poorbarcode commented on code in PR #20179:
URL: https://github.com/apache/pulsar/pull/20179#discussion_r1209955441
##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/service/persistent/PersistentStickyKeyDispatcherMultipleConsumers.java:
##########
@@ -65,23 +67,22 @@ public class PersistentStickyKeyDispatcherMultipleConsumers
extends PersistentDi
private final KeySharedMode keySharedMode;
/**
- * When a consumer joins, it will be added to this map with the current
read position.
- * This means that, in order to preserve ordering, new consumers can only
receive old
- * messages, until the mark-delete position will move past this point.
+ * When a consumer joins, it will be added to this map with the current
last sent position per the message key.
+ * This means that, in order to preserve ordering per the message key, new
consumers can only receive old
+ * messages, until the mark-delete position will move past this point in
the key. New consumers can receive
+ * any new messages with the message key that is not in the last sent
position.
*/
- private final LinkedHashMap<Consumer, PositionImpl>
recentlyJoinedConsumers;
-
- private final Set<Consumer> stuckConsumers;
- private final Set<Consumer> nextStuckConsumers;
+ private final LinkedHashMap<Consumer, LastSentPositions>
recentlyJoinedConsumers;
+ // The lastSentPosition is not thread-safe
+ private final LastSentPositions lastSentPositions;
PersistentStickyKeyDispatcherMultipleConsumers(PersistentTopic topic,
ManagedCursor cursor,
Review Comment:
@equanz
Sorry, I'm not trying to stand in the way of something. the above comments
are based on the premise that the mechanism of Key_Shared does not become more
complicated if the goal of this PR is to fix the above three scenarios, I think
`lock` is a simpler choice.
I also think recording the mapping of the key and last send position is a
good way to prevent the stuck between multi-consumers, but since this adds
complexity, a PIP is needed.
--
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]
