Skip to content

Troubleshooting DR Setups

This page covers common user-facing issues when creating, configuring, and operating DR setups.

Cause: The DR license module is not enabled on your license.

Resolution: Contact your administrator to enable the DR module. The DR tab on the Dashboard and the + New DR Setup button will not appear until the module is active.

CauseResolution
Missing Create permissionRequest the Create permission from your administrator
DR setup quota reachedContact your administrator to increase the maximum number of DR setups on the license

Cluster Linking is not supported on all Kafka platforms. If a cluster fails the compatibility check during the topology step:

PlatformCompatibleResolution
Confluent Cloud; Basic tierNoUpgrade to Dedicated or Enterprise tier
Confluent Cloud; Standard tierNoUpgrade to Dedicated or Enterprise tier
WarpStreamNoUse a Confluent Cloud or Confluent Platform cluster
Amazon MSKNoUse a Confluent Cloud or Confluent Platform cluster
Open Source KafkaNoUse a Confluent Cloud or Confluent Platform cluster

Cause: The cluster is unreachable due to a network issue, incorrect URL, or credential mismatch.

Resolution: Navigate to Pre-Migration Setup -> Kafka Clusters, verify the cluster’s bootstrap URL and credentials, and re-run the connection test. The topology step retests automatically when the cluster is added.

Cause: The schema registry URL or credentials are incorrect.

Resolution: Navigate to Pre-Migration Setup -> Schema Registry, verify the URL and credentials, and re-test. The topology step checks each assigned schema registry when schema linking is enabled.

Cause: A WarpStream cluster is present in the topology. Schema linking is not supported for WarpStream.

Resolution: Remove the WarpStream cluster from the topology, or use a topology that does not include WarpStream clusters.

Cause: ACL Sync is enabled and a mirror prefix is set on the same link. These two options are mutually exclusive.

Resolution: In the Infrastructure Settings step, either disable ACL Sync or remove the mirror prefix for the affected link. The wizard will block Continue until the conflict is resolved.

Cause: The number of selected topics, consumer groups, or topic partitions exceeds the limit on your license.

Resolution: Reduce the selection scope (use wildcard patterns to be more selective, or switch from All to Selected) or contact your administrator to increase license limits.

CheckCauseResolution
Target cluster reachable; FailTarget cluster bootstrap is downRestore cluster connectivity before proceeding. This is a hard fail and cannot be overridden.
All link directions active; BlockOne or more directions are Paused, Degraded, or FailedInspect link health in the Monitor tab. Resolve the link issue or use Emergency Failover if the source is unavailable.
Replication lag within RPO; BlockActive SLA breaches existWait for lag to drain below the RPO threshold, or use Emergency Failover if the primary is unavailable and data loss is acceptable.

Block-level checks (not hard fails) can be overridden with explicit operator acknowledgement when no hard fail checks are present.

Cause: The DR setup’s status is Draft, Configured, or Decommissioned: none of which have active replication.

Resolution: Start replication from the Overview tab. The Monitor tab appears once the status transitions to Replicating, Paused, Degraded, or any in-progress state.

Cause: The Manage Links permission is not assigned to your role.

Resolution: Request the Manage Links permission from your administrator. Without it, the Topics, Consumer Groups, and Schemas sub-tabs in the Monitor tab link accordions are not accessible.

Cause: The Failover permission is not assigned, or the DR status does not support the action.

Resolution: Request the Failover permission if the buttons are hidden. If the status is the issue, refer to the Lifecycle Actions availability table.

There are two independent causes:

CauseResolution
Auto-refresh is disabled or polling was paused (client-side only)Enable the auto-refresh toggle in the status header, or click manual refresh.
The DataReplicatorService health probe feed has stopped writing snapshots for longer than the staleness threshold, 20 seconds by default (backend)Shows a “Monitoring feed down” banner and flips all cluster health to unknown. Toggling auto-refresh will not clear this: restore the DataReplicatorService probe process. It clears automatically on the next snapshot once the feed resumes.

Action Buttons Not Appearing in Overview Tab

Section titled “Action Buttons Not Appearing in Overview Tab”

Lifecycle action buttons appear conditionally based on DR status. Refer to the Start, Pause, Decommission, and Delete page for the full availability table.