Hi,
I can see in our grafana that kong_db_cache & kong_process_events is always 100% on our nodes.
Each kong node has 4GB of memory, with 1.7GB memory available (1.7GB in Buff/cache).
I haven’t been able to find anything related to these two measurement.