[KAFKA-16029] Investigate cause of "Unable to find FetchSessionHandler for node X" in logs

• Type: Bug
• Status: Resolved
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 3.7.0
• Fix Version/s: 3.7.0
• Component/s: clients, consumer
• Labels: consumer-threading-refactor fetcher
• Assignee: Kirk True
• Reporter: Kirk True
• Reviewer: David Jacot
• Created: 18/Dec/23 15:08
• Updated: 30/Jan/24 16:58
• Resolved: 24/Jan/24 12:57

Description

From mjsax:

Looking into AK unit/integration test logs for KS, I regularly see an ERROR log line that is triggered here:

https://github.com/apache/kafka/blob/trunk/clients/src/main/java/org/apache/kafka/clients/consumer/internals/AbstractFetch.java#L150-L151

Given that it seems not to impact the test (it's not failing because of this), I am wondering why we log this at ERROR level or if it might be better to reduce to WARN? It seems to happen fairly frequently, but it also seems that it's nothing one would need to be concerned about, and thus using ERROR might be more alerting to end users than it needs to be? Thoughts?

According to Matthias, the running the EosIntegrationTest locally reproduces the log line.

This is also reproducible by running the Apache Kafka quickstart.

Issue Links

• links to: GitHub Pull Request #15186