MHA的安裝搭建和測試

這篇文章主要講解了“MHA的安裝搭建和測試”，文中的講解內容簡(jiǎn)單清晰，易于學(xué)習與理解，下面請大家跟著(zhù)小編的思路慢慢深入，一起來(lái)研究和學(xué)習“MHA的安裝搭建和測試”吧！

環(huán)境：

192.168.56.221     MHA1

192.168.56.222     MHA2

192.168.56.223     MHA3

192.168.56.224     MHA4

#221為master 222位slave 223為slave 224為管理節點(diǎn)

##1 安裝MHA軟件包

https://code.google.com/p/-master-ha/

https://downloads.mariadb.com/MHA/

yum install -y perl-DBD-MySQL

yum install -y perl cpan

yum install -y perl-DBD-MySQL perl-Config-Tiny perl-Log-Dispatch perl-Parallel-ForkManager

常用rpm搜索網(wǎng)站：

http://rpm.pbone.net/  

perl-Log-Dispatch

perl-Parallel-ForkManager

[root@MHA1 ~]# rpm -ivh perl-Mail-Sender-0.8.16-1.el6.rf.noarch.rpm

warning: perl-Mail-Sender-0.8.16-1.el6.rf.noarch.rpm: Header V3 DSA/SHA1 Signature, key ID 6b8d79e6: NOKEY

Preparing...                ########################################### [100%]

   1:perl-Mail-Sender       ########################################### [100%]

[root@MHA1 ~]# rpm -ivh perl-Mail-Sendmail-0.79_16-4.2.noarch.rpm

warning: perl-Mail-Sendmail-0.79_16-4.2.noarch.rpm: Header V3 DSA/SHA1 Signature, key ID 04b8b50a: NOKEY

Preparing...                ########################################### [100%]

   1:perl-Mail-Sendmail     ########################################### [100%]

[root@MHA1 ~]# yum localinstall perl-Log-Dispatch-2.27-1.el6.noarch.rpm

[root@MHA1 ~]# yum localinstall perl-Parallel-ForkManager-0.7.5-2.2.el6.rf.noarch.rpm

[root@MHA1 ~]# yum localinstall mha4mysql-node-0.54-0.el6.noarch.rpm   #所有節點(diǎn)安裝

[root@MHA1 ~]# yum localinstall mha4mysql-manager-0.55-0.el6.noarch.rpm  #管理節點(diǎn)安裝

##2 主從同步配置：

#主

mysql> show variables like 'log_bin';

+---------------+-------+

| Variable_name | Value |

+---------------+-------+

| log_bin       | ON    |

+---------------+-------+

1 row in set (0.53 sec)

mysql> grant replication slave on *.* to 'rep'@'%' identified by '123456';   #建議所有mysql服務(wù)器配置，以防主從切換

Query OK, 0 rows affected, 1 warning (0.17 sec)

mysql> flush privileges;

Query OK, 0 rows affected (0.01 sec)

#從服務(wù)器

vi my.cnf

[mysqld]

#slave

relay_log=/usr/local/mysql/logs/relay-log.log

master-info-repository=TABLE

relay-log-info-repository=TABLE

sync_master_info = 1    

sync_relay_log = 1   

sync_relay_log_info = 1

mysql> change master to master_host='192.168.56.221', master_port=3306, master_user='rep', master_password='123456', master_log_file='ray-bin.000003', master_log_pos=740;

Query OK, 0 rows affected, 2 warnings (0.05 sec)

mysql> start slave;

Query OK, 0 rows affected (0.03 sec)

mysql> show slave status\G

*************************** 1. row ***************************

               Slave_IO_State: Waiting for master to send event

                  Master_Host: 192.168.56.221

                  Master_User: rep

                  Master_Port: 3306

                Connect_Retry: 60

              Master_Log_File: ray-bin.000003

          Read_Master_Log_Pos: 740

               Relay_Log_File: relay-log.000003

                Relay_Log_Pos: 318

        Relay_Master_Log_File: ray-bin.000003

             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: 740

              Relay_Log_Space: 519

              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: 21

                  Master_UUID: 262ff2d5-2264-11e7-a823-080027a847bd

             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:

            Executed_Gtid_Set:

                Auto_Position: 0

         Replicate_Rewrite_DB:

                 Channel_Name:

           Master_TLS_Version:

