Hello Pinot community, Today we faced an ingestion issue with one of our Pinot REALTIME tables. The table stopped consuming data from its Kafka topic without any config changes. Other tables were ingesting data as expected during this time. Symptoms: No new records were ingested into the table despite active upstream Kafka publishing. Errors: Pinot server returned 500 internal server error for endpoints like: Temporary Fix: We recreated the table with the same schema and stream config, after which ingestion resumed successfully. Also we haven't incorporated any authentication , so there would be no issue due to authentication What are the common root causes for ingestion halts limited to a single table, while other tables remain unaffected? Is table recreation the only resolution in such cases, or is there a way to recover without recreation?

Common causes are segment commit failure, connectivity issues Kafka (network or auth), bad schema. Check for the consuming segment name in server log, it should tell you exactly what happened.

There is no network or auth issue , as you said there maybe segment commit failure or bad schema , from where and how can I check that issue .

Server logs