[GitHub] [spark] juliuszsompolski commented on a diff in pull request #42355: [SPARK-44709][CONNECT] Run ExecuteGrpcResponseSender in reattachable execute in new thread to fix flow control

["juliuszsompolski (via GitHub)" <gi...@apache.org>]

juliuszsompolski commented on code in PR #42355:
URL: https://github.com/apache/spark/pull/42355#discussion_r1286441954


##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/execution/ExecuteResponseObserver.scala:
##########
@@ -85,12 +85,18 @@ private[connect] class ExecuteResponseObserver[T <: MessageLite](val executeHold
    */
   private var responseSender: Option[ExecuteGrpcResponseSender[T]] = None
 
+  // Statistics about cached responses.
+  private var cachedSizeUntilHighestConsumed = CachedSize()
+  private var cachedSizeUntilLastProduced = CachedSize()
+  private var autoRemovedSize = CachedSize()
+  private var totalSize = CachedSize()
+
   /**
    * Total size of response to be held buffered after giving out with getResponse. 0 for none, any
    * value greater than 0 will buffer the response from getResponse.
    */
   private val retryBufferSize = if (executeHolder.reattachable) {
-    SparkEnv.get.conf.get(CONNECT_EXECUTE_REATTACHABLE_SENDER_MAX_STREAM_SIZE).toLong
+    SparkEnv.get.conf.get(CONNECT_EXECUTE_REATTACHABLE_OBSERVER_RETRY_BUFFER_SIZE).toLong

Review Comment:
   this was a bug of using wrong config...



-- 
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: reviews-unsubscribe@spark.apache.org

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