Group Replication uses a distributed recovery process to
synchronize group members when joining them to the group.
Distributed recovery involves transferring transactions from a
donor's binary log to a joining member using a replication
must therefore set up a replication user with the correct
permissions so that Group Replication can establish direct
member-to-member replication channels. If group members have
been set up to support the use of a remote cloning operation as
part of distributed recovery, this replication user is also used
as the clone user on the donor, and requires the correct
permissions for this role too. For a complete description of
distributed recovery, see
Section 20.5.4, “Distributed Recovery”.
The same replication user must be used for distributed recovery on every group member. The process of creating the replication user for distributed recovery can be captured in the binary log, and then you can rely on distributed recovery to replicate the statements used to create the user. Alternatively, you can disable binary logging before creating the replication user, and then create the user manually on each member, for example if you want to avoid the changes being propagated to other server instances. If you do this, ensure you re-enable binary logging once you have configured the user.
If distributed recovery connections for your group use SSL, the replication user must be created on each server before the joining member connects to the donor. For instructions to set up SSL for distributed recovery connections and create a replication user that requires SSL, see Section 20.6.3, “Securing Distributed Recovery Connections”
By default, users created in MySQL 8 use Section 184.108.40.206, “Caching SHA-2 Pluggable Authentication”. If the replication user for distributed recovery uses the caching SHA-2 authentication plugin, and you are not using SSL for distributed recovery connections, RSA key-pairs are used for password exchange. You can either copy the public key of the replication user to the joining member, or configure the donors to provide the public key when requested. For instructions to do this, see Section 220.127.116.11, “Secure User Credentials for Distributed Recovery”.
To create the replication user for distributed recovery, follow these steps:
Start the MySQL server instance, then connect a client to it.
If you want to disable binary logging in order to create the replication user separately on each instance, do so by issuing the following statement:
mysql> SET SQL_LOG_BIN=0;
Create a MySQL user with the following privileges:
REPLICATION SLAVE, which is required for making a distributed recovery connection to a donor to retrieve data.
CONNECTION_ADMIN, which ensures that Group Replication connections are not terminated if one of the servers involved is placed in offline mode.
BACKUP_ADMIN, if the servers in the replication group are set up to support cloning (see Section 18.104.22.168, “Cloning for Distributed Recovery”). This privilege is required for a member to act as the donor in a cloning operation for distributed recovery.
GROUP_REPLICATION_STREAM, if the MySQL communication stack is in use for the replication group (see Section 20.6.1, “Communication Stack for Connection Security Management”). This privilege is required for the user account to be able to establish and maintain connections for Group Replication using the MySQL communication stack.
In this example the user
rpl_userwith the password
passwordis shown. When configuring your servers use a suitable user name and password:
mysql> CREATE USER rpl_user@'%' IDENTIFIED BY 'password'; mysql> GRANT REPLICATION SLAVE ON *.* TO rpl_user@'%'; mysql> GRANT CONNECTION_ADMIN ON *.* TO rpl_user@'%'; mysql> GRANT BACKUP_ADMIN ON *.* TO rpl_user@'%'; mysql> GRANT GROUP_REPLICATION_STREAM ON *.* TO rpl_user@'%'; mysql> FLUSH PRIVILEGES;
If you disabled binary logging, enable it again as soon as you have created the user, by issuing the following statement:
mysql> SET SQL_LOG_BIN=1;
When you have created the replication user, you must supply the user credentials to the server for use with distributed recovery. You can do this by setting the user credentials as the credentials for the
group_replication_recoverychannel, using a
CHANGE REPLICATION SOURCE TOstatement. Alternatively, you can specify the user credentials for distributed recovery in a
User credentials set using
CHANGE REPLICATION SOURCE TOare stored in plain text in the replication metadata repositories on the server. They are applied whenever Group Replication is started, including automatic starts if the
group_replication_start_on_bootsystem variable is set to
User credentials specified on
START GROUP_REPLICATIONare saved in memory only, and are removed by a
STOP GROUP_REPLICATIONstatement or server shutdown. You must issue a
START GROUP_REPLICATIONstatement to provide the credentials again, so you cannot start Group Replication automatically with these credentials. This method of specifying the user credentials helps to secure the Group Replication servers against unauthorized access.
For more information on the security implications of each method of providing the user credentials, see Section 22.214.171.124.3, “Providing Replication User Credentials Securely”. If you choose to provide the user credentials using a
CHANGE REPLICATION SOURCE TOstatement, issue the following statement on the server instance now, replacing
passwordwith the values used when creating the user:
mysql> CHANGE REPLICATION SOURCE TO SOURCE_USER='rpl_user', -> SOURCE_PASSWORD='password' -> FOR CHANNEL 'group_replication_recovery';