AWS Security ChangesHomeSearch

AWS msk: MSK KRaft controller DNS endpoint and monitoring guidance update

Service: msk · 2026-10-23 · Documentation low

File: msk/latest/developerguide/metadata-management.md

Summary

Adds a link to the 'Kafka cluster ID for KRaft clusters' page and rewrites the note about KRaft controller DNS endpoints, explaining that clusters created on Apache Kafka 4.1+ (or migrated from ZooKeeper) have no controller endpoints, so controller JMX metrics cannot be collected via open monitoring; recommends scraping broker endpoints and using CloudWatch metrics like ActiveControllerCount. Also renames a related-links entry.

Security assessment

The change is purely operational/observability documentation about KRaft controller DNS endpoints and open-monitoring metric collection. It does not address a vulnerability, credential handling, encryption, IAM, or exposure; monitoring guidance is only tangentially security-adjacent, so severity is low.

Evidence

+  * You don't need to access your cluster's KRaft controllers directly for any administrative or monitoring actions. Amazon MSK monitors and manages the controllers for you, and you don't need controller metrics to monitor the health of your cluster. Whether your cluster has controller DNS endpoints depends on how it was created:

Diff

diff --git a/msk/latest/developerguide/metadata-management.md b/msk/latest/developerguide/metadata-management.md
index af7e6c0..001ce4f 100644
--- a//msk/latest/developerguide/metadata-management.md
+++ b//msk/latest/developerguide/metadata-management.md
@@ -20,0 +21,2 @@ From Apache Kafka version 3.7.x on Amazon MSK, you can create clusters that use
+  * [Kafka cluster ID for KRaft clusters](./kraft-cluster-id.html)
+
@@ -121 +123,7 @@ Here are some points to note about KRaft mode on MSK:
-  * You do not need to access your cluster's KRaft controllers directly for any administrative actions. However, if you are using open monitoring to collect metrics, you also need the DNS endpoints of your controllers in order to collect some non-controller related metrics about your cluster. You can get these DNS endpoints from the MSK Console or using the [ListNodes](https://docs.aws.amazon.com/msk/1.0/apireference/clusters-clusterarn-nodes.html#ListNodes) API operation. See [Monitor an MSK Provisioned cluster with Prometheus](./open-monitoring.html) for updated steps for setting up open-monitoring for KRaft-based clusters.
+  * You don't need to access your cluster's KRaft controllers directly for any administrative or monitoring actions. Amazon MSK monitors and manages the controllers for you, and you don't need controller metrics to monitor the health of your cluster. Whether your cluster has controller DNS endpoints depends on how it was created:
+
+    * KRaft clusters created on Apache Kafka versions earlier than 4.1 have controller DNS endpoints. These endpoints are retained when you upgrade the cluster to a later Apache Kafka version.
+
+    * KRaft clusters created on Apache Kafka version 4.1 or later don't have controller DNS endpoints. Clusters migrated from ZooKeeper mode to KRaft mode also don't have them.
+
+For clusters without controller DNS endpoints, the [ListNodes](https://docs.aws.amazon.com/msk/1.0/apireference/clusters-clusterarn-nodes.html#ListNodes) API operation doesn't return endpoints for controller nodes. As a result, you can't collect controller JMX metrics with open monitoring. For open monitoring, scrape your broker endpoints and use Amazon MSK CloudWatch metrics, such as `ActiveControllerCount`, for cluster-level health. For more information, see [Monitor an MSK Provisioned cluster with Prometheus](./open-monitoring.html).
@@ -140 +148 @@ CloudTrail events
-Migrate from ZooKeeper to KRaft mode
+Kafka cluster ID for KRaft clusters