Percona-mysql MHA高可用实战方案

Posted 2020-06-17

tags:

篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了Percona-mysql MHA高可用实战方案相关的知识，希望对你有一定的参考价值。

前言

MHA（Master High Availability）目前在mysql高可用方面是一个相对成熟的解决方案，它由日本DeNA公司youshimaton(现就职于Facebook公司)开发，是一套优秀的作为MySQLy高可用环境下故障切换和主从提升的高可用软件。在MySQL故障切换过程中，MHA能做到在0~30秒之内自动完成数据库的故障切换操作，并且在进行故障切换过程中，MHA能在最大程度上保证数据的一致性，以达到真正意义上的高可用。

它由两部分组成：MHA Manager(管理节点)和MHA Node(数据节点)。MHA Manager可以单独部署在一台独立的机器上管理多个master-slave集群，也可以部署在一台slave上。

MHA node运行在每台MySQL服务器上，MHA Manager会定时探测集群中的master节点,当master 出现故障时，它可以自动将最新数据的slave提升为新的master,然后将所有其他的slave重新指向新的master。整个故障转移过程对应用程序是完全透明的。

1. 安装部署MHA前准备

MHA架构图

具体搭建如表：

角色	IP地址	主机名	serverID	类型
Monitor host	192.168.127.100	MHA		监控集群组
Master	192.168.127.101	master	101	写入
Candicate master	192.168.127.102	slave01	102	读
slave	192.168.127.103	slave02	103	读

vi /etc/hosts

127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4

::1 localhost localhost.localdomain localhost6 localhost6.localdomain6

192.168.127.100 MHA

192.168.127.101 master

192.168.127.102 slave01

192.168.127.103 slave02

1.1． percona-mysql安装（master、slave01、slave02 三台安装）

注意：三台的server_id 不一样，为了做主从同步

创建mysql用户：

useradd mysql

mkdir /app

mkdir -p /data/mysql3306

解决percona-mysql软件：

tar zxvf Percona-Server-5.6.27-rel75.0-Linux.x86_64.ssl101.tar.gz

注意：安装的软件需要根据openssl版本来下载

rpm -qa | grep ssl

openssl-1.0.1e-15.el6.x86_64

把解压文件移动相应目录：

mv Percona-Server-5.6.27-rel75.0-Linux.x86_64.ssl101 /app/mysql5.6

创建放慢查询日志目录：

mkdir /app/mysql5.6/logs

给目录权限：

chown -R mysql:mysql /app/mysql5.6

chown -R mysql:mysql /data/mysql3306

创建配置文件

vi /app/mysql5.6/my.cnf

[client]

socket=/app/mysql5.6/mysql.sock

default-character-set=utf8

port=3306

[mysql]

prompt=\\[email protected]\\d \\r:\\m:\\s>

no-auto-rehash

[mysqld_safe]

log-error=/data/mysql3306/mysqld.error

[mysqld]

socket=/app/mysql5.6/mysql.sock

pid-file=/app/mysql5.6/mysqld.pid

basedir=/app/mysql5.6

datadir=/data/mysql3306

port=3306

server_id=101

character-set-server=utf8

skip-external-locking

skip-name-resolve

max_connections=1024

max_connect_errors=1000

wait_timeout = 400

interactive_timeout = 400

table_definition_cache=500

table_open_cache=500

sort_buffer_size = 16M

tmp_table_size = 200M

read_buffer_size = 1M

read_rnd_buffer_size = 4M

myisam_sort_buffer_size = 64M

thread_cache_size = 8

query_cache_type=0

query_cache_size=0

thread_concurrency = 16

lower_case_table_names = 1

log_bin_trust_function_creators = 1

#################slow log####################

slow-query_log=1

slow-query_log_file=/app/mysql5.6/logs/mysql.slow

long_query_time=2

####################binlog######################

log-bin=mysql-bin

binlog-format=ROW

expire_logs_days=5

sync_binlog=1

################replication##########

log-slave-updates=1

################INNODB################

sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES

transaction-isolation=READ-COMMITTED

innodb_buffer_pool_size=10G

innodb_flush_log_at_trx_commit=2

innodb_strict_mode=1

innodb_flush_method=O_DIRECT

innodb_file_format=Barracuda

innodb_log_files_in_group=3

innodb_file_per_table=1

innodb_io_capacity=500

innodb_support_xa=1

innodb_additional_mem_pool_size=16M

innodb_log_buffer_size=64M

[mysqldump]

