zuston commented on code in PR #1445:
URL:
https://github.com/apache/incubator-uniffle/pull/1445#discussion_r1477194993
##########
client-spark/common/src/main/java/org/apache/spark/shuffle/RssSparkConfig.java:
##########
@@ -347,6 +347,13 @@ public class RssSparkConfig {
+ " of concurrent tasks."))
.createWithDefault(RssClientConfig.RSS_ESTIMATE_SERVER_ASSIGNMENT_ENABLED_DEFAULT_VALUE);
+ public static final ConfigEntry<Boolean> RSS_TASK_FAILED_CALLBACK_ENABLED =
Review Comment:
Please use the unified config style.
##########
client-spark/spark3/src/main/java/org/apache/spark/shuffle/RssShuffleManager.java:
##########
@@ -1122,47 +1109,21 @@ public void addFailuresShuffleServerInfos(String
shuffleServerId) {
public synchronized boolean reassignShuffleServers(
Review Comment:
Can you rename this method to `reassignAllShuffleServersForWholeStage`
,current name is similar with the `reassignFaultyShuffleServer`, that is not
clear
##########
client-spark/spark3/src/main/java/org/apache/spark/shuffle/RssShuffleManager.java:
##########
@@ -1187,6 +1148,94 @@ public synchronized boolean reassignShuffleServers(
}
}
+ @Override
+ public ShuffleServerInfo reassignFaultyShuffleServer(
Review Comment:
Why not using `synchronized` , if having multiple request happen at the same
time?
And if the same shuffleServerID is faulty from different tasks, we should
make it use the same assignment rule from the first retry assign here?
##########
client/src/main/java/org/apache/uniffle/client/util/RssClientConfig.java:
##########
@@ -93,4 +93,5 @@ public class RssClientConfig {
public static final int
RSS_ESTIMATE_TASK_CONCURRENCY_PER_SERVER_DEFAULT_VALUE = 80;
public static final String RSS_RESUBMIT_STAGE = "rss.resubmit.stage";
+ public static final String RSS_TASK_FAILED_CALLBACK_ENABLED =
"rss.task.failed.callback.enabled";
Review Comment:
ditto.
##########
client-spark/spark3/src/main/java/org/apache/spark/shuffle/RssShuffleManager.java:
##########
@@ -1187,6 +1148,94 @@ public synchronized boolean reassignShuffleServers(
}
}
+ @Override
+ public ShuffleServerInfo reassignFaultyShuffleServer(
Review Comment:
How about naming `reassignFaultyShuffleServerForTasks`
##########
client/src/main/java/org/apache/uniffle/client/impl/FailedBlockSendTracker.java:
##########
@@ -0,0 +1,75 @@
+/*
+ * 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.uniffle.client.impl;
+
+import java.util.Collection;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import com.google.common.collect.Maps;
+import com.google.common.collect.Sets;
+
+import org.apache.uniffle.common.ShuffleBlockInfo;
+import org.apache.uniffle.common.ShuffleServerInfo;
+import org.apache.uniffle.common.rpc.StatusCode;
+
+public class FailedBlockSendTracker {
+
+ private Map<Long, Set<TrackBlockStatus>> trackBlockStatusMap;
Review Comment:
Why using the `hashset` data structure? Maybe linkedlist will be better,
that will show the real pipeline of the failure chain.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
