nicktelford commented on code in PR #15264:
URL: https://github.com/apache/kafka/pull/15264#discussion_r1476175960
##########
streams/src/main/java/org/apache/kafka/streams/KafkaStreams.java:
##########
@@ -1343,6 +1348,17 @@ private long getCacheSizePerThread(final int
numStreamThreads) {
return totalCacheSize / (numStreamThreads +
(topologyMetadata.hasGlobalTopology() ? 1 : 0));
}
+ private long getMaxUncommittedBytesPerThread(final int numStreamThreads) {
Review Comment:
> Does passing maxUncommittedBytesPerThread already affect commit behavior?
As far as I understand the code it does not because
approximateNumUncommittedBytes() returns always 0 at the moment.
No, for exactly this reason.
> This method does not consider whether default.state.isolation.level is set
to READ_COMMITTED.
Does it need to? Under READ_UNCOMMITTED, since transaction buffers aren't
used, `approximateNumUncommittedBytes()` will always return `0` (as it does
right now), ensuring that we never exceed the configured threshold.
Notably, if we make this dependent on the isolation level, any future
changes that add support for transaction buffers to READ_UNCOMMITTED writes
would break this.
--
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
