (Can't connect to MySQL server on '192.168.142.111' (111))
Wed May 16 23:45:48 2018 - [warning] Connection failed 4 time(s)..
Wed May 16 23:45:48 2018 - [warning] Master is not reachable from health checker!
Wed May 16 23:45:48 2018 - [warning] Master 192.168.142.111(192.168.142.111:3306) is not reachable!
Wed May 16 23:45:48 2018 - [warning] SSH is reachable.
Wed May 16 23:45:48 2018 - [info] Connecting to a master server failed. Reading configuration file /etc/masterha_default.cnf and /etc/masterha/app1.cnf again, and trying to connect to all servers to check server status..
Wed May 16 23:45:48 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.
Wed May 16 23:45:48 2018 - [info] Reading application default configuration from /etc/masterha/app1.cnf..
Wed May 16 23:45:48 2018 - [info] Reading server configuration from /etc/masterha/app1.cnf..
Wed May 16 23:45:49 2018 - [info] GTID failover mode = 1
Wed May 16 23:45:49 2018 - [info] Dead Servers:
Wed May 16 23:45:49 2018 - [info] 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:49 2018 - [info] Alive Servers:
Wed May 16 23:45:49 2018 - [info] 192.168.142.112(192.168.142.112:3306)
Wed May 16 23:45:49 2018 - [info] 192.168.142.113(192.168.142.113:3306)
Wed May 16 23:45:49 2018 - [info] Alive Slaves:
Wed May 16 23:45:49 2018 - [info] 192.168.142.112(192.168.142.112:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:49 2018 - [info] GTID ON
Wed May 16 23:45:49 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:49 2018 - [info] Primary candidate for the new Master (candidate_master is set)
Wed May 16 23:45:49 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:49 2018 - [info] GTID ON
Wed May 16 23:45:49 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:49 2018 - [info] Not candidate for the new Master (no_master is set)
Wed May 16 23:45:49 2018 - [info] Checking slave configurations..
Wed May 16 23:45:49 2018 - [info] read_only=1 is not set on slave 192.168.142.112(192.168.142.112:3306).
Wed May 16 23:45:49 2018 - [info] read_only=1 is not set on slave 192.168.142.113(192.168.142.113:3306).
Wed May 16 23:45:49 2018 - [info] Checking replication filtering settings..
Wed May 16 23:45:49 2018 - [info] Replication filtering check ok.
Wed May 16 23:45:49 2018 - [info] Master is down!
Wed May 16 23:45:49 2018 - [info] Terminating monitoring script.
Wed May 16 23:45:49 2018 - [info] Got exit code 20 (Master dead).
Wed May 16 23:45:49 2018 - [info] MHA::MasterFailover version 0.57.
Wed May 16 23:45:49 2018 - [info] Starting master failover.
Wed May 16 23:45:49 2018 - [info]
Wed May 16 23:45:49 2018 - [info] * Phase 1: Configuration Check Phase..
Wed May 16 23:45:49 2018 - [info]
Wed May 16 23:45:50 2018 - [info] GTID failover mode = 1
Wed May 16 23:45:50 2018 - [info] Dead Servers:
Wed May 16 23:45:50 2018 - [info] 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Checking master reachability via MySQL(double check)...
Wed May 16 23:45:50 2018 - [info] ok.
Wed May 16 23:45:50 2018 - [info] Alive Servers:
Wed May 16 23:45:50 2018 - [info] 192.168.142.112(192.168.142.112:3306)
Wed May 16 23:45:50 2018 - [info] 192.168.142.113(192.168.142.113:3306)
Wed May 16 23:45:50 2018 - [info] Alive Slaves:
Wed May 16 23:45:50 2018 - [info] 192.168.142.112(192.168.142.112:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:50 2018 - [info] GTID ON
Wed May 16 23:45:50 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Primary candidate for the new Master (candidate_master is set)
Wed May 16 23:45:50 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:50 2018 - [info] GTID ON
Wed May 16 23:45:50 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Not candidate for the new Master (no_master is set)
Wed May 16 23:45:50 2018 - [info] Starting GTID based failover.
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] ** Phase 1: Configuration Check Phase completed.
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] * Phase 2: Dead Master Shutdown Phase..
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] Forcing shutdown so that applications never connect to the current master..
Wed May 16 23:45:50 2018 - [info] Executing master IP deactivation script:
Wed May 16 23:45:50 2018 - [info] /usr/bin/master_ip_failover --orig_master_host=192.168.142.111 --orig_master_ip=192.168.142.111 --orig_master_port=3306 --command=stopssh --ssh_user=mha
IN SCRIPT TEST====sudo /sbin/ifconfig eno16777736:2 down==sudo /sbin/ifconfig eno16777736:2 192.168.142.114 netmask 255.255.255.0;/sbin/arping -I eno16777736 -c 3 -s 192.168.142.114 192.168.142.2 >/dev/null 2>&1===
Disabling the VIP on old master: 192.168.142.111
Wed May 16 23:45:50 2018 - [info] done.
Wed May 16 23:45:50 2018 - [warning] shutdown_script is not set. Skipping explicit shutting down of the dead master.
Wed May 16 23:45:50 2018 - [info] * Phase 2: Dead Master Shutdown Phase completed.
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] * Phase 3: Master Recovery Phase..
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] * Phase 3.1: Getting Latest Slaves Phase..
