zhixinwen commented on code in PR #3075:
URL: https://github.com/apache/kvrocks/pull/3075#discussion_r2248508006
##########
src/cluster/replication.cc:
##########
@@ -203,6 +206,21 @@ void FeedSlaveThread::loop() {
Stop();
return;
}
+
+ // Check if this change would unblock any WAIT command
+ auto largest_unblockable_seq =
srv_->LargestTargetSeqToWakeup(batch.sequence);
+ if (largest_unblockable_seq > last_replconf_getack_seq_) {
+ // Send _getack to the slave to get acknowledgment
+ auto s = util::SockSend(conn_->GetFD(), redis::BulkString("_getack"),
conn_->GetBufferEvent());
+ if (!s.IsOK()) {
+ error("Write error while sending _getack to slave: {}", s.Msg());
+ Stop();
+ return;
+ } else {
+ last_replconf_getack_seq_ = largest_unblockable_seq;
Review Comment:
```
Should we set last_replconf_getack_seq_ to largest_unblockable_seq when the
value of largest_unblockable_seq is zero?
```
I am not sure if I get you.
`last_replconf_getack_seq_` is used to prevent unnecessary `_getack` sent.
If we set it back to zero, it would cause unnecessary sends.
The check in `largest_unblockable_seq > last_replconf_getack_seq_` ensures
we would not set `last_replconf_getack_seq_` back to zero when
`largest_unblockable_seq` is zero.
--
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]
