[
https://issues.apache.org/jira/browse/KAFKA-7040?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16508763#comment-16508763
]
Dong Lin edited comment on KAFKA-7040 at 6/11/18 9:14 PM:
----------------------------------------------------------
[~luwang] Say message m0 is truncated in step 5. Is that message acked by
broker1 in before Broker0 accepts m0 in step 3? If no, it seems that the
message is produced with ack=1 and it is within Kafka's contract to truncate
it. This is a known behavior when we have leadership change in Kafka.
If m0 has been acked by broker1 in step 3, then broker0 should still be able to
fetch it again after broker1 after step 5, right?
was (Author: lindong):
[~luwang] Say message m0 is truncated in step 5. Is that message acked by
broker1 in before Broker0 accepts m0 in step 3? If no, it seems that the
message is produced with ack=1 and it is within Kafka's contract to truncate
it. This is a known behavior when we have leadership change in Kafka.
If m0 has been acked by broker1 in step 3, then broker0 should still be able to
fetch it again after broker1 after step 5, right?
> The replica fetcher thread may truncate accepted messages during multiple
> fast leadership transitions
> -----------------------------------------------------------------------------------------------------
>
> Key: KAFKA-7040
> URL: https://issues.apache.org/jira/browse/KAFKA-7040
> Project: Kafka
> Issue Type: Bug
> Reporter: Lucas Wang
> Priority: Minor
>
> Problem Statement:
> Consider the scenario where there are two brokers, broker0, and broker1, and
> there are two partitions "t1p0", and "t1p1"[1], both of which have broker1 as
> the leader and broker0 as the follower. The following sequence of events
> happened on broker0
> 1. The replica fetcher thread on a broker0 issues a LeaderEpoch request to
> broker1, and awaits to get the response
> 2. A LeaderAndISR request causes broker0 to become the leader for one
> partition t1p0, which in turn will remove the partition t1p0 from the replica
> fetcher thread
> 3. Broker0 accepts some messages from a producer
> 4. A 2nd LeaderAndISR request causes broker1 to become the leader, and
> broker0 to become the follower for partition t1p0. This will cause the
> partition t1p0 to be added back to the replica fetcher thread on broker0.
> 5. The replica fetcher thread on broker0 receives a response for the
> LeaderEpoch request issued in step 1, and truncates the accepted messages in
> step3.
> The issue can be reproduced with the test from
> https://github.com/gitlw/kafka/commit/8956e743f0e432cc05648da08c81fc1167b31bea
> [1] Initially we set up broker0 to be the follower of two partitions instead
> of just one, to avoid the shutting down of the replica fetcher thread when it
> becomes idle.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