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] The latest binary log file/position on all slaves is binlog01.000009:246
Wed May 16 23:45:50 2018 - [info] Retrieved Gtid Set: 8d7abed9-d4cd-11e7-a165-000c29c913a2:6-7
Wed May 16 23:45:50 2018 - [info] Latest slaves (Slaves that received relay log files to the latest):
Wed May 16 23:45:50 2018 - [info] 192.168.142.112(192.168.142.112:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:50 2018 - [info] GTID ON
Wed May 16 23:45:50 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Primary candidate for the new Master (candidate_master is set)
Wed May 16 23:45:50 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:50 2018 - [info] GTID ON
Wed May 16 23:45:50 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Not candidate for the new Master (no_master is set)
Wed May 16 23:45:50 2018 - [info] The oldest binary log file/position on all slaves is binlog01.000009:246
Wed May 16 23:45:50 2018 - [info] Retrieved Gtid Set: 8d7abed9-d4cd-11e7-a165-000c29c913a2:6-7
Wed May 16 23:45:50 2018 - [info] Oldest slaves:
Wed May 16 23:45:50 2018 - [info] 192.168.142.112(192.168.142.112:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:50 2018 - [info] GTID ON
Wed May 16 23:45:50 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Primary candidate for the new Master (candidate_master is set)
Wed May 16 23:45:50 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:50 2018 - [info] GTID ON
Wed May 16 23:45:50 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Not candidate for the new Master (no_master is set)
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] * Phase 3.3: Determining New Master Phase..
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] Searching new master from slaves..
Wed May 16 23:45:50 2018 - [info] Candidate masters from the configuration file:
Wed May 16 23:45:50 2018 - [info] 192.168.142.112(192.168.142.112:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:50 2018 - [info] GTID ON
Wed May 16 23:45:50 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Primary candidate for the new Master (candidate_master is set)
Wed May 16 23:45:50 2018 - [info] Non-candidate masters:
Wed May 16 23:45:50 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:45:50 2018 - [info] GTID ON
Wed May 16 23:45:50 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:45:50 2018 - [info] Not candidate for the new Master (no_master is set)
Wed May 16 23:45:50 2018 - [info] Searching from candidate_master slaves which have received the latest relay log events..
Wed May 16 23:45:50 2018 - [info] New master is 192.168.142.112(192.168.142.112:3306)
Wed May 16 23:45:50 2018 - [info] Starting master failover..
Wed May 16 23:45:50 2018 - [info]
From:
192.168.142.111(192.168.142.111:3306) (current master)
+--192.168.142.112(192.168.142.112:3306)
+--192.168.142.113(192.168.142.113:3306)
To:
192.168.142.112(192.168.142.112:3306) (new master)
+--192.168.142.113(192.168.142.113:3306)
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] * Phase 3.3: New Master Recovery Phase..
Wed May 16 23:45:50 2018 - [info]
Wed May 16 23:45:50 2018 - [info] Waiting all logs to be applied..
Wed May 16 23:45:50 2018 - [info] done.
Wed May 16 23:45:50 2018 - [info] Getting new master's binlog name and position..
Wed May 16 23:45:50 2018 - [info] binlog01.000008:286
Wed May 16 23:45:50 2018 - [info] All other slaves should start replication from here. Statement should be: CHANGE MASTER TO MASTER_HOST='192.168.142.112', MASTER_PORT=3306, MASTER_AUTO_POSITION=1, MASTER_USER='repl', MASTER_PASSWORD='xxx';
Wed May 16 23:45:50 2018 - [info] Master Recovery succeeded. File:Pos:Exec_Gtid_Set: binlog01.000008, 286, 42f239e7-5908-11e8-8214-000c2926d694:1,
8d7abed9-d4cd-11e7-a165-000c29c913a2:1-7,
aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa:1-15:1000002-1000075
Wed May 16 23:45:50 2018 - [info] Executing master IP activate script:
Wed May 16 23:45:50 2018 - [info] /usr/bin/master_ip_failover --command=start --ssh_user=mha --orig_master_host=192.168.142.111 --orig_master_ip=192.168.142.111 --orig_master_port=3306 --new_master_host=192.168.142.112 --new_master_ip=192.168.142.112 --new_master_port=3306 --new_master_user='mha' --new_master_password=xxx
Unknown option: new_master_user
Unknown option: new_master_password
IN SCRIPT TEST====sudo /sbin/ifconfig eno16777736:2 down==sudo /sbin/ifconfig eno16777736:2 192.168.142.114 netmask 255.255.255.0;/sbin/arping -I eno16777736 -c 3 -s 192.168.142.114 192.168.142.2 >/dev/null 2>&1===
Enabling the VIP - 192.168.142.114 on the new master - 192.168.142.112
Wed May 16 23:45:54 2018 - [info] OK.
Wed May 16 23:45:54 2018 - [info] ** Finished master recovery successfully.
Wed May 16 23:45:54 2018 - [info] * Phase 3: Master Recovery Phase completed.
Wed May 16 23:45:54 2018 - [info]
Wed May 16 23:45:54 2018 - [info] * Phase 4: Slaves Recovery Phase..
Wed May 16 23:45:54 2018 - [info]
Wed May 16 23:45:54 2018 - [info]
Wed May 16 23:45:54 2018 - [info] * Phase 4.1: Starting Slaves in parallel..
Wed May 16 23:45:54 2018 - [info]
Wed May 16 23:45:54 2018 - [info] -- Slave recovery on host 192.168.142.113(192.168.142.113:3306) started, pid: 3382. Check tmp log /var/log/masterha/app1/192.168.142.113_3306_20180516234549.log if it takes time..
Wed May 16 23:45:55 2018 - [info]
Wed May 16 23:45:55 2018 - [info] Log messages from 192.168.142.113 ...
Wed May 16 23:45:55 2018 - [info]
Wed May 16 23:45:54 2018 - [info] Resetting slave 192.168.142.113(192.168.142.113:3306) and starting replication from the new master 192.168.142.112(192.168.142.112:3306)..
