The DCAE ONSET flood prevention mechanism do not apply to the newly developed Frankfrut controller as it was supported in the "usecases" controller.   The reason is because the ONSETs in the working memory are short lived and retracted shortly after insertion.

It's seen in labs (and junits) that 2 "canonical onsets" representing the same event are processed in parallel even though if they represent the same alarm.

This causes a performance degradation in vDNS, and vFW use cases as duplicated ONSET events are pre-processed though AAI, and PDP-X to them being dropped when determined the resource is locked.