Thanks for the KIP Kevin! The main issue with this metric is that its value is the difference between > a broker's local system time at the time of metric collection and the > leader's append timestamp of the broker's last applied record. This means > the frequency/interval of metrics collection determines this metric's > value, which may present challenges for monitoring the metric.
I might not call it an issue with the metric, just that the metric measures one thing (lag) and that might be unintuitive if we're actually looking for latency. (I think this is what you're getting at with "challenges for monitoring the metric" too.) For instance, what your new metric won't give a good indicator on (that the lag metric does) is if the last applied record on the broker was from a while back. Alyssa On Wed, Feb 18, 2026 at 9:18 AM Kevin Wu <[email protected]> wrote: > Hi all, > > I would like to start a discussion on a KIP to add a latency metric for the > last applied metadata image on brokers. > > KIP link: > > https://cwiki.apache.org/confluence/display/KAFKA/KIP-1286%3A+Add+last-applied-image-latency+metric+for+brokers > > Best, > Kevin Wu >