Wed May 16 23:45:54 2018 - [info] Executed CHANGE MASTER.
Wed May 16 23:45:54 2018 - [info] Slave started.
Wed May 16 23:45:54 2018 - [info] gtid_wait(42f239e7-5908-11e8-8214-000c2926d694:1,
8d7abed9-d4cd-11e7-a165-000c29c913a2:1-7,
aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa:1-15:1000002-1000075) completed on 192.168.142.113(192.168.142.113:3306). Executed 2 events.
Wed May 16 23:45:55 2018 - [info] End of log messages from 192.168.142.113.
Wed May 16 23:45:55 2018 - [info] -- Slave on host 192.168.142.113(192.168.142.113:3306) started.
Wed May 16 23:45:55 2018 - [info] All new slave servers recovered successfully.
Wed May 16 23:45:55 2018 - [info]
Wed May 16 23:45:55 2018 - [info] * Phase 5: New master cleanup phase..
Wed May 16 23:45:55 2018 - [info]
Wed May 16 23:45:55 2018 - [info] Resetting slave info on the new master..
Wed May 16 23:45:55 2018 - [info] 192.168.142.112: Resetting slave info succeeded.
Wed May 16 23:45:55 2018 - [info] Master failover to 192.168.142.112(192.168.142.112:3306) completed successfully.
Wed May 16 23:45:55 2018 - [info]
----- Failover Report -----
app1: MySQL Master failover 192.168.142.111(192.168.142.111:3306) to 192.168.142.112(192.168.142.112:3306) succeeded
Master 192.168.142.111(192.168.142.111:3306) is down!
Check MHA Manager logs at mysqlmha3:/var/log/masterha/app1/manager.log for details.
Started automated(non-interactive) failover.
Invalidated master IP address on 192.168.142.111(192.168.142.111:3306)
Selected 192.168.142.112(192.168.142.112:3306) as a new master.
192.168.142.112(192.168.142.112:3306): OK: Applying all logs succeeded.
192.168.142.112(192.168.142.112:3306): OK: Activated master IP address.
192.168.142.113(192.168.142.113:3306): OK: Slave started, replicating from 192.168.142.112(192.168.142.112:3306)
192.168.142.112(192.168.142.112:3306): Resetting slave info succeeded.
Master failover to 192.168.142.112(192.168.142.112:3306) completed successfully.
[root@mysqlmha3 app1]#
在节点2上面检查VIP是否漂移：
[root@mysqlmha2 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
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: eno16777736: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:26:d6:94 brd ff:ff:ff:ff:ff:ff
inet 192.168.142.112/24 brd 192.168.142.255 scope global eno16777736
valid_lft forever preferred_lft forever
inet 192.168.142.114/24 brd 192.168.142.255 scope global secondary eno16777736:2
valid_lft forever preferred_lft forever
inet6 fe80::20c:29ff:fe26:d694/64 scope link
valid_lft forever preferred_lft forever
[root@mysqlmha2 ~]#
查看mysqlmha3节点查看当前集群状态：
[mha@mysqlmha3 shell]$ masterha_check_repl --conf=/etc/masterha/app1.cnf
Wed May 16 23:55:37 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.
Wed May 16 23:55:37 2018 - [info] Reading application default configuration from /etc/masterha/app1.cnf..
Wed May 16 23:55:37 2018 - [info] Reading server configuration from /etc/masterha/app1.cnf..
Wed May 16 23:55:37 2018 - [info] MHA::MasterMonitor version 0.57.
Wed May 16 23:55:38 2018 - [info] GTID failover mode = 1
Wed May 16 23:55:38 2018 - [info] Dead Servers:
Wed May 16 23:55:38 2018 - [info] 192.168.142.111(192.168.142.111:3306)
Wed May 16 23:55:38 2018 - [info] Alive Servers:
Wed May 16 23:55:38 2018 - [info] 192.168.142.112(192.168.142.112:3306)
Wed May 16 23:55:38 2018 - [info] 192.168.142.113(192.168.142.113:3306)
Wed May 16 23:55:38 2018 - [info] Alive Slaves:
Wed May 16 23:55:38 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Wed May 16 23:55:38 2018 - [info] GTID ON
Wed May 16 23:55:38 2018 - [info] Replicating from 192.168.142.112(192.168.142.112:3306)
Wed May 16 23:55:38 2018 - [info] Not candidate for the new Master (no_master is set)
Wed May 16 23:55:38 2018 - [info] Current Alive Master: 192.168.142.112(192.168.142.112:3306)
Wed May 16 23:55:38 2018 - [info] Checking slave configurations..
Wed May 16 23:55:38 2018 - [info] read_only=1 is not set on slave 192.168.142.113(192.168.142.113:3306).
Wed May 16 23:55:38 2018 - [info] Checking replication filtering settings..
Wed May 16 23:55:38 2018 - [info] binlog_do_db= , binlog_ignore_db=
Wed May 16 23:55:38 2018 - [info] Replication filtering check ok.
Wed May 16 23:55:38 2018 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln364] None of slaves can be master. Check failover configuration file or log-bin settings in my.cnf
Wed May 16 23:55:38 2018 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln427] Error happened on checking configurations. at /bin/masterha_check_repl line 48.
Wed May 16 23:55:38 2018 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln525] Error happened on monitoring servers.
Wed May 16 23:55:38 2018 - [info] Got exit code 1 (Not master dead).
