guanziyue commented on a change in pull request #4264:
URL: https://github.com/apache/hudi/pull/4264#discussion_r809748931
##########
File path:
hudi-common/src/main/java/org/apache/hudi/common/util/queue/BoundedInMemoryQueue.java
##########
@@ -204,6 +204,7 @@ private boolean expectMoreRecords() {
* singleton iterator for this queue.
*/
private Option<O> readNextRecord() {
+ checkIfInterrupted();
Review comment:
I do this so that consumer can stop asap when queue received a
termination signal. I just have a simple test, it seems that there is no
difference on perf if I do a sampling here to determine whether to check it or
not. Or we can simply remove it here. As far as I know, add sync to parquet
writer is much enough to solve this problem. The modification of
BoundedInMemoryQueue is to make memory executor stop gracefully. We may pick
one of them. Any suggestion for 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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