quick

max_allowed_packet=128M

myisam_max_sort_sort_file_size=2G

初始化数据库

/app/mysql5.6/scripts/mysql_install_db --user=mysql --basedir=/app/mysql5.6 --datadir=/data/mysql3306 --defaults-file=/app/mysql5.6/my.cnf

启动脚本

cp /app/mysql5.6/support-files/mysql.server /etc/init.d/mysql

vi /etc/init.d/mysql

basedir=/app/mysql5.6

datadir=/data/mysql3306

注意：修改以上两处即可

启动数据库

/etc/init.d/mysql start

Starting MySQL (Percona Server).... [ OK ]

环境变量配置

vi /etc/profile

export MYSQL_HOME=/app/mysql5.6

export MY_BASEDIR_VERSION=/app/mysql5.6

export PATH=/app/mysql5.6/bin:/app/mysql5.6/scripts:$PATH

export LD_LIBRARY_PATH=/app/mysql5.6/lib

生效环境变量

source /etc/profile

1.2 .主从同步搭建

注意：防火墙需要关闭

创建复制账号（master、slave1(mha新主)）

GRANT REPLICATION SLAVE ON *.* TO ‘repl‘@‘192.168.127.%‘ IDENTIFIED BY ‘repl‘;

flush privileges;

查看master binlog POS点信息

[email protected](none) 06:47:05>show master status;

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

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

| mysql-bin.000004 | 409 | | | |

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

1 row in set (0.01 sec)

建立主从复制（slave01、slave02）

[[email protected] ~]# mysql

Welcome to the MySQL monitor. Commands end with ; or \g.

Your MySQL connection id is 2

Server version: 5.6.27-75.0-log Percona Server (GPL), Release 75.0, Revision 8bb53b6

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.

[email protected](none) 07:03:39>CHANGE MASTER TO MASTER_HOST=‘192.168.127.101‘,MASTER_PORT=3306,MASTER_USER=‘repl‘,MASTER_PASSWORD=‘repl‘,MASTER_LOG_FILE=‘mysql-bin.000004‘,MASTER_LOG_POS=409;

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

[email protected](none) 07:03:41>start slave;

Query OK, 0 rows affected (0.02 sec)

查看主从复制

[email protected](none) 07:03:42>show slave status\G;

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

Slave_IO_State: Waiting for master to send event

Master_Host: 192.168.127.101

Master_User: repl

Master_Port: 3306

Connect_Retry: 60

Master_Log_File: mysql-bin.000004

Read_Master_Log_Pos: 409

Relay_Log_File: mysqld-relay-bin.000002

Relay_Log_Pos: 283

Relay_Master_Log_File: mysql-bin.000004

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: 409

Relay_Log_Space: 457

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: 101

Master_UUID: 8b1cf62d-e063-11e5-84ba-000c2908253f

Master_Info_File: /data/mysql3306/master.info

SQL_Delay: 0

SQL_Remaining_Delay: NULL

Slave_SQL_Running_State: Slave has read all relay log; waiting for the slave I/O thread to update it

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

1 row in set (0.02 sec)

ERROR:

No query specified

以上主从已经搭建好，下面我们安装与配置MHA

（1）slave服务器(192.168.127.102,192.168.103)设置read only;

mysql> set global read_only=1;

（2）设置relay log清除方式（在每个slave 下）

mysql> set global relay_log_purge=0;

(3)创建监控用户,在所有MYSQL上执行

mysql> grant all privileges on *.* to ‘root‘@‘192.168.127.%‘ identified by ‘123456‘;

mysql>flush privileges;

