Have been looking around for a way to reverse scale pods with KEDA, so reduce number of pods the larger a rabbitmq queue is.
Our idea was to reduce the number of consumers to our rabbitmq ingest queue, if another queue is getting to big - thereby allowing the rest of the system to catch up in case a service goes down and slows our whole pipeline down.
2.1m questions
2.1m answers
60 comments
57.0k users