[分享]mysql-mmm_MySQL, Oracle及数据库讨论区_Weblogic技术|Tuxedo技术|中间件技术|Oracle论坛|JAVA论坛|Linux/Unix技术|hadoop论坛

总帖数

每页帖数

1/1页

返回列表

发起投票

查看: 3166 | 回复: 0

主题： [分享]mysql-mmm

shary

注册用户

等级：新兵
经验：54
发帖：66
精华：0
注册：2011-8-31
状态：离线
发送短消息息给shary

加好友发送短消息息给shary

发消息

发表于：

2014-11-18 11:40:41 | [全部帖] [楼主帖]

楼主

一、关系图

二、安装mysql-mmm

1)安装epel

相关文档参见：http://fedoraproject.org/wiki/EPEL

rpm -Uvh http://download.fedora.redhat.com/pub/epel/5/i386/epel-release-5-4.noarch.rpm

2) monitor软件安装

[root@localhost ~]# yum install –y mysql-mmm-monitor-*

包如下：

[root@localhost ~]# rpm -qa|grep mysql
mysql-mmm-monitor-2.2.1-1.el5
mysql-mmm-2.2.1-1.el5
mysql-5.0.77-4.el5_5.4

3）所在机器安装(包括master,slave)

[root@localhost ~]# yum install –y -mmm-agent*

包如下：

[root@localhost ~]# rpm -qa|grep mysql
mysql-test-5.0.77-4.el5_5.4
mysql-mmm-agent-2.2.1-1.el5
mysql-5.0.77-4.el5_5.4
mysql-server-5.0.77-4.el5_5.4
mysql-devel-5.0.77-4.el5_5.4
mysql-bench-5.0.77-4.el5_5.4
mysql-mmm-2.2.1-1.el5
mysql-connector-odbc-3.51.26r1127-1.el5

三、配置文件

1)monitor配置文件

[root@localhost ~]# vi /etc/mysql-mmm/mmm_mon.conf
include mmm_common.conf
<monitor>
ip                  127.0.0.1
pid_path            /var/run/mysql-mmm/mmm_mond.pid
bin_path            /usr/libexec/mysql-mmm
status_path         /var/lib/mysql-mmm/mmm_mond.status
ping_ips            172.16.71.128,172.16.71.130 #master,slave实际数据库ip
auto_set_online     60
</monitor>
<host default>
monitor_user        mmm_monitor
monitor_password    10jqka
</host>
Debug

2）通用的配置文件，在monitor及master,slave上放置的这个文件都一致

[root@localhost ~]# vi /etc/mysql-mmm/mmm_common.conf
active_master_role      writer
<host default>
cluster_interface       eth0
pid_path                /var/run/mysql-mmm/mmm_agentd.pid
bin_path                /usr/libexec/mysql-mmm/
replication_user        replicant
replication_password    10jqka
agent_user              mmm_agent
agent_password          10jqka
</host>
<host db1>
ip      172.16.71.128 #真实master数据库ip
mode    master
peer    db2
</host>
<host db2>
ip      172.16.71.130 #真实master数据库ip
mode    master
peer    db1
</host>
#<host db3>
# ip 192.168.100.51 #真实slave数据库ip
#    mode    slave
#</host>
<role writer>
hosts   db1, db2
ips     172.16.71.253 #虚拟写数据库ip
mode    exclusive
</role>
<role reader>
hosts   db1, db2
ips     172.16.71.251,172.16.71.252 #虚拟读数据库ip
mode    balanced
</role>

3）master,slave配置文件

[root@localhost ~]# vi /etc/mysql-mmm/mmm_agent.conf
include mmm_common.conf
# The 'this' variable refers to this server. Proper operation requires
# that 'this' server (db1 by default), as well as all other servers, have the
# proper IP addresses set in mmm_common.conf.
this db1#如为第一台master db则为db1如为第二台则为db2与mmm_common.conf里的机器对应

三、查看效果

1）启动db1的相关服务

service mysqld start
service mysql-mmm-agent start

数据库授权：所有数据库的授权一致

GRANT REPLICATION CLIENT ON *.* TO 'mmm_monitor'@'172.16.71.%' IDENTIFIED BY '10jqka';
GRANT SUPER, REPLICATION CLIENT, PROCESS ON *.* TO 'mmm_agent'@'172.16.71.%' IDENTIFIED BY '10jqka';
grant all on *.* to replicant@'172.16.71.%' identified by '10jqka';
flush privileges;

2）启动monitor相关服务

service mysql-mmm-monitor start

查看状态：

mmm_control show

把db1设置为online状态：

mmm_control set_online db1

开放mmm_agentd端口9989

真实机器查看分配的vip地址

[root@localhost ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 00:0c:29:ef:79:70 brd ff:ff:ff:ff:ff:ff
inet 172.16.71.132/24 brd 172.16.71.255 scope global eth0
inet 172.16.71.251/32 scope global eth0
inet 172.16.71.253/32 scope global eth0
inet6 fe80::20c:29ff:feef:7970/64 scope link
valid_lft forever preferred_lft forever

四、主主（master-master）同步配置

1）172.16.71.132机器my.cnf

skip-name-resolve
server_id = 132
set-variable = key_buffer_size=512M
set-variable = table_cache=32
set-variable = sort_buffer_size=32M
set-variable = read_buffer_size=32M
set-variable = max_connections=2100
set-variable = log_slow_queries=/var/lib/mysql/log/mysql-slow.log
set-variable = log_bin=/var/lib/mysql/log/mysql-bin
set-variable = long_query_time=1
set-variable = thread_concurrency=16
set-variable = tmp_table_size=4000000000
set-variable = slave_skip_errors=all
#replicate-ignore-db=mysql
#replicate-ignore-db=test 发觉加上这个ignore选项后会导致mmm接管出现点小问题不知道为什么
log-slave-updates
sync_binlog=1
auto_increment_increment=2
auto_increment_offset=1

进入mysql配置设置为从

mysql> change master to master_host='172.16.71.128';
mysql>change master to master_user='repl';
mysql>change master to master_password='10jqka';
mysql>start slave

主（master）服务器配置：

GRANT REPLICATION SLAVE ON *.* TO 'repl'@'172.16.71.%' IDENTIFIED BY '10jqka';

2）172.16.71.128机器my.cnf配置

skip-name-resolve
server_id = 1
set-variable = key_buffer_size=512M
set-variable = table_cache=32
set-variable = sort_buffer_size=32M
set-variable = read_buffer_size=32M
set-variable = max_connections=2100
set-variable = log_slow_queries=/var/lib/mysql/log/mysql-slow.log
set-variable = log_bin=/var/lib/mysql/log/mysql-bin
set-variable = long_query_time=1
set-variable = thread_concurrency=16
set-variable = tmp_table_size=4000000000
set-variable = slave_skip_errors=all
#replicate-ignore-db=mysql
#replicate-ignore-db=test
log-slave-updates
sync_binlog=1
auto_increment_increment=2
auto_increment_offset=2

进入mysql配置设置为从

mysql> change master to master_host='172.16.71.132';
mysql>change master to master_user='repl';
mysql>change master to master_password='10jqka';
mysql>start slave

主（master）服务器配置：

GRANT REPLICATION SLAVE ON *.* TO 'repl'@'172.16.71.%' IDENTIFIED BY '10jqka';

五、其它从(slave)服务器配置：

my.cnf配置

skip-name-resolve
server_id           = 1
set-variable = key_buffer_size=512M
set-variable = table_cache=32
set-variable = sort_buffer_size=32M
set-variable = read_buffer_size=32M
set-variable = max_connections=2100
set-variable = log_slow_queries=/var/lib/mysql/log/mysql-slow.log
set-variable = log_bin=/var/lib/mysql/log/mysql-bin
set-variable = long_query_time=1
set-variable = thread_concurrency=16
set-variable = tmp_table_size=4000000000
set-variable = slave_skip_errors=all
mysql>change master to master_host='172.16.71.253';
mysql>change master to master_user='repl';
mysql>change master to master_password='10jqka';                           mysql>change master to master_log_file=’mysql-bin.00001’;
mysql>change master to master_log_pos=128;                                mysql>start slave