（4）在slave01(192.168.127.102)上创建复制用户：

mysql> grant replication slave on *.* to ‘repl‘@‘192.168.127.%‘ identified by ‘repl‘;

mysql>flush privileges;

2.安装部署MHA

2.1安装MHA node(在所有Mysql服务器上安装)

（1）安装依赖包

rpm -Uvh http://dl.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-8.noarch.rpm

rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-6

yum -y install perl-DBD-MySQL perl-Config-Tiny perl-Log-Dispatch perl-Parallel-ForkManager perl-Config-IniFiles perl-Time-HiRes perl-Time-HiRes perl-CPAN

（2）在所有的节点上安装mha node:

tar zxvf mha4mysql-node-0.56.tar.gz

cd mha4mysql-node-0.56

perl Makefile.PL

make

make install

2.2.安装MHA Manager

MHA Manager中主要包括了几个管理员的命令行工具，例如masterha_manager，masterha_master_switch等。

（1）安装依赖包

rpm -Uvh http://dl.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-8.noarch.rpm

rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-6

yum -y install perl-DBD-MySQL perl-Config-Tiny perl-Log-Dispatch perl-Parallel-ForkManager perl-Config-IniFiles perl-Time-HiRes perl-Time-HiRes perl-CPAN

（2）安装MHA node软件包。注意，在MHA Manger的主机上也要安装MHA node.

tar zxvf mha4mysql-node-0.56.tar.gz

cd mha4mysql-node-0.56

perl Makefile.PL

make

make install

（3）安装MHA Manager软件包。

tar zxvf mha4mysql-manager-0.56.tar.gz

cd mha4mysql-manager-0.56

perl Makefile.PL

make

make install

2.3. 配置SSH 登录无密码验证

（1）在manager （192.168.127.100）上配置到所有节点的无密码验证

ssh-keygen -t rsa

ssh-copy-id -i ~/.ssh/id_rsa.pub [email protected]

（2）在MHA Node master(192.168.127.101)上：

ssh-keygen -t rsa

ssh-copy-id -i ~/.ssh/id_rsa.pub [email protected]

（3）在MHA Node slave01(192.168.127.102)上：

ssh-keygen -t rsa

ssh-copy-id -i ~/.ssh/id_rsa.pub [email protected]

（4）在MHA Node slave02(192.168.127.103)上：

ssh-keygen -t rsa

ssh-copy-id -i ~/.ssh/id_rsa.pub [email protected]

在每台做以下步骤

ln -s /app/mysql5.6/bin/* /usr/local/bin/

把脚本拷贝相关目录

[[email protected] ~]# cp /root/mha4mysql-manager-0.56/samples/scripts/master_ip_failover /usr/local/bin/

[[email protected] ~]# cp /root/mha4mysql-manager-0.56/samples/scripts/master_ip_online_change /usr/local/bin/

[[email protected]~]#cp /root/mha4mysql-manager-0.56/samples/scripts/send_report /usr/local/bin/

[[email protected] ~]# cp /root/mha4mysql-manager-0.56/bin/masterha_secondary_check /usr/bin/

3.配置MHA

配置MHA的步骤如下。

（1）创建MHA工作目录，并且创建相关配置文件：

mkdir -p /etc/masterha

mkdir -p /masterha/app1

配置如下

vi /etc/masterha/app1.cnf

[server default]

manager_workdir=/masterha/app1

manager_log=/masterha/app1/app1.log

master_ip_failover_script=/usr/local/bin/master_ip_failover

master_ip_online_change_script=/usr/local/bin/master_ip_online_change

user=root

password=123456

ssh_user=root

repl_user=repl

repl_password=repl

ping_interval=1

remote_workdir=/tmp

report_script=/usr/local/bin/send_report

secondary_check_script=/usr/bin/masterha_secondary_check -s MHA -s slave02 --user=root --master_host=master --master_ip=192.168.127.101 --master_port=3306 --password=123456

shutdown_script=""

report_script=""

[server1]

hostname=192.168.127.101

master_binlog_dir=/data/mysql3306

candidate_master=1

[server2]

hostname=192.168.127.102

master_binlog_dir=/data/mysql3306

candidate_master=1

check_repl_delay=0

[server3]

hostname=192.168.127.103

master_binlog_dir=/data/mysql3306

no_master=1

4.检查SSH的配置

检查MHA Manager到所有MHA node的SSH连接状态：

[[email protected] ~]# masterha_check_ssh --conf=/etc/masterha/app1.cnf

Wed Mar 2 19:03:30 2016 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.