1 row in set (0.00 sec)

mysql> use mysql

Database changed

mysql> alter table slave_master_info engine=innodb;

Query OK, 0 rows affected (0.23 sec)

Records: 0  Duplicates: 0  Warnings: 0

mysql> alter table slave_relay_log_info engine=innodb;

Query OK, 0 rows affected (0.08 sec)

Records: 0  Duplicates: 0  Warnings: 0

mysql> alter table slave_worker_info engine=innodb;

Query OK, 0 rows affected (0.08 sec)

Records: 0  Duplicates: 0  Warnings: 0

##3 配置互信

[root@MHA1 ~]# ssh-keygen -t rsa  #所有節點(diǎn)

Generating public/private rsa key pair.

Enter file in which to save the key (/root/.ssh/id_rsa):

Enter passphrase (empty for no passphrase):

Enter same passphrase again:

Your identification has been saved in /root/.ssh/id_rsa.

Your public key has been saved in /root/.ssh/id_rsa.pub.

The key fingerprint is:

e2:1c:6a:a3:ae:e5:38:0e:e7:cd:21:50:a5:14:d2:3b root@MHA1

The key's randomart image is:

+--[ RSA 2048]----+

|..o..            |

| o.o             |

|  o.             |

| .E              |

|.  .  o S        |

| .   + o         |

|. + = o          |

|o* * o           |

|+== o            |

+-----------------+

[root@MHA1 ~]# ssh slave1 cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys

[root@MHA1 ~]# ssh slave2 cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys

[root@MHA1 ~]# ssh master cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys

[root@MHA1 ~]# ssh MHAManager cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys

[root@MHA1 ~]# scp /root/.ssh/authorized_keys slave1:/root/.ssh/authorized_keys

[root@MHA1 ~]# scp /root/.ssh/authorized_keys slave2:/root/.ssh/authorized_keys

[root@MHA1 ~]# scp /root/.ssh/authorized_keys MHAManager:/root/.ssh/authorized_keys

[root@MHA1 ~]# ssh slave1 date;ssh slave2 date;ssh master date;ssh MHAManager date;

Fri Apr 28 10:46:10 CST 2017

Fri Apr 28 10:46:14 CST 2017

Fri Apr 28 10:46:12 CST 2017

Fri Apr 28 10:46:13 CST 2017

[root@MHA1 ~]#

[root@MHA4 ~]# ssh slave1 date;ssh slave2 date;ssh master date;ssh MHAManager date;

Fri Apr 28 10:46:10 CST 2017

Fri Apr 28 10:46:14 CST 2017

Fri Apr 28 10:46:12 CST 2017

Fri Apr 28 10:46:13 CST 2017

[root@MHA4 ~]#

##4 slave服務(wù)器上設置只讀權限，并關(guān)閉自動(dòng)清除執行完畢的中繼日志(relay log)

[mysqld]

read_only = 1

relay_log_purge = 0

mysql -uroot -p123456 -e "set global read_only=on;"

注意：

    設置只讀是防止人在slave服務(wù)器上誤寫(xiě)入，保證主從數據一致

    關(guān)閉自動(dòng)清除執行完畢的中級日志是：假如一臺從庫沒(méi)有接受完主庫的binlog，那么可以通過(guò)MHA把最新的slave上的中級日志發(fā)送到最老的slave上，識別差異中級日志并補齊數據

    如果想數據庫自動(dòng)刪除執行完畢的日志，可以通過(guò)crontab實(shí)現，例如想要每天5點(diǎn)刪除中繼日志

    0 5 * * * /usr/local/mysql/bin/mysql -uroot -p123456 -e "set global relay_log_purge=1;flush logs;set global relay_log_purge=0;flush logs;"

##5 配置MHA監控、管理服務(wù)

#創(chuàng )建mha管理賬號

mysql> grant all privileges on *.* to 'mhaadmin'@'%' identified by '123456';

Query OK, 0 rows affected, 1 warning (0.54 sec)

mysql> flush privileges;

Query OK, 0 rows affected (0.23 sec)

#編輯MHA配置文件

############

#manager_workdir和remote_workdir用來(lái)定義MHA管理目錄，以便存放日志。

#user和password為MHA的管理賬號密碼

#repl_user和repl_password為復制的賬號密碼

#如果你的環(huán)境是鏈式復制架構，那么試著(zhù)multi_tier_slave=1

#ping_interval=1每個(gè)一秒檢測一次

#ping_type、master_ip_failover_script和master_ip_online_change_script定義故障切換和在線(xiàn)切換

