mumrah commented on PR #14028: URL: https://github.com/apache/kafka/pull/14028#issuecomment-1640226629
Thanks for taking a look @showuon! > MetadataType has 2 kinds of value, one is for brokers, and the other is for controllers. The one for controllers contain a 3rd value: Dual (3), which is missing from this PR. Since we already have ZkMigrationState, I think the "Dual" state is a bit redundant. I think it's simpler to let KafkaServer/MetadataType and KafkaController/MetadataType have the same values > From the KIP, many metrics only reported by active controller, but I can see the value are reported by all controllers, including followers. Is it expected? Ah, that's probably just an oversight in the KIP. All controllers should report metrics. For metadata-derived metrics (like number of migrating brokers), each controller will report the same value. For runtime metrics like the dual-write timings, only the active controller will be updating this metric (the inactive controllers will report zeros) > There's still Features metric is missing from current implementation. Will that be added in a follow-up PR? Yea, I think a follow-up PR makes sense for this one. It's really a bit less important than the others for monitoring a migration. Once we close on this PR, I'll update the KIP with any small differences -- 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
