junrao commented on code in PR #14111: URL: https://github.com/apache/kafka/pull/14111#discussion_r1288946333
########## clients/src/main/java/org/apache/kafka/clients/Metadata.java: ########## @@ -144,13 +159,25 @@ public long metadataExpireMs() { */ public synchronized int requestUpdate() { this.needFullUpdate = true; + this.backoffUpdateRequests = 0L; Review Comment: > My understanding of the PR is this, metadata request won't backoff, but produce request would backoff. So likely metadata is going to be updated next time around produce request is retried(post backoff). @msn-tldr : To me, the common reason why a produce request needs to backoff is that the metadata is stable since the latest metadata hasn't been propagated to the brokers yet. So, if we don't backoff the metadata request, the returned metadata may still be stale, which won't help the backed off produce request. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: jira-unsubscr...@kafka.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org