温馨提示×

温馨提示×

您好,登录后才能下订单哦!

密码登录×
登录注册×
其他方式登录
点击 登录注册 即表示同意《亿速云用户服务条款》

MySQL MHA switchover在线切换过程

发布时间:2020-08-08 00:01:41 来源:网络 阅读:445 作者:xingzhehxiang 栏目:MySQL数据库

1、切换过程,注意需要关闭masterha_manager

[root@es3 ~]#  masterha_master_switch --master_state=alive --conf=/root/app1.cnf --new_master_host=es3 --orig_master_is_new_slave
Tue Aug 20 10:12:02 2019 - [info] MHA::MasterRotate version 0.58.
Tue Aug 20 10:12:02 2019 - [info] Starting online master switch..
Tue Aug 20 10:12:02 2019 - [info] 
Tue Aug 20 10:12:02 2019 - [info] * Phase 1: Configuration Check Phase..
Tue Aug 20 10:12:02 2019 - [info] 
Tue Aug 20 10:12:02 2019 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.
Tue Aug 20 10:12:02 2019 - [info] Reading application default configuration from /root/app1.cnf..
Tue Aug 20 10:12:02 2019 - [info] Reading server configuration from /root/app1.cnf..
Tue Aug 20 10:12:03 2019 - [info] GTID failover mode = 1
Tue Aug 20 10:12:03 2019 - [info] Current Alive Master: es1(192.168.56.14:3306)
Tue Aug 20 10:12:03 2019 - [info] Alive Slaves:
Tue Aug 20 10:12:03 2019 - [info]   es2(192.168.56.15:3306)  Version=5.7.24-log (oldest major version between slaves) log-bin:enabled
Tue Aug 20 10:12:03 2019 - [info]     GTID ON
Tue Aug 20 10:12:03 2019 - [info]     Replicating from 192.168.56.14(192.168.56.14:3306)
Tue Aug 20 10:12:03 2019 - [info]   es3(192.168.56.16:3306)  Version=5.7.24-log (oldest major version between slaves) log-bin:enabled
Tue Aug 20 10:12:03 2019 - [info]     GTID ON
Tue Aug 20 10:12:03 2019 - [info]     Replicating from es1(192.168.56.14:3306)

It is better to execute FLUSH NO_WRITE_TO_BINLOG TABLES on the master before switching. Is it ok to execute on es1(192.168.56.14:3306)? (YES/no): yes
Tue Aug 20 10:12:06 2019 - [info] Executing FLUSH NO_WRITE_TO_BINLOG TABLES. This may take long time..
Tue Aug 20 10:12:06 2019 - [info]  ok.
Tue Aug 20 10:12:06 2019 - [info] Checking MHA is not monitoring or doing failover..
Tue Aug 20 10:12:06 2019 - [info] Checking replication health on es2..
Tue Aug 20 10:12:06 2019 - [info]  ok.
Tue Aug 20 10:12:06 2019 - [info] Checking replication health on es3..
Tue Aug 20 10:12:06 2019 - [info]  ok.
Tue Aug 20 10:12:06 2019 - [info] es3 can be new master.
Tue Aug 20 10:12:06 2019 - [info] 
From:
es1(192.168.56.14:3306) (current master)
 +--es2(192.168.56.15:3306)
 +--es3(192.168.56.16:3306)

To:
es3(192.168.56.16:3306) (new master)
 +--es2(192.168.56.15:3306)
 +--es1(192.168.56.14:3306)

Starting master switch from es1(192.168.56.14:3306) to es3(192.168.56.16:3306)? (yes/NO): yes
Tue Aug 20 10:12:11 2019 - [info] Checking whether es3(192.168.56.16:3306) is ok for the new master..
Tue Aug 20 10:12:11 2019 - [info]  ok.
Tue Aug 20 10:12:11 2019 - [info] es1(192.168.56.14:3306): SHOW SLAVE STATUS returned empty result. To check replication filtering rules, temporarily executing CHANGE MASTER to a dummy host.
Tue Aug 20 10:12:11 2019 - [info] es1(192.168.56.14:3306): Resetting slave pointing to the dummy host.
Tue Aug 20 10:12:11 2019 - [info] ** Phase 1: Configuration Check Phase completed.
Tue Aug 20 10:12:11 2019 - [info] 
Tue Aug 20 10:12:11 2019 - [info] * Phase 2: Rejecting updates Phase..
Tue Aug 20 10:12:11 2019 - [info] 
Tue Aug 20 10:12:11 2019 - [info] Executing master ip online change script to disable write on the current master:
Tue Aug 20 10:12:11 2019 - [info]   /usr/local/bin/master_ip_online_change --command=stop --orig_master_host=es1 --orig_master_ip=192.168.56.14 --orig_master_port=3306 --orig_master_user='repl' --new_master_host=es3 --new_master_ip=192.168.56.16 --new_master_port=3306 --new_master_user='repl' --orig_master_ssh_user=root --new_master_ssh_user=root   --orig_master_is_new_slave --orig_master_password=xxx --new_master_password=xxx



***************************************************************
Disabling the VIP - 192.168.56.191/24 on old master: es1
***************************************************************



