jiafu1115 commented on code in PR #20203:
URL: https://github.com/apache/kafka/pull/20203#discussion_r2462709991
##########
core/src/main/scala/kafka/server/BrokerServer.scala:
##########
@@ -591,6 +600,15 @@ class BrokerServer(
"all of the SocketServer Acceptors to be started",
enableRequestProcessingFuture, startupDeadline, time)
+ brokerReadyCallbacks.foreach { callback =>
+ try {
+ callback.onBrokerReady()
Review Comment:
@chia7712
Good question. For external clusters, we also delay the initialization, even
though it’s not strictly necessary — but it doesn’t seem to have any negative
effect either.
That said, it did remind me of one thing: the current updated Javadoc says
“begins after the broker is ready to handle requests” but it should explicitly
clarify that this refers to the current broker, not others. I think we should
update the Javadoc to make that clear;
The current code change actually keeps the design consistent for both local
and external clusters when used as remote storage. Considering that local
clusters are usually the more common setup — given their simpler deployment and
lower latency — this approach seems reasonable. So I think updating the Javadoc
should be sufficient. What do you think? Any better idea？
I already updated the javadoc
(https://github.com/apache/kafka/pull/20203/files#diff-38f94f707537f1c07dc8bb3d6e35544beeaed67f7ab46243d5f9eaf8f55145ecR72).
You can help to take a look at it. Thanks
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
