There was interruption between GCP's monitoring & services on GCP. I also have my own Prometheus/Grafana stack in there and it kept chugging along fine, even t

There was interruption between GCP's monitoring & services on GCP. I also have my own Prometheus/Grafana stack in there and it kept chugging along fine, even though their DB graphs show severe dips in tx/sec in the time period in question.