Tue Aug 20 10:12:12 2019 - [info]  ok.
Tue Aug 20 10:12:12 2019 - [info] Locking all tables on the orig master to reject updates from everybody (including root):
Tue Aug 20 10:12:12 2019 - [info] Executing FLUSH TABLES WITH READ LOCK..
Tue Aug 20 10:12:12 2019 - [info]  ok.
Tue Aug 20 10:12:12 2019 - [info] Orig master binlog:pos is mysqlbin.000005:194.
Tue Aug 20 10:12:12 2019 - [info]  Waiting to execute all relay logs on es3(192.168.56.16:3306)..
Tue Aug 20 10:12:12 2019 - [info]  master_pos_wait(mysqlbin.000005:194) completed on es3(192.168.56.16:3306). Executed 0 events.
Tue Aug 20 10:12:12 2019 - [info]   done.
Tue Aug 20 10:12:12 2019 - [info] Getting new master's binlog name and position..
Tue Aug 20 10:12:12 2019 - [info]  mysqlbin.000004:154
Tue Aug 20 10:12:12 2019 - [info]  All other slaves should start replication from here. Statement should be: CHANGE MASTER TO MASTER_HOST='es3 or 192.168.56.16', MASTER_PORT=3306, MASTER_AUTO_POSITION=1, MASTER_USER='repl', MASTER_PASSWORD='xxx';
Tue Aug 20 10:12:12 2019 - [info] Executing master ip online change script to allow write on the new master:
Tue Aug 20 10:12:12 2019 - [info]   /usr/local/bin/master_ip_online_change --command=start --orig_master_host=es1 --orig_master_ip=192.168.56.14 --orig_master_port=3306 --orig_master_user='repl' --new_master_host=es3 --new_master_ip=192.168.56.16 --new_master_port=3306 --new_master_user='repl' --orig_master_ssh_user=root --new_master_ssh_user=root   --orig_master_is_new_slave --orig_master_password=xxx --new_master_password=xxx



***************************************************************
Enabling the VIP - 192.168.56.191/24 on new master: es3 
***************************************************************



Tue Aug 20 10:12:13 2019 - [info]  ok.
Tue Aug 20 10:12:13 2019 - [info] 
Tue Aug 20 10:12:13 2019 - [info] * Switching slaves in parallel..
Tue Aug 20 10:12:13 2019 - [info] 
Tue Aug 20 10:12:13 2019 - [info] -- Slave switch on host es2(192.168.56.15:3306) started, pid: 4076
Tue Aug 20 10:12:13 2019 - [info] 
Tue Aug 20 10:12:14 2019 - [info] Log messages from es2 ...
Tue Aug 20 10:12:14 2019 - [info] 
Tue Aug 20 10:12:13 2019 - [info]  Waiting to execute all relay logs on es2(192.168.56.15:3306)..
Tue Aug 20 10:12:13 2019 - [info]  master_pos_wait(mysqlbin.000005:194) completed on es2(192.168.56.15:3306). Executed 0 events.
Tue Aug 20 10:12:13 2019 - [info]   done.
Tue Aug 20 10:12:13 2019 - [info]  Resetting slave es2(192.168.56.15:3306) and starting replication from the new master es3(192.168.56.16:3306)..
Tue Aug 20 10:12:13 2019 - [info]  Executed CHANGE MASTER.
Tue Aug 20 10:12:13 2019 - [info]  Slave started.
Tue Aug 20 10:12:14 2019 - [info] End of log messages from es2 ...
Tue Aug 20 10:12:14 2019 - [info] 
Tue Aug 20 10:12:14 2019 - [info] -- Slave switch on host es2(192.168.56.15:3306) succeeded.
Tue Aug 20 10:12:14 2019 - [info] Unlocking all tables on the orig master:
Tue Aug 20 10:12:14 2019 - [info] Executing UNLOCK TABLES..
Tue Aug 20 10:12:14 2019 - [info]  ok.
Tue Aug 20 10:12:14 2019 - [info] Starting orig master as a new slave..
Tue Aug 20 10:12:14 2019 - [info]  Resetting slave es1(192.168.56.14:3306) and starting replication from the new master es3(192.168.56.16:3306)..
Tue Aug 20 10:12:14 2019 - [info]  Executed CHANGE MASTER.
Tue Aug 20 10:12:14 2019 - [info]  Slave started.
Tue Aug 20 10:12:14 2019 - [info] All new slave servers switched successfully.
Tue Aug 20 10:12:14 2019 - [info] 
Tue Aug 20 10:12:14 2019 - [info] * Phase 5: New master cleanup phase..
Tue Aug 20 10:12:14 2019 - [info] 
Tue Aug 20 10:12:14 2019 - [info]  es3: Resetting slave info succeeded.
Tue Aug 20 10:12:14 2019 - [info] Switching master to es3(192.168.56.16:3306) completed successfully.
[root@es3 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 08:00:27:ed:8b:51 brd ff:ff:ff:ff:ff:ff
    inet 10.0.2.15/24 brd 10.0.2.255 scope global noprefixroute dynamic enp0s3
       valid_lft 84197sec preferred_lft 84197sec
    inet6 fe80::c377:b5dd:2937:f269/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
3: enp0s8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 08:00:27:1f:f9:c8 brd ff:ff:ff:ff:ff:ff
    inet 192.168.56.16/24 brd 192.168.56.255 scope global noprefixroute dynamic enp0s8
       valid_lft 1030sec preferred_lft 1030sec
    inet 192.168.56.191/24 brd 192.168.56.255 scope global secondary enp0s8:1
       valid_lft forever preferred_lft forever
    inet6 fe80::90e0:253d:ab32:8ba4/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
[root@es3 ~]#


向AI问一下细节

免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。

AI