#secondary_check_script是為了防止網(wǎng)絡(luò )抖動(dòng)誤切換，造成數據不一致。原理：當監控節點(diǎn)無(wú)法連接mysql主庫是，會(huì )視圖從其他從服務(wù)器去ping和連接mysql主庫，只有雙方都連接失敗，才認定mysql主庫宕機，如果有以防可以連接mysql主庫，則不發(fā)生切換

#ignore_fail=0缺省情況下，當某個(gè)Slave故障時(shí)（比如：不能通過(guò)MySQL/SSH連接，SQL線(xiàn)程因錯誤停止等），MHA不開(kāi)啟故障切換；若設置為1，則對應的那個(gè)Slave出現故障時(shí)自動(dòng)切換。

#candidate_master用于設置某個(gè)Slave是否可以?xún)?yōu)先成為Master。若設置為1，則對應的那個(gè)Slave可優(yōu)先成為新的Master；若多個(gè)Slave均設置該參數為1，則成為Master的優(yōu)先級按照[server_1]/[server_2]/…排序；

#master_binlog_dir=  MySQL主庫的binlog日志文件的全路徑，缺省為/var/lib/mysql，根據實(shí)際情況設置為真實(shí)的路徑。該參數適用于這么一種情況：Master實(shí)例故障，但OS系統運行正常，此時(shí)MHA可通過(guò)SSH登錄，然后讀取并拷貝必須的二進(jìn)制日志事件（即差異的日志）。#可見(jiàn)該參數是必要而有用的，因為Master死掉后，MHA無(wú)法自動(dòng)獲取binlog日志文件的路徑。另外，可用逗號隔開(kāi)設置多個(gè)路徑。

#check_repl_delay 缺省情況下，如果某個(gè)Slave延遲超過(guò)100MB的relay logs，MHA則不會(huì )選擇該Slave成為新的Master，因為這需要太長(cháng)的時(shí)間來(lái)恢復。若設置該參數為0，MHA在選擇新的Master時(shí)，會(huì )忽略復制延遲。

############

vi /etc/mha.cnf

[server default]

user=mhaadmin

password=123456

manager_workdir=/usr/local/mha

manager_log=/usr/local/mha/manager.log

remote_workdir=/usr/local/mha

ssh_user=root

ssh_port=22

repl_user=rep

repl_password=123456

ping_interval=1

ping_type=CONNECT

master_ip_failover_script=/usr/local/mha/scripts/master_ip_failover

master_ip_online_change_script=/usr/local/mha/scripts/master_ip_online_change

secondary_check_script=/usr/bin/masterha_secondary_check -s 192.168.56.222 -s 192.168.56.223 -s 192.168.56.224 --user=root --master_host=master --master_ip=192.168.56.221 --master_port=3306

[server1]

ignore_fail=0

check_repl_delay=1

hostname=MHA1

ip=192.168.56.221

ssh_port=22

port=3306

master_binlog_dir=/usr/local/mysql/logs/

candidate_master=1

[server2]

ignore_fail=0

check_repl_delay=1

hostname=MHA2

ip=192.168.56.222

ssh_port=22

port=3306

master_binlog_dir=/usr/local/mysql/logs/

candidate_master=1

[server3]

ignore_fail=0

check_repl_delay=1

hostname=MHA3

ip=192.168.56.223

ssh_port=22

port=3306

master_binlog_dir=/usr/local/mysql/logs/

candidate_master=1

##使用工具檢測ssh

[root@MHA4 logs]# masterha_check_ssh --conf=/etc/mha.cnf

Fri Apr 28 15:11:14 2017 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.

Fri Apr 28 15:11:14 2017 - [info] Reading application default configurations from /etc/mha.cnf..

Fri Apr 28 15:11:14 2017 - [info] Reading server configurations from /etc/mha.cnf..

Fri Apr 28 15:11:14 2017 - [info] Starting SSH connection tests..

Fri Apr 28 15:11:17 2017 - [debug]

Fri Apr 28 15:11:15 2017 - [debug]  Connecting via SSH from root@master(192.168.56.221:22) to root@slave1(192.168.56.222:22)..

Fri Apr 28 15:11:15 2017 - [debug]   ok.

