Set up MySQL Source-Replica Replication
Source-replica data replication enables you to copy replicated data to
multiple computers for backup and analysis by multiple parties. You should submit
necessary changes identified by a group member to the
designated primary of the node. This differs from source-source
replication,
in which any authorized contributor of the group can update data.
This article provides steps for setting up MySQL® source-replica database
replication between two cloud servers. The operating system used for the
examples in the article is CentOS® 6, built from a Rackspace Cloud
Servers base image.
Before you begin
The steps in this article use two cloud servers, db01
and db02
.
Cloud servers have two IP addresses (one public, one private). The
examples demonstrate configuring replication over the private IP
interface so that no bandwidth charges are incurred. For the duration of
the article, db01
is considered the source MySQL server (running in
read-write mode), and db02
is considered the replica server (running in
read-only mode).
If you already have a MySQL database running on the source node, a dump
and restore into the replica node is required before configuring
replication between them. You use the mysqldump
command to dump a
database into a file, then transfer it and restore it to the replica.
After the necessary configuration has been performed, replication is in
effect. For more information, see the Configure replication
section.
Create the Cloud Servers
Create two Linux® cloud servers, using the Centos 6 base image. Use the
following steps to create each server separately.
- Log in to the Cloud Control Panel.
- In the top navigation bar, click Select a Product then Rackspace Cloud.
- Select Servers then Cloud Servers.
- Click Create Server.
- Name the servers so that you can easily identify them during
the setup. - Select the Centos 6 base image.
- Select the RAM configuration (flavor) appropriate for your
database requirements. - Click Create Server.
The commands outlined in the following sections need to be executed by a
privileged root or sudo group user. Any strings or values specified in
brackets should be replaced with data specific to your setup.
Install MySQL
You must install the mysql-server package on both CentOS cloud
servers.
-
Before installing MySQL, confirm that the package database is
up-to-date by running the following command:#yum update
-
Install MySQL and enable it to run at boot automatically:
#yum install mysql-server #chkconfig mysqld on
-
Start the
mysqld
service:#service mysqld start
-
After the
mysqld
service has been started, set your MySQL server
root password by using following commands:/usr/bin/mysqladmin -u root password 'new-password' /usr/bin/mysqladmin -u root -h web01 password 'new-password'
Note: Alternatively, you can run the secure installation script
packaged with the MySQL installation:# /usr/bin/mysql_secure_installation Enter current password for root (enter for none): ... Set root password? [Y/n] Y ... Remove anonymous users? [Y/n] Y ... Disallow root login remotely? [Y/n] Y ... Remove test database and access to it? [Y/n] Y ... Reload privilege tables now? [Y/n] Y
-
To permit connections on port 3306 (the
mysqld
default port), add a
TCP port 3306 rule with an insert at the last line number in the
RH-Firewall-1-INPUT
chain (in this case, line 10):# iptables -I RH-Firewall-1-INPUT 10 -p tcp --dport 3306 -j ACCEPT
-
Save the firewall configuration:
# service iptables save
Complete the following section to make relevant configuration changes to
enable replication.
Configure replication
A MySQL user is required on the source server (db01
) to be used for
replication.
-
Run the following commands to set up the MySQL user, updating the
entries in brackets with strings or values that you plan to use
with your setup:Note: You might not need to create the user in the follow code.
# mysql -u root -p mysql> create user [replication_username]@'[private IP of db02]' identified by '[some password]'; mysql> grant replication slave on *.* TO [replication_username]@'[private IP of db02]'; mysql> flush privileges; mysql> quit
-
Edit the /etc/my.cnf file, and add the following entries:
bind-address = 0.0.0.0 server-id = 1 log-bin = mysql-bin binlog-ignore-db = "mysql"
-
After you have finished updating the /etc/my.cnf file, restart
themysqld
service.#service mysqld restart
Before starting replication, the data on the source
and replica servers must be the same. To accomplish this duplication, dump
the data from the source (db01
) server and add it to the
replica (db02
) server. -
Use the following command to ensure that nothing can write to the
source database during a database dump. Also note the filename and
position of the binary log because you need these values to
complete the replication configuration on db02:# mysql -u root -p mysql> FLUSH TABLES WITH READ LOCK; mysql> SHOW MASTER STATUS; +------------------+--------------------------+------------------+ | File | Position | Binlog_Do_DB | Binlog_Ignore_DB | +------------------+--------------------------+------------------+ | mysql-bin.000010 | 10 | | mysql | +------------------+--------------------------+------------------+ 1 row in set (0.00 sec)
Note 1: Record the filename and position of the binary log because you need these values to
complete the replication configuration ondb02
.Note 2: Keep this session open, closing it releases the lock!
-
Perform a database dump by using
mysqldump
as follows:# mysqldump -u root -p --databases [database-1] [database-2] ... > /root/db_dump.sql
-
After the database dump has completed, lift the read lock from the source
(db01
) by typing the following, or by exiting the open session:mysql> UNLOCK TABLES;
-
Copy the database dump file to the replica server so that it can
be restored by using the following command:scp /root/db_dump.sql [private-IP-of-db02]:/root/
-
On
db02
, edit the /etc/my.cnf file and add the following
entries:bind-address = 0.0.0.0 server-id = 2 master-host = [private-IP-of-db01] master-user = [replication-username] master-password = [replication-password] master-connect-retry = 60
-
Import the db_dump.sql file copied earlier and restart the
MySQL service.# mysql -u root -p < /root/db_dump.sql # service mysqld restart
-
Complete the replica replication steps:
# mysql -u root -p mysql> SLAVE STOP; mysql> CHANGE MASTER TO MASTER_HOST='[private-IP-of-db01]', MASTER_USER='[replication-username]', MASTER_PASSWORD='[replication-password]', MASTER_LOG_FILE='[file-listed-on-master-status]', MASTER_LOG_POS=[log-position-listed-on-master-status]; mysql> START SLAVE; mysql> SHOW SLAVE STATUS\G
Note: The Slave_IO_State field should show "Waiting for master to
send event". If it shows "Connecting to Master", check your
MySQL log file. By default, it is /var/log/mysqld.log,
but it might be configured differently on your system. As
always, /etc/my.cnf defines the location of your log file.
Test replication
To test the replication setup, create a new database and associated
table on db01
and insert data to confirm that the changes are mirrored
on db02
. In the following example, the new database is named
testing, and the new table is named users:
# mysql -u root -p
mysql> create database testing;
mysql> use testing
mysql> create table users(id int not null auto_increment, primary key(id), username varchar(30) not null);
mysql> insert into users (username) values ('foo');
mysql> insert into users (username) values ('bar');
mysql> exit
You should see the changes on db02
immediately.
Updated 12 months ago