Managed Apache Iceberg™ tables: Replicated by default in replication and failover groups (Pending)

Attention

This behavior change is in the 2026_05 bundle.

For the current status of the bundle, refer to Bundle history.

When this behavior change bundle is enabled, managed Iceberg tables are replicated by default as part of a replication or failover group, without the Public Preview parameters that were previously required.

Before the change:

Managed Iceberg tables are replicated only when the Public Preview parameters are enabled. When those parameters are disabled, managed Iceberg tables are silently skipped: the replication or failover group succeeds without replicating them.

After the change:

Managed Iceberg tables are replicated by default whenever they exist in a database or schema that’s included in the replication or failover group. The Public Preview parameters are no longer required.

Because these tables are now in scope, the external volumes they depend on must be replicated and configured for access from the secondary account. If an external volume isn’t replicated, or is misconfigured for the secondary account, the refresh of the replication or failover group fails until you address the external volume.

What you need to do

If you aren’t already onboarded to the Public Preview for managed Iceberg tables, take action before this change takes effect.

First, identify the managed Iceberg tables in the databases that are included in your replication or failover group. Run the following query, replacing the database names with the databases in your group:

SHOW ICEBERG TABLES IN ACCOUNT;

SELECT "database_name", "schema_name", "name",
       "external_volume_name", "catalog_name", "iceberg_table_type"
FROM TABLE(RESULT_SCAN(LAST_QUERY_ID()))
WHERE "iceberg_table_type" = 'MANAGED'
  AND "database_name" IN ('DB_ONE', 'DB_TWO', 'DB_THREE')
ORDER BY 1, 2, 3;

The results show each managed Iceberg table along with the external volume it depends on, so you know which external volumes must be replicated and configured for the secondary account.

To replicate your managed Iceberg tables (recommended for disaster recovery):

  • Configure your replication or failover group to replicate the external volumes that the Iceberg tables use, and make sure storage locations are configured for both the primary and secondary accounts.
  • Make sure those external volumes are set up to be accessed by both the primary and secondary accounts.

To opt out of replicating them:

  • Move the managed Iceberg tables out of the databases that are included in the replication or failover group.

For managed Iceberg tables, to avoid replication failures:

  • Don’t delete any data files directly, and don’t add lifecycle policies to the data files.
  • Don’t revoke or change access to the external volume’s storage location. For example, don’t edit the bucket or IAM policy in a way that prevents the secondary account from reading or writing to the storage location.

For help:

Ref: 2341