HDFS-10691. FileDistribution fails in hdfs oiv command due to ArrayIndexOutOfBoundsException. Contributed by Yiqun Lin.

(cherry picked from commit 204a2055b1)
(cherry picked from commit 0cff416c35)

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/offlineImageViewer/FileDistributionCalculator.java

@@ -128,6 +128,12 @@ final class FileDistributionCalculator {
 
         int bucket = fileSize > maxSize ? distribution.length - 1 : (int) Math
             .ceil((double)fileSize / steps);
+        // Compare the bucket value with distribution's length again,
+        // because sometimes the bucket value will be equal to
+        // the length when maxSize can't be divided completely by step.
+        if (bucket >= distribution.length) {
+          bucket = distribution.length - 1;
+        }
         ++distribution[bucket];
 
       } else if (p.getType() == INodeSection.INode.Type.DIRECTORY) {

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/offlineImageViewer/FileDistributionVisitor.java

@@ -145,6 +145,10 @@ class FileDistributionVisitor extends TextWriterImageVisitor {
       high = distribution.length-1;
     else
       high = (int)Math.ceil((double)current.fileSize / step);
+
+    if (high >= distribution.length) {
+      high = distribution.length - 1;
+    }
     distribution[high]++;
     if(totalFiles % 1000000 == 1)
       System.out.println("Files processed: " + totalFiles

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/tools/offlineImageViewer/TestOfflineImageViewer.java

@@ -70,6 +70,7 @@ import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.FileSystemTestHelper;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.hdfs.DFSConfigKeys;
+import org.apache.hadoop.hdfs.DFSTestUtil;
 import org.apache.hadoop.hdfs.DistributedFileSystem;
 import org.apache.hadoop.hdfs.MiniDFSCluster;
 import org.apache.hadoop.hdfs.protocol.HdfsConstants.SafeModeAction;
@@ -552,4 +553,52 @@ public class TestOfflineImageViewer {
       GenericTestUtils.assertExceptionContains("Layout version mismatch.", t);
     }
   }
+
+  @Test
+  public void testFileDistributionCalculatorForException() throws Exception {
+    File fsimageFile = null;
+    Configuration conf = new Configuration();
+    HashMap<String, FileStatus> files = Maps.newHashMap();
+
+    // Create a initial fsimage file
+    try (MiniDFSCluster cluster =
+        new MiniDFSCluster.Builder(conf).numDataNodes(1).build()) {
+      cluster.waitActive();
+      DistributedFileSystem hdfs = cluster.getFileSystem();
+
+      // Create a reasonable namespace
+      Path dir = new Path("/dir");
+      hdfs.mkdirs(dir);
+      files.put(dir.toString(), pathToFileEntry(hdfs, dir.toString()));
+      // Create files with byte size that can't be divided by step size,
+      // the byte size for here are 3, 9, 15, 21.
+      for (int i = 0; i < FILES_PER_DIR; i++) {
+        Path file = new Path(dir, "file" + i);
+        DFSTestUtil.createFile(hdfs, file, 6 * i + 3, (short) 1, 0);
+
+        files.put(file.toString(),
+            pathToFileEntry(hdfs, file.toString()));
+      }
+
+      // Write results to the fsimage file
+      hdfs.setSafeMode(SafeModeAction.SAFEMODE_ENTER, false);
+      hdfs.saveNamespace();
+      // Determine location of fsimage file
+      fsimageFile =
+          FSImageTestUtil.findLatestImageFile(FSImageTestUtil
+              .getFSImage(cluster.getNameNode()).getStorage().getStorageDir(0));
+      if (fsimageFile == null) {
+        throw new RuntimeException("Didn't generate or can't find fsimage");
+      }
+    }
+
+    // Run the test with params -maxSize 23 and -step 4, it will not throw
+    // ArrayIndexOutOfBoundsException with index 6 when deals with
+    // 21 byte size file.
+    int status =
+        OfflineImageViewerPB.run(new String[] {"-i",
+            fsimageFile.getAbsolutePath(), "-o", "-", "-p",
+            "FileDistribution", "-maxSize", "23", "-step", "4"});
+    assertEquals(0, status);
+  }
 }