MySQL Shell 8.4  /  MySQL InnoDB ReplicaSet  /  Adopting an Existing Replication Setup

9.6 Adopting an Existing Replication Setup

As an alternative to creating a ReplicaSet from scratch, you can adopt an existing replication setup using the adoptFromAR option with dba.createReplicaSet(). The replication setup is scanned, and if it is compatible with the InnoDB ReplicaSet Limitations, AdminAPI creates the necessary metadata. Once the replication setup has been adopted, you can only use AdminAPI to administer the InnoDB ReplicaSet.

To convert an existing replication setup to an InnoDB ReplicaSet connect to the primary, also referred to as the source. The replication topology is automatically scanned and validated, beginning from the connected global session of the instance MySQL Shell. The configuration of all instances is checked during the adoption to ensure they are compatible with InnoDB ReplicaSet usage:

  • All replication channels must be active, and their transaction sets as verified through GTID sets must be consistent.

  • Instances are assumed to have the same state or be able to converge.


All new replication channels are created with SSL enabled. This is not true for replication groups adopted with MySQL Shell 8.0.32. Their replication channels remain unencrypted.

The replication topology is automatically scanned and validated, starting from the instance MySQL Shell's global session is connected to. The only changes made by this operation to an adopted ReplicaSet are the creation of the metadata schema. Existing replication channels are not changed during adoption, although you can change them during subsequent primary switch operations.

For example, to adopt a replication topology consisting of the MySQL server instances on example1 and example2 to an InnoDB ReplicaSet.

Connect to the primary at example1 and issue:

mysql-js> rs = dba.createReplicaSet('testadopt', {'adoptFromAR':1})
A new replicaset with the topology visible from 'example1:3306' will be created.

* Scanning replication topology...
** Scanning state of instance example1:3306
** Scanning state of instance example2:3306

* Discovering async replication topology starting with example1:3306
Discovered topology:
- example1:3306: uuid=00371d66-3c45-11ea-804b-080027337932 read_only=no
- example2:3306: uuid=59e4f26e-3c3c-11ea-8b65-080027337932 read_only=no
    - replicates from example1:3306
	source="localhost:3310" channel= status=ON receiver=ON applier=ON

* Checking configuration of discovered instances...

This instance reports its own address as example1:3306
example1:3306: Instance configuration is suitable.

This instance reports its own address as example2:3306
example2:3306: Instance configuration is suitable.

* Checking discovered replication topology...
example1:3306 detected as the PRIMARY.
Replication state of example2:3306 is OK.

Validations completed successfully.

* Updating metadata...

ReplicaSet object successfully created for example1:3306.
Use rs.addInstance() to add more asynchronously replicated instances to
this replicaset and rs.status() to check its status.

Once the InnoDB ReplicaSet has been adopted, you can use it in the same way that you would use a ReplicaSet which was created.


From this point, you must administer the InnoDB ReplicaSet using only AdminAPI.