Issue Summary

There is a known issue with importing large datasets in Assets for Data Center in a short period of time that can flood the clustermessage table with INSIGHT_DATA_CACHE. Due to this volume, several potential causes may prevent the service ClusterMessageCleaningService from deleting the messages, including:

• Rapid message generation: If messages are being created faster than they can be deleted (e.g., during large imports, automation etc.)
• Insufficient cleanup frequency: Default daily interval might be too infrequent for high-volume environments
• Index contention: Heavy read/write activity on the clustermessage table during cleanup

Steps to Reproduce

1. Have a large dataset for clustermessage table with over 180M rows for ticket connections.
2. Configure data retention period for clustermessage table
   • RETENTION_PERIOD: 2880m
   • Execution time: daily at 03:30 AM

Expected Results

The only criteria or condition it checks is MESSAGE_TIME < (current_time - retention_period) where the retention period can be configured in the advanced settings

All records older than the retention_period should be deleted

Actual Results

No rows from clustermessage are being deleted.

Workaround

We recommend the customers to delete old entries in the clustermessage table for messages relating to INSIGHT. Example query below, where 'message_time' should be replaced!

DELETE FROM clustermessage WHERE message like '%INSIGHT%' AND message_time < '2025-06-06 00:00:00'

Once this is done, we recommend the customer to perform Assets reindexing.

Always back up your data before making any database modifications. If possible, test any alter, insert, update, or delete SQL commands on a staging server first.

Versions Tested

Please confirm all versions that have been tested for this issue, and indicate whether the tested version is affected or not affected, below: