HADOOP-10295. Merge change r1563019 from trunk.

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/branches/branch-2@1563020 13f79535-47bb-0310-9956-ffa450edef68

hadoop-common-project/hadoop-common/CHANGES.txt

@@ -11,6 +11,9 @@ Release 2.4.0 - UNRELEASED
     HADOOP-10139. Update and improve the Single Cluster Setup document.
     (Akira Ajisaka via Arpit Agarwal)
 
+    HADOOP-10295. Allow distcp to automatically identify the checksum type of 
+    source files and use it for the target. (jing9 and Laurent Goujon)
+
   OPTIMIZATIONS
 
   BUG FIXES

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/FileChecksum.java

@@ -21,6 +21,7 @@ import java.util.Arrays;
 
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
+import org.apache.hadoop.fs.Options.ChecksumOpt;
 import org.apache.hadoop.io.Writable;
 
 /** An abstract class representing file checksums for files. */
@@ -36,6 +37,10 @@ public abstract class FileChecksum implements Writable {
   /** The value of the checksum in bytes */
   public abstract byte[] getBytes();
 
+  public ChecksumOpt getChecksumOpt() {
+    return null;
+  }
+
   /** Return true if both the algorithms and the values are the same. */
   @Override
   public boolean equals(Object other) {

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/MD5MD5CRC32FileChecksum.java

@@ -88,6 +88,7 @@ public class MD5MD5CRC32FileChecksum extends FileChecksum {
     return DataChecksum.Type.CRC32;
   }
 
+  @Override
   public ChecksumOpt getChecksumOpt() {
     return new ChecksumOpt(getCrcType(), bytesPerCRC);
   }

hadoop-tools/hadoop-distcp/src/main/java/org/apache/hadoop/tools/DistCpOptionSwitch.java

@@ -37,15 +37,16 @@ public enum DistCpOptionSwitch {
   /**
    * Preserves status of file/path in the target.
    * Default behavior with -p, is to preserve replication,
-   * block size, user, group and permission on the target file
+   * block size, user, group, permission and checksum type on the target file.
+   * Note that when preserving checksum type, block size is also preserved.
    *
-   * If any of the optional switches are present among rbugp, then
+   * If any of the optional switches are present among rbugpc, then
-   * only the corresponding file attribute is preserved
+   * only the corresponding file attribute is preserved.
    *
    */
   PRESERVE_STATUS(DistCpConstants.CONF_LABEL_PRESERVE_STATUS,
-      new Option("p", true, "preserve status (rbugp)" +
+      new Option("p", true, "preserve status (rbugpc)" +
-          "(replication, block-size, user, group, permission)")),
+          "(replication, block-size, user, group, permission, checksum-type)")),
 
   /**
    * Update target location by copying only files that are missing

hadoop-tools/hadoop-distcp/src/main/java/org/apache/hadoop/tools/DistCpOptions.java

@@ -61,7 +61,7 @@ public class DistCpOptions {
   private Path targetPath;
 
   public static enum FileAttribute{
-    REPLICATION, BLOCKSIZE, USER, GROUP, PERMISSION;
+    REPLICATION, BLOCKSIZE, USER, GROUP, PERMISSION, CHECKSUMTYPE;
 
     public static FileAttribute getAttribute(char symbol) {
       for (FileAttribute attribute : values()) {

hadoop-tools/hadoop-distcp/src/main/java/org/apache/hadoop/tools/OptionsParser.java

@@ -50,7 +50,7 @@ public class OptionsParser {
     protected String[] flatten(Options options, String[] arguments, boolean stopAtNonOption) {
       for (int index = 0; index < arguments.length; index++) {
         if (arguments[index].equals("-" + DistCpOptionSwitch.PRESERVE_STATUS.getSwitch())) {
-          arguments[index] = "-prbugp";
+          arguments[index] = "-prbugpc";
         }
       }
       return super.flatten(options, arguments, stopAtNonOption);

hadoop-tools/hadoop-distcp/src/main/java/org/apache/hadoop/tools/mapred/CopyMapper.java

@@ -322,7 +322,7 @@ public class CopyMapper extends Mapper<Text, FileStatus, Text, Text> {
                    targetFileStatus.getLen() != source.getLen()
                 || (!skipCrc &&
                        !DistCpUtils.checksumsAreEqual(sourceFS,
-                                          source.getPath(), targetFS, target))
+                          source.getPath(), null, targetFS, target))
                 || (source.getBlockSize() != targetFileStatus.getBlockSize() &&
                       preserve.contains(FileAttribute.BLOCKSIZE))
                );

hadoop-tools/hadoop-distcp/src/main/java/org/apache/hadoop/tools/mapred/RetriableFileCopyCommand.java

@@ -18,23 +18,33 @@
 
 package org.apache.hadoop.tools.mapred;
 
-import org.apache.hadoop.tools.util.RetriableCommand;
+import java.io.BufferedInputStream;
-import org.apache.hadoop.tools.util.ThrottledInputStream;
+import java.io.BufferedOutputStream;
-import org.apache.hadoop.tools.util.DistCpUtils;
+import java.io.IOException;
-import org.apache.hadoop.tools.DistCpOptions.*;
+import java.io.InputStream;
-import org.apache.hadoop.tools.DistCpConstants;
+import java.io.OutputStream;
-import org.apache.hadoop.fs.*;
+import java.util.EnumSet;
-import org.apache.hadoop.mapreduce.Mapper;
+
-import org.apache.hadoop.conf.Configuration;
-import org.apache.hadoop.io.IOUtils;
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.CreateFlag;
+import org.apache.hadoop.fs.FileChecksum;
+import org.apache.hadoop.fs.FileStatus;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Options.ChecksumOpt;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.fs.permission.FsPermission;
+import org.apache.hadoop.io.IOUtils;
+import org.apache.hadoop.mapreduce.Mapper;
+import org.apache.hadoop.tools.DistCpConstants;
+import org.apache.hadoop.tools.DistCpOptions.FileAttribute;
+import org.apache.hadoop.tools.util.DistCpUtils;
+import org.apache.hadoop.tools.util.RetriableCommand;
+import org.apache.hadoop.tools.util.ThrottledInputStream;
 
 import com.google.common.annotations.VisibleForTesting;
 
