Hello, recently we upgraded kafka from 2.8.0 to 3.7.0 and noticed a 5x value increase from the kafka.request.produce.time.avg datadog metric. There was also a considerable increase in the kafka.request.metadata.time.avg value. It is worth noting the 99percentile metrics are comparable to before the upgrade. We upgraded zookeeper and confluent-kafka clients to the newest stable version after seeing these issues, which didn't resolve the issue. Any ideas on what could have caused this change between versions, and steps to resolve this?