Fri Apr 28 15:11:15 2017 - [debug]  Connecting via SSH from root@master(192.168.56.221:22) to root@slave2(192.168.56.223:22)..

Fri Apr 28 15:11:17 2017 - [debug]   ok.

Fri Apr 28 15:11:17 2017 - [debug]

Fri Apr 28 15:11:15 2017 - [debug]  Connecting via SSH from root@slave1(192.168.56.222:22) to root@master(192.168.56.221:22)..

Fri Apr 28 15:11:16 2017 - [debug]   ok.

Fri Apr 28 15:11:16 2017 - [debug]  Connecting via SSH from root@slave1(192.168.56.222:22) to root@slave2(192.168.56.223:22)..

Fri Apr 28 15:11:17 2017 - [debug]   ok.

Fri Apr 28 15:11:17 2017 - [debug]

Fri Apr 28 15:11:16 2017 - [debug]  Connecting via SSH from root@slave2(192.168.56.223:22) to root@master(192.168.56.221:22)..

Fri Apr 28 15:11:17 2017 - [debug]   ok.

Fri Apr 28 15:11:17 2017 - [debug]  Connecting via SSH from root@slave2(192.168.56.223:22) to root@slave1(192.168.56.222:22)..

Fri Apr 28 15:11:17 2017 - [debug]   ok.

Fri Apr 28 15:11:17 2017 - [info] All SSH connection tests passed successfully.

[root@MHA4 ~]# masterha_check_repl --conf=/etc/mha.cnf

Fri Apr 28 16:12:44 2017 - [info]   /usr/local/mha/scripts/master_ip_failover --command=status --ssh_user=root --orig_master_host=MHA1 --orig_master_ip=192.168.56.221 --orig_master_port=3306

Bareword "FIXME_xxx" not allowed while "strict subs" in use at /usr/local/mha/scripts/master_ip_failover line 93.

Execution of /usr/local/mha/scripts/master_ip_failover aborted due to compilation errors.

Fri Apr 28 16:12:44 2017 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln216]  Failed to get master_ip_failover_script status with return code 255:0.

Fri Apr 28 16:12:44 2017 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln386] Error happend on checking configurations.  at /usr/bin/masterha_check_repl line 48

Fri Apr 28 16:12:44 2017 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln482] Error happened on monitoring servers.

Fri Apr 28 16:12:44 2017 - [info] Got exit code 1 (Not master dead).

MySQL Replication Health is NOT OK!

解決方案：

[root@MHA4 ~]# grep master_ip_failover /etc/mha.cnf

#master_ip_failover_script=/usr/local/mha/scripts/master_ip_failover

#再次檢查主從復制

[root@MHA4 ~]# masterha_check_repl --conf=/etc/mha.cnf

Fri Apr 28 16:21:55 2017 - [info] Slaves settings check done.

Fri Apr 28 16:21:55 2017 - [info]

MHA1 (current master)

+--MHA2

+--MHA3

Fri Apr 28 16:21:55 2017 - [info] Checking replication health on MHA2..

Fri Apr 28 16:21:55 2017 - [info]  ok.

Fri Apr 28 16:21:55 2017 - [info] Checking replication health on MHA3..

Fri Apr 28 16:21:55 2017 - [info]  ok.

Fri Apr 28 16:21:55 2017 - [warning] master_ip_failover_script is not defined.

Fri Apr 28 16:21:55 2017 - [warning] shutdown_script is not defined.

Fri Apr 28 16:21:55 2017 - [info] Got exit code 0 (Not master dead).

MySQL Replication Health is OK.

#檢查mha  Manager狀態(tài)

[root@MHA4 ~]# masterha_check_status --conf=/etc/mha.cnf

mha is stopped(2:NOT_RUNNING).

