Jason Gustafson created KAFKA-9886:
--------------------------------------
Summary: Validate segment range before reading in `Log.read`
Key: KAFKA-9886
URL: https://issues.apache.org/jira/browse/KAFKA-9886
Project: Kafka
Issue Type: Improvement
Reporter: Jason Gustafson
Assignee: Jason Gustafson
Log.read uses the following logic to set the upper limit on a segment read.
{code}
val maxPosition = {
// Use the max offset position if it is on this segment; otherwise, the
segment size is the limit.
if (maxOffsetMetadata.segmentBaseOffset == segment.baseOffset) {
maxOffsetMetadata.relativePositionInSegment
} else {
segment.size
}
}
{code}
In the else branch, the expectation is that
`maxOffsetMetadata.segmentBaseOffset > segment.baseOffset`. In KAFKA-9838, we
found a bug where this assumption failed which led to reads above the high
watermark. We should validate the expectation explicitly so that we don't leave
the door open for similar bugs in the future.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