MySQL Replication Health is NOT OK! ###这里输出为正常，因为节点一没有修复
[mha@mysqlmha3 shell]$
修复节mysqlmha1节点数据库为备库：
查看在自动切换脚本时记录了change master脚本，在节点一执行时需要把密码加入。
[root@mysqlmha3 app1]# cd /var/log/masterha/app1
[root@mysqlmha3 app1]# grep -i change manager.log
Wed May 16 23:45:50 2018 - [info] All other slaves should start replication from here. Statement should be: CHANGE MASTER TO MASTER_HOST='192.168.142.112', MASTER_PORT=3306, MASTER_AUTO_POSITION=1, MASTER_USER='repl', MASTER_PASSWORD='xxx';
Wed May 16 23:45:54 2018 - [info] Executed CHANGE MASTER.
[root@mysqlmha3 app1]#
启动mysqlmha1节点数据库，确认数据库启动没有报错后往后操作
[root@mysqlmha1 ~]# mysqld --defaults-file=/etc/mymha.cnf &
[1] 3636
[root@mysqlmha1 ~]#
使mysqlmha1节点为备库
[root@mysqlmha1 ~]# mysql -uroot -p -P3306 --protocol=tcp
Enter password:
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 4
Server version: 5.7.18-log MySQL Community Server (GPL)
Copyright (c) 2000, 2017, Oracle and/or its affiliates. All rights reserved.
Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.
Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.
mysql> CHANGE MASTER TO MASTER_HOST='192.168.142.112', MASTER_PORT=3306, MASTER_AUTO_POSITION=1, MASTER_USER='repl', MASTER_PASSWORD='repl';
Query OK, 0 rows affected, 2 warnings (0.36 sec)
mysql> start slave;
Query OK, 0 rows affected (0.01 sec)
mysql> show slave status\G
*************************** 1. row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 192.168.142.112
Master_User: repl
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: binlog01.000008
Read_Master_Log_Pos: 286
Relay_Log_File: relaylog01.000004
Relay_Log_Pos: 393
Relay_Master_Log_File: binlog01.000008
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: 0
Last_Error:
Skip_Counter: 0
Exec_Master_Log_Pos: 286
Relay_Log_Space: 669
Until_Condition: None
Until_Log_File:
Until_Log_Pos: 0
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno: 0
Last_IO_Error:
Last_SQL_Errno: 0
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id: 112
Master_UUID: 42f239e7-5908-11e8-8214-000c2926d694
Master_Info_File: mysql.slave_master_info
SQL_Delay: 0
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
Master_Retry_Count: 86400
Master_Bind:
Last_IO_Error_Timestamp:
Last_SQL_Error_Timestamp:
Master_SSL_Crl:
Master_SSL_Crlpath:
Retrieved_Gtid_Set: 42f239e7-5908-11e8-8214-000c2926d694:1
Executed_Gtid_Set: 42f239e7-5908-11e8-8214-000c2926d694:1,
8d7abed9-d4cd-11e7-a165-000c29c913a2:1-7,
aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa:1-15:1000002-1000075
Auto_Position: 1
Replicate_Rewrite_DB:
Channel_Name:
Master_TLS_Version:
1 row in set (0.00 sec)
mysql> exit
Bye
[root@mysqlmha1 ~]#
在MHA 管理节点检查数据库当前状态：
[mha@mysqlmha3 shell]$ masterha_check_repl --conf=/etc/masterha/app1.cnf
Thu May 17 00:02:39 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.
Thu May 17 00:02:39 2018 - [info] Reading application default configuration from /etc/masterha/app1.cnf..
Thu May 17 00:02:39 2018 - [info] Reading server configuration from /etc/masterha/app1.cnf..
Thu May 17 00:02:39 2018 - [info] MHA::MasterMonitor version 0.57.
Thu May 17 00:02:40 2018 - [info] GTID failover mode = 1
Thu May 17 00:02:40 2018 - [info] Dead Servers:
Thu May 17 00:02:40 2018 - [info] Alive Servers:
Thu May 17 00:02:40 2018 - [info] 192.168.142.111(192.168.142.111:3306) ###已经识别到节点恢复
Thu May 17 00:02:40 2018 - [info] 192.168.142.112(192.168.142.112:3306)
Thu May 17 00:02:40 2018 - [info] 192.168.142.113(192.168.142.113:3306)
Thu May 17 00:02:40 2018 - [info] Alive Slaves:
Thu May 17 00:02:40 2018 - [info] 192.168.142.111(192.168.142.111:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Thu May 17 00:02:40 2018 - [info] GTID ON
Thu May 17 00:02:40 2018 - [info] Replicating from 192.168.142.112(192.168.142.112:3306)
Thu May 17 00:02:40 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Thu May 17 00:02:40 2018 - [info] GTID ON
Thu May 17 00:02:40 2018 - [info] Replicating from 192.168.142.112(192.168.142.112:3306)
Thu May 17 00:02:40 2018 - [info] Not candidate for the new Master (no_master is set)
Thu May 17 00:02:40 2018 - [info] Current Alive Master: 192.168.142.112(192.168.142.112:3306) ###当前主库已经切换至节点2
Thu May 17 00:02:40 2018 - [info] Checking slave configurations..
Thu May 17 00:02:40 2018 - [info] read_only=1 is not set on slave 192.168.142.111(192.168.142.111:3306). ###节点1已经成为备库
Thu May 17 00:02:40 2018 - [info] read_only=1 is not set on slave 192.168.142.113(192.168.142.113:3306). ###节点3没有改边，仍然是备库
Thu May 17 00:02:40 2018 - [info] Checking replication filtering settings..
Thu May 17 00:02:40 2018 - [info] binlog_do_db= , binlog_ignore_db=
Thu May 17 00:02:40 2018 - [info] Replication filtering check ok.
