The HVR_STATS Table Purge Causes Reporting Delays

Issue

The HVR_STATS table purge causes reporting delays throughout the day.

Environment

• HVR 6
• Job: hvrstats

Resolution

To resolve this issue, use one of the following methods to control when the purge occurs:

• Change the retention policy to UNBOUNDED to disable automatic purging. You must then purge the HVR_STATS table manually.
• Schedule a cron job that runs the hvrstats command with the -pMEDIUM option at your preferred interval. For example:

  hvrstats -pMEDIUM <hubname>
  

  This applies the MEDIUM aging rules during the scheduled purge instead of after every statistics collection cycle.

With scheduled purging, the HVR_STATS table grows between purges, causing its size to fluctuate. In contrast, the built-in purge keeps the table size consistent through continuous operation. Ensure you have sufficient disk space for the table's peak size.

Cause

By default, the purge runs at the end of every statistics collection cycle. With you use the MEDIUM retention policy, this inline cleanup can cause database lock contention on the HVR_STATS table. These locks can interfere with monitoring workflows and cause reporting delays.