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