Thu May 17 00:02:40 2018 - [info] GTID (with auto-pos) is supported. Skipping all SSH and Node package checking.
Thu May 17 00:02:40 2018 - [info] Checking SSH publickey authentication settings on the current master..
Thu May 17 00:02:40 2018 - [info] HealthCheck: SSH to 192.168.142.112 is reachable.
Thu May 17 00:02:40 2018 - [info]
192.168.142.112(192.168.142.112:3306) (current master)
+--192.168.142.111(192.168.142.111:3306)
+--192.168.142.113(192.168.142.113:3306)
Thu May 17 00:02:40 2018 - [info] Checking replication health on 192.168.142.111..
Thu May 17 00:02:40 2018 - [info] ok.
Thu May 17 00:02:40 2018 - [info] Checking replication health on 192.168.142.113..
Thu May 17 00:02:40 2018 - [info] ok.
Thu May 17 00:02:40 2018 - [info] Checking master_ip_failover_script status:
Thu May 17 00:02:40 2018 - [info] /usr/bin/master_ip_failover --command=status --ssh_user=mha --orig_master_host=192.168.142.112 --orig_master_ip=192.168.142.112 --orig_master_port=3306
IN SCRIPT TEST====sudo /sbin/ifconfig eno16777736:2 down==sudo /sbin/ifconfig eno16777736:2 192.168.142.114 netmask 255.255.255.0;/sbin/arping -I eno16777736 -c 3 -s 192.168.142.114 192.168.142.2 >/dev/null 2>&1===
Checking the Status of the script.. OK
Thu May 17 00:02:44 2018 - [info] OK.
Thu May 17 00:02:44 2018 - [warning] shutdown_script is not defined.
Thu May 17 00:02:44 2018 - [info] Got exit code 0 (Not master dead).
MySQL Replication Health is OK.
[mha@mysqlmha3 shell]$
【测试步骤：手工切换】
当前数据库状态：
|
系统 |
IP |
主机名 |
备注 |
版本 |
|
xx系统 |
192.168.142.111 |
mysqlmha1 |
备库 |
5.7.18 -log MySQL Community Server (GPL) |
|
192.168.142.112 |
mysqlmha2 |
主库 |
||
|
192.168.142.113 |
mysqlmha3 |
备库&MHA MGM |
||
|
192.168.142.112 |
mysqlmha1 |
VIP |
切换后数据库状态：
|
系统 |
IP |
主机名 |
备注 |
版本 |
|
xx系统 |
192.168.142.111 |
mysqlmha1 |
主库 |
5.7.18 -log MySQL Community Server (GPL) |
|
192.168.142.112 |
mysqlmha2 |
备库(预主库) |
||
|
192.168.142.113 |
mysqlmha3 |
备库&MHA MGM |
||
|
192.168.142.111 |
mysqlmha1 |
VIP |
确认当前数据库状态情况，每次检查或者操作前都需要确认当前数据库状态：
检查节点间的ssh互信状态是否正常，如果有报错，确认用户，用户互信及密码
$masterha_check_ssh --conf=/etc/masterha/app1.cnf
检查mysql主从复制是否正常，如果有报错，确认脚本文件权限是否准确，配置文件信息是否准确
$masterha_check_repl --conf=/etc/masterha/app1.cnf
在mysqlmha3管理节点查看人工切换脚本，信息一定要准确
[mha@mysqlmha3 shell]$ cat masterha_switch.sh
#!/bin/sh
conf="/etc/masterha/app1.cnf"
dead_master="192.168.142.112" ###当前master的IP节点
new_master="192.168.142.111" ###需要切换目标的IP节点
port="3306"
user=$(whoami)
if [ $user != "mha" ]
then
echo "the user is not mha,please check !!!"
exit
fi
function manual_failover(){
##人工切换需要停掉manager进程
/usr/bin/masterha_stop --conf=/etc/masterha/app1.cnf
/usr/bin/masterha_master_switch --conf=${conf} --master_state=dead --dead_master_host=${dead_master} --dead_master_port=${port} --new_master_host=${new_master} --new_master_port=${port} --ignore_last_failover
}
function manual_online_change(){
##人工切换需要停掉manager进程
/usr/bin/masterha_stop --conf=/etc/masterha/app1.cnf
/usr/bin/masterha_master_switch --conf=${conf} --master_state=alive --new_master_host=${new_master} --new_master_port=${port} --orig_master_is_new_slave --running_updates_limit=10000 --interactive=0 < /dev/null > /var/log/masterha/app1/manual_online_change.log 2>&1 &
}
case "$1" in
manual_failover)
manual_failover
;;
manual_online_change)
manual_online_change
;;
*)
echo "Usage: $0 manual_failover|manual_online_change"
esac
[mha@mysqlmha3 shell]$
手工执行切换脚本，没有报错即可
[mha@mysqlmha3 shell]$ sh masterha_switch.sh manual_online_change
MHA Manager is not running on app1(2:NOT_RUNNING). ###此输出为正常，因为当前没有启动自动切换脚本进程，如果有启动的话，也会停止自动切换进程
查看手工切换日志
[mha@mysqlmha3 app1]$ cat manual_online_change.log
Thu May 17 00:22:39 2018 - [info] MHA::MasterRotate version 0.57.
Thu May 17 00:22:39 2018 - [info] Starting online master switch..
Thu May 17 00:22:39 2018 - [info]
Thu May 17 00:22:39 2018 - [info] * Phase 1: Configuration Check Phase..
Thu May 17 00:22:39 2018 - [info]
Thu May 17 00:22:39 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.
Thu May 17 00:22:39 2018 - [info] Reading application default configuration from /etc/masterha/app1.cnf..
