Manage cross-cluster replication in Aiven for Apache Cassandra®#

Important

Aiven for Apache Cassandra® cross-cluster replication (CCR) is a limited availability feature. If you’re interested in trying out this feature, contact the sales team at sales@Aiven.io.

Learn how to update Apache Cassandra® services that has cross-cluster replication (CCR) enabled: change the service plan and add an extra disk space. Find out how to set up the replication factor and the consistency level for your CCR-enabled Apache Cassandra® services.

Prerequisites#

Aiven-wise#

  • Aiven account

  • Pair of Aiven for Apache Cassandra services with CCR enabled

Tools#

  • To update the service plan or add an extra disk space, use Aiven Console.

  • To set up the replication factor on the database side, issue the CREATE KEYSPACE statement from any supported client driver. This guide uses the cqlsh Cassandra client for that purpose to ensure general applicability of the instruction.

  • To set up the consistency level on the client side, configure it in your software. This guide uses the cqlsh Cassandra client for that purpose to ensure general applicability of the instruction.

Change the service plan#

Important

When you change the plan for your CCR-enabled service, the plan for the connected replica service changes accordingly since the services constituting a CCR pair always share the same service plan.

Note

It’s recommended to use Aiven Console for changing the plan for a CCR-enabled service.

  1. Log in to Aiven Console.

  2. From the Services page, select a CCR-enabled Aiven for Apache Cassandra service that you want to update.

  3. On the Overview page of your service, select Service settings from the sidebar.

  4. On the Service settings page of your service, navigate to the Service plan section, and select Change plan from the Actions (…) menu.

  5. In the Change service plan window, select a new plan you want to use for your service.

    Tip

    You can also add extra disk space for your service by using the slider in the Additional disk storage section.

  6. Select Change.

Result

You’ve changed the plan for your CCR-enabled service and its CCR-replica service.

Add an extra disk space#

Important

Changes to the storage space are applied to both services constituting the CCR pair so also affect the replica service.

Note

It’s recommended to use Aiven Console for adding storage space for CCR-enabled services.

  1. Log in to Aiven Console.

  2. From the Services page, select a CCR-enabled Aiven for Apache Cassandra service that you want to update.

  3. On the Overview page of your service, select Service settings from the sidebar.

  4. On the Service settings page of your service, navigate to the Service plan section, and select Add additional storage from the Actions (…) menu.

  5. In the Upgrade service storage window, use the slider to add extra disk space for your service.

    Tip

    You can also change your service plan by selecting Change plan in the Your current plan section.

  6. Select Save changes.

Result

You’ve added extra disk storage space for your CCR-enabled service and its CCR-replica service.

Set up the replication factor#

You can specify how many replicas of your data you’d like to have on either of datacenters hosting your service. For that purpose, you need a keyspace with the NetworkTopologyStrategy replication. To create a keyspace that supports CCR and defines the replication factor, you need to run the CREATE KEYSPACE query with a set of parameters configuring the keyspace as needed.

Note

This instruction uses the cqlsh Cassandra CLI client to configure the replication factor. cqlsh is used for demonstration purposes and the same statements can be executed using any supported client driver.

  1. Connect to your service via cqlsh.

    Note

    You can connect to either of the two services constituting the CCR pair to set up the replication factor.

  2. From the cqlsh shell, check out

    • Existing keyspaces with the DESCRIBE keyspaces; query (for a new service, only system keyspaces are returned)

    • Datacenters available for your service with the SELECT data_center from system.peers_v2; query.

  3. Create a keyspace by running a query in which you specify

    • Replication strategy ('class': 'NetworkTopologyStrategy')

    • Number of replicas to be created in the first datacenter ('datacenter_1_name': 'number_of_replicas')

    • Number of replicas to be created in the second datacenter ('datacenter_2_name': 'number_of_replicas')

    CREATE KEYSPACE keyspace_name WITH replication =    /
{                                                   /
  'class': 'NetworkTopologyStrategy',               /
  'datacenter_1_name': 'number_of_replicas',        /
  'datacaenter_2_name': 'number_of_replicas'        /
}                                                   /
AND durable_writes = true;
    Example#
    CREATE KEYSPACE default WITH replication =           /
{                                                    /
  'class': 'NetworkTopologyStrategy',                /
  'dc_1': '3',                                       /
  'dc_2': '3'                                        /
}                                                    /
AND durable_writes = true;

Result

You’ve set up the replication factor for your keyspace. Now all data within this keyspace gets replicated to the datacenters according to the specified factor.

See also

For more details on the replication factor for Apache Cassandra, see NetworkTopologyStrategy in the Apache Cassandra documentation.

Set up the consistency level#

For Apache Cassandra, you can set up the CONSISTENCY parameter, which regulates when the client can consider an operation as successfully completed. The CONSISTENCY parameter defines how many nodes need to confirm the operation as finalized before the client can acknowledge the operation as successfully completed.

Note

You can configure the consistency level in the shell or in a client library. While using the cqlsh CLI client is convenient for setting up keyspaces or testing, configuring and using a client driver is recommended for operations in the production environment, such as data imports, data querying, or data reads/writes from/to databases.

In the shell#

Note

This instruction uses the cqlsh Cassandra CLI client to configure the consistency level.

  1. Connect to your service via cqlsh.

  2. Run CONSISTENCY; to check your current setting for the consistency level.

    Expected output: The query can return, for example, Current consistency level is ONE., which means that a confirmation of an operation completion on one node is enough for this operation to be considered as successful.

  1. To set up the consistency level to a specific value, run the CONSISTENCY consistency_level_argument; query.

    Allowed consistency level arguments: For the list of the allowed consistency level arguments for Apache Cassandra, see CONSISTENCY in the Apache Cassandra documentation.

Example#
CONSISTENCY QUORUM;

In a client library#

To configure the consistency level in a client library, add an extra parameter or object to define the consistency level on your software component before running a particular query.

Example:

In Python, you can specify consistency_level as a parameter for the SimpleStatement object.

session.execute(SimpleStatement("LIST ROLES", consistency_level=ConsistencyLevel.ALL))

You’ve set up the consistency level for your service. Now operations on your data are considered as successfully completed according to the consistency level you specified.

See also

For more details on consistency levels for Apache Cassandra, see CONSISTENCY in the Apache Cassandra documentation.

More on Apache Cassandra CCR#

More on CCR with Aiven#