[GitHub] [spark] AngersZhuuuu commented on a change in pull request #30139: [SPARK-31069][CORE] high cpu caused by chunksBeingTransferred in external shuffle service

[GitBox <gi...@apache.org>]

AngersZhuuuu commented on a change in pull request #30139:
URL: https://github.com/apache/spark/pull/30139#discussion_r519652381



##########
File path: common/network-common/src/main/java/org/apache/spark/network/server/ChunkFetchRequestHandler.java
##########
@@ -88,12 +88,14 @@ public void processFetchRequest(
       logger.trace("Received req from {} to fetch block {}", getRemoteAddress(channel),
         msg.streamChunkId);
     }
-    long chunksBeingTransferred = streamManager.chunksBeingTransferred();
-    if (chunksBeingTransferred >= maxChunksBeingTransferred) {
-      logger.warn("The number of chunks being transferred {} is above {}, close the connection.",
-        chunksBeingTransferred, maxChunksBeingTransferred);
-      channel.close();
-      return;
+    if (maxChunksBeingTransferred < Long.MAX_VALUE) {

Review comment:
       > If the concern is we may miss some places that update the state, then maybe only keeping this change is safer.
   
   OK, so I will just revert change about  `if (maxChunksBeingTransferred < Long.MAX_VALUE) {`




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org