Thu May 17 00:22:39 2018 - [info] Reading server configuration from /etc/masterha/app1.cnf..
Thu May 17 00:22:40 2018 - [info] GTID failover mode = 1
Thu May 17 00:22:40 2018 - [info] Current Alive Master: 192.168.142.112(192.168.142.112:3306)
Thu May 17 00:22:40 2018 - [info] Alive Slaves:
Thu May 17 00:22:40 2018 - [info] 192.168.142.111(192.168.142.111:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Thu May 17 00:22:40 2018 - [info] GTID ON
Thu May 17 00:22:40 2018 - [info] Replicating from 192.168.142.112(192.168.142.112:3306)
Thu May 17 00:22:40 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Thu May 17 00:22:40 2018 - [info] GTID ON
Thu May 17 00:22:40 2018 - [info] Replicating from 192.168.142.112(192.168.142.112:3306)
Thu May 17 00:22:40 2018 - [info] Not candidate for the new Master (no_master is set)
Thu May 17 00:22:40 2018 - [info] Executing FLUSH NO_WRITE_TO_BINLOG TABLES. This may take long time..
Thu May 17 00:22:40 2018 - [info] ok.
Thu May 17 00:22:40 2018 - [info] Checking MHA is not monitoring or doing failover..
Thu May 17 00:22:40 2018 - [info] Checking replication health on 192.168.142.111..
Thu May 17 00:22:40 2018 - [info] ok.
Thu May 17 00:22:40 2018 - [info] Checking replication health on 192.168.142.113..
Thu May 17 00:22:40 2018 - [info] ok.
Thu May 17 00:22:40 2018 - [info] 192.168.142.111 can be new master.
Thu May 17 00:22:40 2018 - [info]
From:
192.168.142.112(192.168.142.112:3306) (current master)
+--192.168.142.111(192.168.142.111:3306)
+--192.168.142.113(192.168.142.113:3306)
To:
192.168.142.111(192.168.142.111:3306) (new master)
+--192.168.142.113(192.168.142.113:3306)
+--192.168.142.112(192.168.142.112:3306) ###跟需求一致
Thu May 17 00:22:40 2018 - [info] Checking whether 192.168.142.111(192.168.142.111:3306) is ok for the new master..
Thu May 17 00:22:40 2018 - [info] ok.
Thu May 17 00:22:40 2018 - [info] 192.168.142.112(192.168.142.112:3306): SHOW SLAVE STATUS returned empty result. To check replication filtering rules, temporarily executing CHANGE MASTER to a dummy host.
Thu May 17 00:22:40 2018 - [info] 192.168.142.112(192.168.142.112:3306): Resetting slave pointing to the dummy host.
Thu May 17 00:22:40 2018 - [info] ** Phase 1: Configuration Check Phase completed.
Thu May 17 00:22:40 2018 - [info]
Thu May 17 00:22:40 2018 - [info] * Phase 2: Rejecting updates Phase..
Thu May 17 00:22:40 2018 - [info]
Thu May 17 00:22:40 2018 - [info] Executing master ip online change script to disable write on the current master:
Thu May 17 00:22:40 2018 - [info] /usr/bin/master_ip_online_change --command=stop --orig_master_host=192.168.142.112 --orig_master_ip=192.168.142.112 --orig_master_port=3306 --orig_master_user='mha' --new_master_host=192.168.142.111 --new_master_ip=192.168.142.111 --new_master_port=3306 --new_master_user='mha' --orig_master_ssh_user=mha --new_master_ssh_user=mha --orig_master_is_new_slave --orig_master_password=xxx --new_master_password=xxx
Thu May 17 00:22:40 2018 694778 Set read_only on the new master.. ok.
Thu May 17 00:22:40 2018 699533 Waiting all running 2 threads are disconnected.. (max 1500 milliseconds)
{'Time' => '939','db' => undef,'Id' => '61','User' => 'repl','State' => 'Master has sent all binlog to slave; waiting for more updates','Command' => 'Binlog Dump GTID','Info' => undef,'Host' => 'mysqlmha3:34819'}
{'Time' => '939','db' => undef,'Id' => '62','User' => 'repl','State' => 'Master has sent all binlog to slave; waiting for more updates','Command' => 'Binlog Dump GTID','Info' => undef,'Host' => 'mysqlmha1:47539'}
Thu May 17 00:22:41 2018 202982 Waiting all running 2 threads are disconnected.. (max 1000 milliseconds)
{'Time' => '940','db' => undef,'Id' => '61','User' => 'repl','State' => 'Master has sent all binlog to slave; waiting for more updates','Command' => 'Binlog Dump GTID','Info' => undef,'Host' => 'mysqlmha3:34819'}
{'Time' => '940','db' => undef,'Id' => '62','User' => 'repl','State' => 'Master has sent all binlog to slave; waiting for more updates','Command' => 'Binlog Dump GTID','Info' => undef,'Host' => 'mysqlmha1:47539'}
Thu May 17 00:22:41 2018 706427 Waiting all running 2 threads are disconnected.. (max 500 milliseconds)
{'Time' => '940','db' => undef,'Id' => '61','User' => 'repl','State' => 'Master has sent all binlog to slave; waiting for more updates','Command' => 'Binlog Dump GTID','Info' => undef,'Host' => 'mysqlmha3:34819'}
{'Time' => '940','db' => undef,'Id' => '62','User' => 'repl','State' => 'Master has sent all binlog to slave; waiting for more updates','Command' => 'Binlog Dump GTID','Info' => undef,'Host' => 'mysqlmha1:47539'}
Thu May 17 00:22:42 2018 210670 Set read_only=1 on the orig master.. ok.