或者在my.cnf加入

server-id=2
master-host=172.16.71.253
master-user=repl
master-password=10jqka

问题1：

两台机器互为对方的主，此时不能使用master的VIP作为该两机器的主，如果使用VIP作为master的话没有一台先起来作为vip机器的话些会导致monitor不能使用，日志如下：

2010/12/06 16:29:46 WARN db1 is not replicating from db2
2010/12/06 16:29:46 WARN db2 is not replicating from db1

问题2：

当/etc/mysql-mmm/mmm_mon.conf的auto_set_online 3时间设置为较长时，在瞬间关闭现有主，开启现未开启的备时，会导致monitor不能自动切换VIP导致mmm不可用，需要手工设置online状态，有时候甚至会导致monitor启动问题。为防止这情况发生可以把该时间设置短些，或者先让一其中的一台主起来后，再关闭另一台。

问题3：ERROR：Can't connect to monitor daemon

查看进程一般少了perl相关进程，莫名其妙的感觉，经多次尝试试验，一般都是在mysql那些拥有agent机器上的配置问题，还有monitor的auto_set_online时间设置

[root@localhost ~]# ps axu|grep mmm
root      3113 0.0 7.6 14000 8956 ?        S    16:27   0:00 mmm_mond
root      3115 0.9 33.2 106440 38976 ?        Sl   16:27   0:00 mmm_mond
root      3136 0.4 5.5 10388 6456 ?        S    16:28   0:00 perl /usr/libexec/mysql-mmm/monitor/checker ping_ip
root      3140 0.4 6.5 15208 7640 ?        S    16:28   0:00 perl /usr/libexec/mysql-mmm/monitor/checker mysql
root      3141 0.3 5.4 10388 6448 ?        S    16:28   0:00 perl /usr/libexec/mysql-mmm/monitor/checker ping
root      3142 0.5 6.5 15208 7636 ?        S    16:28   0:00 perl /usr/libexec/mysql-mmm/monitor/checker rep_backlog
root      3144 0.6 6.4 15208 7592 ?        S    16:28   0:00 perl /usr/libexec/mysql-mmm/monitor/checker rep_threads
root      3148 0.0 0.5   3916   652 pts/0    R+   16:28   0:00 grep mmm

六、其它补充资料

主从状态查看：

查看master的状态

SHOW MASTER STATUS;

Position不应为0

查看slave的状态

show slave status;

Slave_IO_Running | Slave_SQL_Running这两个字段应为YES|YES.

show processlist;查看进程

slave start; #启动从复制
STOP SLAVE; #停止从复制
RESET MASTER; #重设master日志
RESET SLAVE;#重设slave日志

当Slave_IO_Running: 为No时，从服务器无法同步时操作下如

SLAVE STOP;#停止slave
CHANGE MASTER TO MASTER_LOG_FILE='mysql-bin.000001', MASTER_LOG_POS=0;
#把master_log_file设置为从主show master status;得出来的日志名，POS设置为0
SLAVE START;#再次启用slave

Mysql普通主从配置（未使用mmm参考）

配置主从my.cnf文件：

skip-innodb
skip-name-resolve
server_id = 132
set-variable = key_buffer_size=512M
set-variable = table_cache=32
set-variable = sort_buffer_size=32M
set-variable = read_buffer_size=32M
set-variable = max_connections=2100
set-variable = log_slow_queries=/var/lib/mysql/log/mysql-slow.log
set-variable = log_bin=/var/lib/mysql/log/mysql-bin#从库这条语句注释掉
set-variable = long_query_time=1
set-variable = thread_concurrency=16
set-variable = tmp_table_size=4000000000
set-variable = slave_skip_errors=all

因为mysql版本不同在从服务器下开启了下面选项导致同步不能进行

replicate-ignore-db=mysql
replicate-do-db =test

--转自