I was also facing the same problem. I was using "replication.latency.ms" jmx metric which does not get reset. I started using "replication.latency.avg" which resets when there is no new message to replicate.
On Thu, Oct 1, 2020, 7:06 PM Shiva Sagar Reddy Veeram <saagar2...@gmail.com> wrote: > I’ve noticed the replication latencies are not reset when the destination > topic is all caught up and there is no more data to replicate. So polling > metrics after still returns the last replication latency that was recorded > when there was data. Is there a way to reset the metrics once replication > is done ? We use latency for alerting purposes. > > Thanks, > Sag > > Sent from my iPhone