Thu May 17 00:22:42 2018 212115 Waiting all running 2 queries are disconnected.. (max 500 milliseconds)
{'Time' => '941','db' => undef,'Id' => '61','User' => 'repl','State' => 'Master has sent all binlog to slave; waiting for more updates','Command' => 'Binlog Dump GTID','Info' => undef,'Host' => 'mysqlmha3:34819'}
{'Time' => '941','db' => undef,'Id' => '62','User' => 'repl','State' => 'Master has sent all binlog to slave; waiting for more updates','Command' => 'Binlog Dump GTID','Info' => undef,'Host' => 'mysqlmha1:47539'}
Thu May 17 00:22:42 2018 714337 Killing all application threads..
Thu May 17 00:22:42 2018 716387 done.
Disabling the VIP on old master: 192.168.142.112
Thu May 17 00:22:42 2018 - [info] ok.
Thu May 17 00:22:42 2018 - [info] Locking all tables on the orig master to reject updates from everybody (including root):
Thu May 17 00:22:42 2018 - [info] Executing FLUSH TABLES WITH READ LOCK..
Thu May 17 00:22:42 2018 - [info] ok.
Thu May 17 00:22:42 2018 - [info] Orig master binlog:pos is binlog01.000008:286.
Thu May 17 00:22:42 2018 - [info] Waiting to execute all relay logs on 192.168.142.111(192.168.142.111:3306)..
Thu May 17 00:22:42 2018 - [info] master_pos_wait(binlog01.000008:286) completed on 192.168.142.111(192.168.142.111:3306). Executed 0 events.
Thu May 17 00:22:42 2018 - [info] done.
Thu May 17 00:22:42 2018 - [info] Getting new master's binlog name and position..
Thu May 17 00:22:42 2018 - [info] binlog01.000011:529
Thu May 17 00:22:42 2018 - [info] All other slaves should start replication from here. Statement should be: CHANGE MASTER TO MASTER_HOST='192.168.142.111', MASTER_PORT=3306, MASTER_AUTO_POSITION=1, MASTER_USER='repl', MASTER_PASSWORD='xxx';
Thu May 17 00:22:42 2018 - [info] Executing master ip online change script to allow write on the new master:
Thu May 17 00:22:42 2018 - [info] /usr/bin/master_ip_online_change --command=start --orig_master_host=192.168.142.112 --orig_master_ip=192.168.142.112 --orig_master_port=3306 --orig_master_user='mha' --new_master_host=192.168.142.111 --new_master_ip=192.168.142.111 --new_master_port=3306 --new_master_user='mha' --orig_master_ssh_user=mha --new_master_ssh_user=mha --orig_master_is_new_slave --orig_master_password=xxx --new_master_password=xxx
Thu May 17 00:22:42 2018 907496 Set read_only=0 on the new master.
Enabling the VIP - 192.168.142.114 on the new master - 192.168.142.111
Thu May 17 00:22:46 2018 - [info] ok.
Thu May 17 00:22:46 2018 - [info]
Thu May 17 00:22:46 2018 - [info] * Switching slaves in parallel..
Thu May 17 00:22:46 2018 - [info]
Thu May 17 00:22:46 2018 - [info] -- Slave switch on host 192.168.142.113(192.168.142.113:3306) started, pid: 3645
Thu May 17 00:22:46 2018 - [info]
Thu May 17 00:22:47 2018 - [info] Log messages from 192.168.142.113 ...
Thu May 17 00:22:47 2018 - [info]
Thu May 17 00:22:46 2018 - [info] Waiting to execute all relay logs on 192.168.142.113(192.168.142.113:3306)..
Thu May 17 00:22:46 2018 - [info] master_pos_wait(binlog01.000008:286) completed on 192.168.142.113(192.168.142.113:3306). Executed 0 events.
Thu May 17 00:22:46 2018 - [info] done.
Thu May 17 00:22:46 2018 - [info] Resetting slave 192.168.142.113(192.168.142.113:3306) and starting replication from the new master 192.168.142.111(192.168.142.111:3306)..
Thu May 17 00:22:46 2018 - [info] Executed CHANGE MASTER.
Thu May 17 00:22:46 2018 - [info] Slave started.
Thu May 17 00:22:47 2018 - [info] End of log messages from 192.168.142.113 ...
Thu May 17 00:22:47 2018 - [info]
Thu May 17 00:22:47 2018 - [info] -- Slave switch on host 192.168.142.113(192.168.142.113:3306) succeeded.
Thu May 17 00:22:47 2018 - [info] Unlocking all tables on the orig master:
Thu May 17 00:22:47 2018 - [info] Executing UNLOCK TABLES..
Thu May 17 00:22:47 2018 - [info] ok.
Thu May 17 00:22:47 2018 - [info] Starting orig master as a new slave..
Thu May 17 00:22:47 2018 - [info] Resetting slave 192.168.142.112(192.168.142.112:3306) and starting replication from the new master 192.168.142.111(192.168.142.111:3306)..
Thu May 17 00:22:47 2018 - [info] Executed CHANGE MASTER.
Thu May 17 00:22:47 2018 - [info] Slave started.
Thu May 17 00:22:47 2018 - [info] All new slave servers switched successfully.
Thu May 17 00:22:47 2018 - [info]
Thu May 17 00:22:47 2018 - [info] * Phase 5: New master cleanup phase..
Thu May 17 00:22:47 2018 - [info]
Thu May 17 00:22:47 2018 - [info] 192.168.142.111: Resetting slave info succeeded.
Thu May 17 00:22:47 2018 - [info] Switching master to 192.168.142.111(192.168.142.111:3306) completed successfully.