Wed Mar 2 19:03:30 2016 - [info] Reading application default configuration from /etc/masterha/app1.cnf..

Wed Mar 2 19:03:30 2016 - [info] Reading server configuration from /etc/masterha/app1.cnf..

Wed Mar 2 19:03:30 2016 - [info] Starting SSH connection tests..

Wed Mar 2 19:03:31 2016 - [debug]

Wed Mar 2 19:03:30 2016 - [debug] Connecting via SSH from [email protected](192.168.127.101:22) to [email protected](192.168.127.102:22)..

Wed Mar 2 19:03:30 2016 - [debug] ok.

Wed Mar 2 19:03:30 2016 - [debug] Connecting via SSH from [email protected](192.168.127.101:22) to [email protected](192.168.127.103:22)..

Wed Mar 2 19:03:30 2016 - [debug] ok.

Wed Mar 2 19:03:31 2016 - [debug]

Wed Mar 2 19:03:30 2016 - [debug] Connecting via SSH from [email protected](192.168.127.102:22) to [email protected](192.168.127.101:22)..

Wed Mar 2 19:03:31 2016 - [debug] ok.

Wed Mar 2 19:03:31 2016 - [debug] Connecting via SSH from [email protected](192.168.127.102:22) to [email protected](192.168.127.103:22)..

Wed Mar 2 19:03:31 2016 - [debug] ok.

Wed Mar 2 19:03:32 2016 - [debug]

Wed Mar 2 19:03:31 2016 - [debug] Connecting via SSH from [email protected](192.168.127.103:22) to [email protected](192.168.127.101:22)..

Wed Mar 2 19:03:31 2016 - [debug] ok.

Wed Mar 2 19:03:31 2016 - [debug] Connecting via SSH from [email protected](192.168.127.103:22) to [email protected](192.168.127.102:22)..

Wed Mar 2 19:03:32 2016 - [debug] ok.

Wed Mar 2 19:03:32 2016 - [info] All SSH connection tests passed successfully.

5.检查整个复制环境

[[email protected] ~]# masterha_check_ssh --conf=/etc/masterha/app1.cnf

Wed Mar 2 19:03:30 2016 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.

Wed Mar 2 19:03:30 2016 - [info] Reading application default configuration from /etc/masterha/app1.cnf..

Wed Mar 2 19:03:30 2016 - [info] Reading server configuration from /etc/masterha/app1.cnf..

Wed Mar 2 19:03:30 2016 - [info] Starting SSH connection tests..

Wed Mar 2 19:03:31 2016 - [debug]

Wed Mar 2 19:03:30 2016 - [debug] Connecting via SSH from [email protected](192.168.127.101:22) to [email protected](192.168.127.102:22)..

Wed Mar 2 19:03:30 2016 - [debug] ok.

Wed Mar 2 19:03:30 2016 - [debug] Connecting via SSH from [email protected](192.168.127.101:22) to [email protected](192.168.127.103:22)..

Wed Mar 2 19:03:30 2016 - [debug] ok.

Wed Mar 2 19:03:31 2016 - [debug]

以上是关于Percona-mysql MHA高可用实战方案的主要内容，如果未能解决你的问题，请参考以下文章

基于半同步复制的MHA高可用MySql集群架构搭建实战

MySQL数据库——MHA高可用集群架构（实战！！！）

MHA架构概述及实战

MySQL架构之MHA架构实战

MySQL高可用之MHA架构企业实战