Troubleshooting
Problem
In IBM Cloud Pak for Watson AIOps 4.12, ELK data is not being processed, resulting in the error 'RecordTooLargeException' due to Kafka rejecting log messages exceeding the maximum allowed size of 1 MB. The issue arises when using a wildcard (*) endpoint in the Elasticsearch URL, causing bulk data aggregation and oversized messages to be sent to Kafka.
Symptom
The customer is experiencing issues with OpenSearch integration in AIOPS 4.12, where topology data ingestion is not occurring due to empty Kafka topics related to topology inputs and outputs. The ELK log data processing is also not functioning, with empty training data sets and log indices. Configuration issues, such as insufficient ephemeral storage allocation, were identified as contributing factors. The customer also faced challenges with executing certain scripts and encountered permission errors when running curl commands.
Document Location
Worldwide
Log InLog in to view more of this document
Was this topic helpful?
Document Information
Modified date:
18 June 2026
UID
ibm17277075