[mha@mysqlmha3 app1]$
使用脚本检查当前数据库节点信息
[mha@mysqlmha3 shell]$ masterha_check_repl --conf=/etc/masterha/app1.cnf
Thu May 17 00:24:01 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.
Thu May 17 00:24:01 2018 - [info] Reading application default configuration from /etc/masterha/app1.cnf..
Thu May 17 00:24:01 2018 - [info] Reading server configuration from /etc/masterha/app1.cnf..
Thu May 17 00:24:01 2018 - [info] MHA::MasterMonitor version 0.57.
Thu May 17 00:24:02 2018 - [info] GTID failover mode = 1
Thu May 17 00:24:02 2018 - [info] Dead Servers:
Thu May 17 00:24:02 2018 - [info] Alive Servers:
Thu May 17 00:24:02 2018 - [info] 192.168.142.111(192.168.142.111:3306)
Thu May 17 00:24:02 2018 - [info] 192.168.142.112(192.168.142.112:3306)
Thu May 17 00:24:02 2018 - [info] 192.168.142.113(192.168.142.113:3306)
Thu May 17 00:24:02 2018 - [info] Alive Slaves:
Thu May 17 00:24:02 2018 - [info] 192.168.142.112(192.168.142.112:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Thu May 17 00:24:02 2018 - [info] GTID ON
Thu May 17 00:24:02 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Thu May 17 00:24:02 2018 - [info] Primary candidate for the new Master (candidate_master is set)
Thu May 17 00:24:02 2018 - [info] 192.168.142.113(192.168.142.113:3306) Version=5.7.18-log (oldest major version between slaves) log-bin:enabled
Thu May 17 00:24:02 2018 - [info] GTID ON
Thu May 17 00:24:02 2018 - [info] Replicating from 192.168.142.111(192.168.142.111:3306)
Thu May 17 00:24:02 2018 - [info] Not candidate for the new Master (no_master is set)
Thu May 17 00:24:02 2018 - [info] Current Alive Master: 192.168.142.111(192.168.142.111:3306)
Thu May 17 00:24:02 2018 - [info] Checking slave configurations..
Thu May 17 00:24:02 2018 - [info] read_only=1 is not set on slave 192.168.142.113(192.168.142.113:3306).
Thu May 17 00:24:02 2018 - [info] Checking replication filtering settings..
Thu May 17 00:24:02 2018 - [info] binlog_do_db= , binlog_ignore_db=
Thu May 17 00:24:02 2018 - [info] Replication filtering check ok.
Thu May 17 00:24:02 2018 - [info] GTID (with auto-pos) is supported. Skipping all SSH and Node package checking.
Thu May 17 00:24:02 2018 - [info] Checking SSH publickey authentication settings on the current master..
Thu May 17 00:24:03 2018 - [info] HealthCheck: SSH to 192.168.142.111 is reachable.
Thu May 17 00:24:03 2018 - [info]
192.168.142.111(192.168.142.111:3306) (current master) ###已经识别到切换后的主从关系信息
+--192.168.142.112(192.168.142.112:3306)
+--192.168.142.113(192.168.142.113:3306)
Thu May 17 00:24:03 2018 - [info] Checking replication health on 192.168.142.112..
Thu May 17 00:24:03 2018 - [info] ok.
Thu May 17 00:24:03 2018 - [info] Checking replication health on 192.168.142.113..
Thu May 17 00:24:03 2018 - [info] ok.
Thu May 17 00:24:03 2018 - [info] Checking master_ip_failover_script status:
Thu May 17 00:24:03 2018 - [info] /usr/bin/master_ip_failover --command=status --ssh_user=mha --orig_master_host=192.168.142.111 --orig_master_ip=192.168.142.111 --orig_master_port=3306
IN SCRIPT TEST====sudo /sbin/ifconfig eno16777736:2 down==sudo /sbin/ifconfig eno16777736:2 192.168.142.114 netmask 255.255.255.0;/sbin/arping -I eno16777736 -c 3 -s 192.168.142.114 192.168.142.2 >/dev/null 2>&1===
Checking the Status of the script.. OK
Thu May 17 00:24:06 2018 - [info] OK.
Thu May 17 00:24:06 2018 - [warning] shutdown_script is not defined.
Thu May 17 00:24:06 2018 - [info] Got exit code 0 (Not master dead).
MySQL Replication Health is OK.
[mha@mysqlmha3 shell]$
查看mysqlmha节点1VIP信息
[root@mysqlmha1 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
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: eno16777736: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:96:18:d2 brd ff:ff:ff:ff:ff:ff
inet 192.168.142.111/24 brd 192.168.142.255 scope global eno16777736
valid_lft forever preferred_lft forever
inet 192.168.142.114/24 brd 192.168.142.255 scope global secondary eno16777736:2
valid_lft forever preferred_lft forever
inet6 fe80::20c:29ff:fe96:18d2/64 scope link
valid_lft forever preferred_lft forever
[root@mysqlmha1 ~]#
【总结】
MHA的VIP和其他备库节点，可根据业务需求适当的来分配读写分离，过多的配置可能会导致出现问题时业务的各种接口报错，延迟了恢复业务时间。
自动切换和人工切换要根据需求来定，虽然部署了自动切换的进程脚本，但是主机和数据库层面也是需要部署监控，在MHA自动维护或者正常状态前发现问题，避免影响业务。
自动切换，有可能因为网络或者主机原因导致误判主节点状态，导致误切换。但是时效性较高。
人工切换，人工切换比较灵活，满足容灾备份演练及节点维护需求。但是存在时效性低的情况。
