mridulm commented on code in PR #2609:
URL: https://github.com/apache/celeborn/pull/2609#discussion_r1722710447
##########
client/src/main/java/org/apache/celeborn/client/read/CelebornInputStream.java:
##########
@@ -388,6 +391,15 @@ private ByteBuf getNextChunk() throws IOException {
throw new CelebornIOException(
"Fetch data from excluded worker! " +
currentReader.getLocation());
}
+
+ // read one chunk first, then throw CelebornIOException to let spark
rerun stage
+ if (testRandomPushForStageRerun && shuffleId == 0 &&
!alreadyReadChunk) {
+ alreadyReadChunk = true;
+ } else if (testRandomPushForStageRerun && shuffleId == 0 &&
alreadyReadChunk) {
+ alreadyReadChunk = false;
+ throw new CelebornIOException("already read chunk");
+ }
Review Comment:
super nit:
```suggestion
if (testRandomPushForStageRerun) {
if (shuffleId == 0 && !alreadyReadChunk) {
alreadyReadChunk = true;
} else if (shuffleId == 0 && alreadyReadChunk) {
alreadyReadChunk = false;
throw new CelebornIOException("already read chunk");
}
}
```
##########
client-spark/common/src/main/java/org/apache/spark/shuffle/celeborn/SparkCommonUtils.java:
##########
@@ -0,0 +1,53 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.shuffle.celeborn;
+
+import org.apache.spark.SparkConf;
+import org.apache.spark.TaskContext;
+import org.apache.spark.scheduler.DAGScheduler;
+
+public class SparkCommonUtils {
+ public static void validateAttemptConfig(SparkConf conf) throws
IllegalArgumentException {
+ int maxStageAttempts =
+ conf.getInt(
+ "spark.stage.maxConsecutiveAttempts",
+ DAGScheduler.DEFAULT_MAX_CONSECUTIVE_STAGE_ATTEMPTS());
+ // In Spark 2, the parameter is referred to as MAX_TASK_FAILURES, while in
Spark 3, it has been
+ // changed to TASK_MAX_FAILURES. The default value for both is
consistently set to 4.
+ int maxTaskAttempts = conf.getInt("spark.task.maxFailures", 4);
+ if (maxStageAttempts >= (1 << 15) || maxTaskAttempts >= (1 << 16)) {
+ // The map attemptId is a non-negative number constructed from
+ // both stageAttemptNumber and taskAttemptNumber.
+ // The high 16 bits of the map attemptId are used for the
stageAttemptNumber,
+ // and the low 16 bits are used for the taskAttemptNumber.
+ // So spark.stage.maxConsecutiveAttempts should be less than 32768 (1 <<
15)
+ // and spark.task.maxFailures should be less than 65536 (1 << 16).
+ throw new IllegalArgumentException(
+ "The spark.stage.maxConsecutiveAttempts should be less than 32768
(currently "
+ + maxStageAttempts
+ + ")"
+ + "and spark.task.maxFailures should be less than 65536
(currently "
+ + maxTaskAttempts
+ + ").");
+ }
+ }
+
+ public static int getEncodedAttemptNumber(TaskContext context) {
+ return (context.stageAttemptNumber() << 16) | context.attemptNumber();
Review Comment:
As we discussed earlier (I cant seem to find the ref :-) ) - please do
submit a PR to Apache Spark as well for this - and ensure the communities can
align on this assumption.
--
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]
