[ https://issues.apache.org/jira/browse/KAFKA-12713?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17399940#comment-17399940 ]
Kai Huang commented on KAFKA-12713: ----------------------------------- [~ijuma] I would like to follow up on this ticket, and continue the discussion. I replied to your [discussion|https://lists.apache.org/thread.html/r7f82dde9133bf9d3a8b688ca7ae02ad761c52e7b79212c9247b276c5%40%3Cdev.kafka.apache.org%3E] thread and illustrate how the fetch latency metric should work in [KIP-736|https://cwiki.apache.org/confluence/display/KAFKA/KIP-736%3A+Report+the+true+end+to+end+fetch+latency]. Could you please take a look and see if that clarifies your question? > Report "REAL" follower/consumer fetch latency > --------------------------------------------- > > Key: KAFKA-12713 > URL: https://issues.apache.org/jira/browse/KAFKA-12713 > Project: Kafka > Issue Type: Bug > Reporter: Ming Liu > Assignee: Kai Huang > Priority: Major > > The fetch latency is an important metrics to monitor for the cluster > performance. With ACK=ALL, the produce latency is affected primarily by > broker fetch latency. > However, currently the reported fetch latency didn't reflect the true fetch > latency because it sometimes need to stay in purgatory and wait for > replica.fetch.wait.max.ms when data is not available. This greatly affect the > real P50, P99 etc. > I like to propose a KIP to be able track the real fetch latency for both > broker follower and consumer. > -- This message was sent by Atlassian Jira (v8.3.4#803005)