we've done as much tuning as we can think of to give our servers resources to monitor process, but we're still running into a issue with max_events getting crushed by an ever increasing process_events log