注意：如果正常，會(huì )顯示"PING_OK"，否則會(huì )顯示"NOT_RUNNING"，這代表MHA監控沒(méi)有開(kāi)啟。

#開(kāi)啟Manager服務(wù)

[root@MHA4 ~]# nohup masterha_manager --conf=/etc/mha.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /usr/local/mha/manager.log 2>&1 &

[1] 5073

啟動(dòng)參數介紹：

--remove_dead_master_conf      該參數代表當發(fā)生主從切換后，老的主庫的ip將會(huì )從配置文件中移除。

--manger_log                            日志存放位置

--ignore_last_failover                 在缺省情況下，如果MHA檢測到連續發(fā)生宕機，且兩次宕機間隔不足8小時(shí)的話(huà)，則不會(huì )進(jìn)行Failover，之所以這樣限制是為了避免ping-pong效應。該參數代表忽略上次MHA觸發(fā)切換產(chǎn)生的文件，默認情況下，MHA發(fā)生切換后會(huì )在日志目錄，也就是上面我設置的/data產(chǎn)生app1.failover.complete文件，下次再次切換的時(shí)候如果發(fā)現該目錄下存在該文件將不允許觸發(fā)切換，除非在第一次切換后收到刪除該文件，為了方便，這里設置為--ignore_last_failover。

[root@MHA4 ~]# masterha_check_status --conf=/etc/mha.cnf

mha (pid:5073) is running(0:PING_OK), master:MHA1

[root@MHA4 ~]# tail -f /usr/local/mha/manager.log

##停止MHA Manager服務(wù)

[root@MHA4 ~]# masterha_stop --conf=/etc/mha.cnf

Stopped mha successfully.

[1]+  Exit 1                  nohup masterha_manager --conf=/etc/mha.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /usr/local/mha/manager.log 2>&1

[root@MHA4 ~]#

## 6 MHA 管理和切換VIP

[root@MHA1 ~]# /sbin/ifconfig eth2:1 192.168.56.225

[root@MHA1 ~]# ifconfig

eth2      Link encap:Ethernet  HWaddr 08:00:27:8B:0A:CE  

          inet addr:192.168.56.221  Bcast:192.168.56.255  Mask:255.255.255.0

          inet6 addr: fe80::a00:27ff:fe8b:ace/64 Scope:Link

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

          RX packets:3348 errors:0 dropped:0 overruns:0 frame:0

          TX packets:2758 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:1000

          RX bytes:316425 (309.0 KiB)  TX bytes:292848 (285.9 KiB)

eth2:1    Link encap:Ethernet  HWaddr 08:00:27:8B:0A:CE  

          inet addr:192.168.56.225  Bcast:192.168.56.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

lo        Link encap:Local Loopback  

          inet addr:127.0.0.1  Mask:255.0.0.0

          inet6 addr: ::1/128 Scope:Host

          UP LOOPBACK RUNNING  MTU:16436  Metric:1

          RX packets:4 errors:0 dropped:0 overruns:0 frame:0

          TX packets:4 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:0

          RX bytes:240 (240.0 b)  TX bytes:240 (240.0 b)

#編輯master_ip_failover_script腳本

##7 測試MHA 故障切換和日志追趕

#主服務(wù)器

[root@MHA1 ~]# pkill -9 mysqld

[root@MHA4 ~]# tail -f /usr/local/mha/manager.log

......

----- Failover Report -----

mha: MySQL Master failover MHA1 to MHA2 succeeded

Master MHA1 is down!

Check MHA Manager logs at MHA4:/usr/local/mha/manager.log for details.

Started automated(non-interactive) failover.

Invalidated master IP address on MHA1.

The latest slave MHA2(192.168.56.222:3306) has all relay logs for recovery.

Selected MHA2 as a new master.

MHA2: OK: Applying all logs succeeded.

MHA2: OK: Activated master IP address.

MHA3: This host has the latest relay log events.

Generating relay diff files from the latest slave succeeded.

MHA3: OK: Applying all logs succeeded. Slave started, replicating from MHA2.

MHA2: Resetting slave info succeeded.

Master failover to MHA2(192.168.56.222:3306) completed successfully.

切換完成后，MHA的manager服務(wù)就會(huì )停止，這是需要把MHA的配置文件中的server選項調整，把主服務(wù)器放在server1的內容下。把secondary_check_script內容修改。

