liangyepianzhou commented on code in PR #15137:
URL: https://github.com/apache/pulsar/pull/15137#discussion_r857653945
##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/transaction/pendingack/impl/MLPendingAckStore.java:
##########
@@ -292,6 +249,69 @@ public void addFailed(ManagedLedgerException exception,
Object ctx) {
return completableFuture;
}
+ private void handleMetadataEntry(PositionImpl logPosition,
PendingAckMetadataEntry pendingAckMetadataEntry) {
+ // store the persistent position in to memory
+ // store the max position of this entry retain
+ if (pendingAckMetadataEntry.getPendingAckOp() != PendingAckOp.ABORT
+ && pendingAckMetadataEntry.getPendingAckOp() !=
PendingAckOp.COMMIT) {
+ Optional<PendingAckMetadata> optional =
pendingAckMetadataEntry.getPendingAckMetadatasList()
+ .stream().max((o1, o2) ->
ComparisonChain.start().compare(o1.getLedgerId(),
+ o2.getLedgerId()).compare(o1.getEntryId(),
o2.getEntryId()).result());
+
+ optional.ifPresent(pendingAckMetadata -> {
+ PositionImpl nowPosition =
PositionImpl.get(pendingAckMetadata.getLedgerId(),
+ pendingAckMetadata.getEntryId());
+
+ if (nowPosition.compareTo(maxAckPosition) > 0) {
+ maxAckPosition = nowPosition;
+ }
+ if (logAppendTimes.get() > upperLimitOfLogAppendTimes) {
Review Comment:
>We use a LagBackoff to calculate the next maxIndexLag, more entries are
added to the entry log, the currentIndexLog increases, if reach the
maxIndexLag, to get the next maxIndexLag and reset the currentIndexLag.
>Or we can avoid resetting the currentIndexLag, just keep it increasing and
use a nextIndexLag. If reach the currentIndexLag == nextIndexLag, reset the
nextIndexLag.
The first implementation is a bit more readable than the second, I tend to
use the first implementation.
--
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]
