MAPREDUCE-6724. Single shuffle to memory must not exceed Integer#MAX_VALUE. (Haibo Chen via gera)

(cherry picked from commit 6890d5b472320fa7592ed1b08b623c55a27089c6)

Conflicts:

	hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/test/java/org/apache/hadoop/mapreduce/task/reduce/TestMergeManager.java

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/MergeManagerImpl.java

@@ -99,7 +99,9 @@ public class MergeManagerImpl<K, V> implements MergeManager<K, V> {
 
   private long usedMemory;
   private long commitMemory;
-  private final long maxSingleShuffleLimit;
+
+  @VisibleForTesting
+  final long maxSingleShuffleLimit;
   
   private final int memToMemMergeOutputsThreshold; 
   private final long mergeThreshold;
@@ -187,10 +189,16 @@ public class MergeManagerImpl<K, V> implements MergeManager<K, V> {
 
     usedMemory = 0L;
     commitMemory = 0L;
-    this.maxSingleShuffleLimit = 
-      (long)(memoryLimit * singleShuffleMemoryLimitPercent);
-    this.memToMemMergeOutputsThreshold = 
-            jobConf.getInt(MRJobConfig.REDUCE_MEMTOMEM_THRESHOLD, ioSortFactor);
+    long maxSingleShuffleLimitConfiged =
+        (long)(memoryLimit * singleShuffleMemoryLimitPercent);
+    if(maxSingleShuffleLimitConfiged > Integer.MAX_VALUE) {
+      maxSingleShuffleLimitConfiged = Integer.MAX_VALUE;
+      LOG.info("The max number of bytes for a single in-memory shuffle cannot" +
+          " be larger than Integer.MAX_VALUE. Setting it to Integer.MAX_VALUE");
+    }
+    this.maxSingleShuffleLimit = maxSingleShuffleLimitConfiged;
+    this.memToMemMergeOutputsThreshold =
+        jobConf.getInt(MRJobConfig.REDUCE_MEMTOMEM_THRESHOLD, ioSortFactor);
     this.mergeThreshold = (long)(this.memoryLimit * 
                           jobConf.getFloat(
                             MRJobConfig.SHUFFLE_MERGE_PERCENT,
@@ -249,17 +257,13 @@ public class MergeManagerImpl<K, V> implements MergeManager<K, V> {
   public void waitForResource() throws InterruptedException {
     inMemoryMerger.waitForMerge();
   }
-  
-  private boolean canShuffleToMemory(long requestedSize) {
-    return (requestedSize < maxSingleShuffleLimit); 
-  }
-  
+
   @Override
   public synchronized MapOutput<K,V> reserve(TaskAttemptID mapId, 
                                              long requestedSize,
                                              int fetcher
                                              ) throws IOException {
-    if (!canShuffleToMemory(requestedSize)) {
+    if (requestedSize > maxSingleShuffleLimit) {
       LOG.info(mapId + ": Shuffling to disk since " + requestedSize + 
                " is greater than maxSingleShuffleLimit (" + 
                maxSingleShuffleLimit + ")");

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/test/java/org/apache/hadoop/mapreduce/task/reduce/TestMergeManager.java

@@ -41,6 +41,7 @@ import org.apache.hadoop.mapred.JobConf;
 import org.apache.hadoop.mapred.MROutputFiles;
 import org.apache.hadoop.mapred.MapOutputFile;
 import org.apache.hadoop.mapreduce.MRJobConfig;
+import org.apache.hadoop.mapreduce.TaskAttemptID;
 import org.apache.hadoop.mapreduce.task.reduce.MergeManagerImpl.CompressAwarePath;
 import org.junit.Assert;
 import org.junit.Test;
@@ -288,5 +289,18 @@ public class TestMergeManager {
     final long maxInMemReduce = mgr.getMaxInMemReduceLimit();
     assertTrue("Large in-memory reduce area unusable: " + maxInMemReduce,
         maxInMemReduce > Integer.MAX_VALUE);
+    assertEquals("maxSingleShuffleLimit to be capped at Integer.MAX_VALUE",
+        Integer.MAX_VALUE, mgr.maxSingleShuffleLimit);
+    verifyReservedMapOutputType(mgr, 10L, "MEMORY");
+    verifyReservedMapOutputType(mgr, 1L + Integer.MAX_VALUE, "DISK");
+  }
+
+  private void verifyReservedMapOutputType(MergeManagerImpl<Text, Text> mgr,
+      long size, String expectedShuffleMode) throws IOException {
+    final TaskAttemptID mapId = TaskAttemptID.forName("attempt_0_1_m_1_1");
+    final MapOutput<Text, Text> mapOutput = mgr.reserve(mapId, size, 1);
+    assertEquals("Shuffled bytes: " + size, expectedShuffleMode,
+        mapOutput.getDescription());
+    mgr.unreserve(size);
   }
 }