[ https://issues.apache.org/jira/browse/KAFKA-3442?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15206990#comment-15206990 ]
Jiangjie Qin commented on KAFKA-3442: ------------------------------------- I am thinking about the same thing if we want to maintain the behavior, that should be trivial to do. I'll update the patch. I agree that we have to check the client code path to make sure the per partition error code handling still works if we return RecordSizeTooLargeException. Protocol wise, I also agree it is less risky if we bump up version for FetchRequest because we are returning a new error code although RecordSizeTooLargeException has already existed in the ApiExceptions. > FetchResponse size exceeds max.partition.fetch.bytes > ---------------------------------------------------- > > Key: KAFKA-3442 > URL: https://issues.apache.org/jira/browse/KAFKA-3442 > Project: Kafka > Issue Type: Bug > Affects Versions: 0.10.0.0 > Reporter: Dana Powers > Assignee: Jiangjie Qin > Priority: Blocker > Fix For: 0.10.0.0 > > > Produce 10000 byte message to topic foobar > Fetch foobar w/ max.partition.fetch.bytes=1024 > Test expects to receive a truncated message (~1024 bytes). 0.8 and 0.9 pass > this test, but 0.10 FetchResponse has full message, exceeding the max > specified in the FetchRequest. > I tested with v0 and v1 apis, both fail. Have not tested w/ v2 -- This message was sent by Atlassian JIRA (v6.3.4#6332)