Configure HCD Unified Authentication in a live cluster

To implement HCD Unified Authentication in an existing deployment (a live cluster) without downtime or interrupted access, use transitional modes to temporarily support applications and users that don’t have an authenticated role:

  1. Update all applications to provide credentials and ensure that they use the latest version of a supported Apache Cassandra driver.

    When using authentication.transitional_mode, you can provide a blank username and password to log in with the anonymous role.

    When authentication and authorization are disabled, the credentials portion of the connection request is ignored. After authentication is enabled all connections must provide credentials.

  2. Change the replication factor for the security keyspaces.

  3. On each node, enable authentication and authorization in transitional mode:

    • Set authentication.transitional_mode to permissive in the cassandra.yaml file.

    • Set authorization.transitional_mode to normal in the cassandra.yaml file.

    For more information, see Enable HCD Unified Authentication.

  4. Enable logging so you can verify that applications are authenticating successfully.

  5. Enable JMX authentication.

  6. Perform a rolling restart of all nodes in the cluster.

    Nodes are vulnerable to malicious activity following this restart. Anybody can access the system using the default cassandra role with password cassandra. DataStax recommends using firewall rules to isolate the cluster until you disable the cassandra role.

  7. After all nodes restart, use cqlsh to replace the cassandra role with a non-default superuser role.

  8. Use the new superuser role to configure additional roles and assign permissions.

  9. Monitor audit logs to verify that all applications can access nodes and have the permissions required to execute requests.

  10. In the cassandra.yaml file, disable transitional_mode, and then configure specific authentication schemes, such as LDAP, if needed.

  11. Perform a rolling restart of the cluster.

Was this helpful?

Give Feedback

How can we improve the documentation?

© Copyright IBM Corporation 2026 | Privacy policy | Terms of use Manage Privacy Choices

Apache, Apache Cassandra, Cassandra, Apache Tomcat, Tomcat, Apache Lucene, Apache Solr, Apache Hadoop, Hadoop, Apache Pulsar, Pulsar, Apache Spark, Spark, Apache TinkerPop, TinkerPop, Apache Kafka and Kafka are either registered trademarks or trademarks of the Apache Software Foundation or its subsidiaries in Canada, the United States and/or other countries. Kubernetes is the registered trademark of the Linux Foundation.

General Inquiries: Contact IBM