[root@MHA4 ~]# vi /etc/mha.cnf

[server default]

manager_log=/usr/local/mha/manager.log

manager_workdir=/usr/local/mha

master_ip_failover_script=/usr/local/mha/scripts/master_ip_failover

master_ip_online_change_script=/usr/local/mha/scripts/master_ip_online_change

password=123456

ping_interval=1

ping_type=CONNECT

remote_workdir=/usr/local/mha

repl_password=123456

repl_user=rep

secondary_check_script=/usr/bin/masterha_secondary_check -s 192.168.56.221 -s 192.168.56.223 -s 192.168.56.224 --user=root --master_host=MHA2 --master_ip=192.168.56.222--master_port=3306

ssh_port=22

ssh_user=root

user=mhaadmin

[server1]

candidate_master=1

check_repl_delay=1

hostname=MHA2

ignore_fail=0

ip=192.168.56.222

master_binlog_dir=/usr/local/mysql/logs/

port=3306

ssh_port=22

[server2]

candidate_master=1

check_repl_delay=1

hostname=MHA1

ignore_fail=0

ip=192.168.56.221

master_binlog_dir=/usr/local/mysql/logs/

port=3306

ssh_port=22

[server3]

candidate_master=1

check_repl_delay=1

hostname=MHA3

ignore_fail=0

ip=192.168.56.223

master_binlog_dir=/usr/local/mysql/logs/

port=3306

ssh_port=22

#修改當原主庫的配置文件把slave參數開(kāi)放，修改當前主庫配置文件把slave參數注釋掉，重啟mysql服務(wù)器

#slave

#relay_log=/usr/local/mysql/logs/relay-log.log

#master-info-repository=TABLE

#relay-log-info-repository=TABLE

#sync_master_info = 1

#sync_relay_log = 1

#sync_relay_log_info = 1

#read_only = 1

relay_log_purge = 0

[root@MHA2 ~]# /etc/init.d/mysqld restart                              

Shutting down MySQL............                            [  OK  ]

Starting MySQL.                                                  [  OK  ]

#檢查MHA的配置文件和啟動(dòng)MHA服務(wù)

[root@MHA4 ~]# masterha_check_ssh --conf=/etc/mha.cnf

Wed May  3 13:38:32 2017 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.

Wed May  3 13:38:32 2017 - [info] Reading application default configurations from /etc/mha.cnf..

Wed May  3 13:38:32 2017 - [info] Reading server configurations from /etc/mha.cnf..

Wed May  3 13:38:32 2017 - [info] Starting SSH connection tests..

Wed May  3 13:38:33 2017 - [debug]

Wed May  3 13:38:32 2017 - [debug]  Connecting via SSH from root@MHA2(192.168.56.221:22) to root@MHA2(192.168.56.222:22)..

Wed May  3 13:38:33 2017 - [debug]   ok.

Wed May  3 13:38:33 2017 - [debug]  Connecting via SSH from root@MHA2(192.168.56.221:22) to root@MHA3(192.168.56.223:22)..

Wed May  3 13:38:33 2017 - [debug]   ok.

Wed May  3 13:38:33 2017 - [debug]

Wed May  3 13:38:33 2017 - [debug]  Connecting via SSH from root@MHA3(192.168.56.223:22) to root@MHA2(192.168.56.222:22)..

Wed May  3 13:38:33 2017 - [debug]   ok.

Wed May  3 13:38:33 2017 - [debug]  Connecting via SSH from root@MHA3(192.168.56.223:22) to root@MHA2(192.168.56.221:22)..

Wed May  3 13:38:33 2017 - [debug]   ok.

Wed May  3 13:38:33 2017 - [info] All SSH connection tests passed successfully.

[root@MHA4 ~]# masterha_check_repl --conf=/etc/mha.cnf

Wed May  3 13:38:56 2017 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.

Wed May  3 13:38:56 2017 - [info] Reading application default configurations from /etc/mha.cnf..

Wed May  3 13:38:56 2017 - [info] Reading server configurations from /etc/mha.cnf..

Wed May  3 13:38:56 2017 - [info] MHA::MasterMonitor version 0.55.

Wed May  3 13:38:56 2017 - [error][/usr/share/perl5/vendor_perl/MHA/Server.pm, ln241]  Getting relay log directory or current relay logfile from replication table failed on MHA2(192.168.56.222:3306)!

Wed May  3 13:38:56 2017 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln386] Error happend on checking configurations.  at /usr/share/perl5/vendor_perl/MHA/ServerManager.pm line 269

Wed May  3 13:38:56 2017 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln482] Error happened on monitoring servers.

Wed May  3 13:38:56 2017 - [info] Got exit code 1 (Not master dead).

MySQL Replication Health is NOT OK!

注意主庫的slave參數relay-log等參數已經(jīng)開(kāi)啟，所以檢查不過(guò)，注釋掉重啟就可以了