-import java.io.*;
-import java.util.EnumSet;
-
 /**
  * This class extends RetriableCommand to implement the copy of files,
  * with retries on failure.
@@ -99,15 +109,21 @@ public class RetriableFileCopyCommand extends RetriableCommand {
         LOG.debug("Copying " + sourceFileStatus.getPath() + " to " + target);
         LOG.debug("Tmp-file path: " + tmpTargetPath);
       }
-      FileSystem sourceFS = sourceFileStatus.getPath().getFileSystem(
+      final Path sourcePath = sourceFileStatus.getPath();
-              configuration);
+      final FileSystem sourceFS = sourcePath.getFileSystem(configuration);
-      long bytesRead = copyToTmpFile(tmpTargetPath, targetFS, sourceFileStatus,
+      final FileChecksum sourceChecksum = fileAttributes
-                                     context, fileAttributes);
+          .contains(FileAttribute.CHECKSUMTYPE) ? sourceFS
+          .getFileChecksum(sourcePath) : null;
 
-      compareFileLengths(sourceFileStatus, tmpTargetPath, configuration, bytesRead);
+      long bytesRead = copyToTmpFile(tmpTargetPath, targetFS, sourceFileStatus,
+          context, fileAttributes, sourceChecksum);
+
+      compareFileLengths(sourceFileStatus, tmpTargetPath, configuration,
+          bytesRead);
       //At this point, src&dest lengths are same. if length==0, we skip checksum
       if ((bytesRead != 0) && (!skipCrc)) {
-        compareCheckSums(sourceFS, sourceFileStatus.getPath(), targetFS, tmpTargetPath);
+        compareCheckSums(sourceFS, sourceFileStatus.getPath(), sourceChecksum,
+            targetFS, tmpTargetPath);
       }
       promoteTmpToTarget(tmpTargetPath, target, targetFS);
       return bytesRead;
@@ -118,14 +134,33 @@ public class RetriableFileCopyCommand extends RetriableCommand {
     }
   }
 
+  /**
+   * @return the checksum spec of the source checksum if checksum type should be
+   *         preserved
+   */
+  private ChecksumOpt getChecksumOpt(EnumSet<FileAttribute> fileAttributes,
+      FileChecksum sourceChecksum) {
+    if (fileAttributes.contains(FileAttribute.CHECKSUMTYPE)
+        && sourceChecksum != null) {
+      return sourceChecksum.getChecksumOpt();
+    }
+    return null;
+  }
+
   private long copyToTmpFile(Path tmpTargetPath, FileSystem targetFS,
       FileStatus sourceFileStatus, Mapper.Context context,
-                             EnumSet<FileAttribute> fileAttributes)
+      EnumSet<FileAttribute> fileAttributes, final FileChecksum sourceChecksum)
       throws IOException {
-    OutputStream outStream = new BufferedOutputStream(targetFS.create(
+    FsPermission permission = FsPermission.getFileDefault().applyUMask(
-            tmpTargetPath, true, BUFFER_SIZE,
+        FsPermission.getUMask(targetFS.getConf()));
-            getReplicationFactor(fileAttributes, sourceFileStatus, targetFS, tmpTargetPath),
+    OutputStream outStream = new BufferedOutputStream(
-            getBlockSize(fileAttributes, sourceFileStatus, targetFS, tmpTargetPath), context));
+        targetFS.create(tmpTargetPath, permission,
+            EnumSet.of(CreateFlag.CREATE, CreateFlag.OVERWRITE), BUFFER_SIZE,
+            getReplicationFactor(fileAttributes, sourceFileStatus, targetFS,
+                tmpTargetPath),
+            getBlockSize(fileAttributes, sourceFileStatus, targetFS,
+                tmpTargetPath),
+            context, getChecksumOpt(fileAttributes, sourceChecksum)));
     return copyBytes(sourceFileStatus, outStream, BUFFER_SIZE, context);
   }
 
@@ -140,9 +175,10 @@ public class RetriableFileCopyCommand extends RetriableCommand {
   }
 
   private void compareCheckSums(FileSystem sourceFS, Path source,
-                                FileSystem targetFS, Path target)
+      FileChecksum sourceChecksum, FileSystem targetFS, Path target)
       throws IOException {
-    if (!DistCpUtils.checksumsAreEqual(sourceFS, source, targetFS, target)) {
+    if (!DistCpUtils.checksumsAreEqual(sourceFS, source, sourceChecksum,
+        targetFS, target)) {
       StringBuilder errorMessage = new StringBuilder("Check-sum mismatch between ")
           .append(source).append(" and ").append(target).append(".");
       if (sourceFS.getFileStatus(source).getBlockSize() != targetFS.getFileStatus(target).getBlockSize()) {
@@ -249,11 +285,18 @@ public class RetriableFileCopyCommand extends RetriableCommand {
             sourceFile.getReplication() : targetFS.getDefaultReplication(tmpTargetPath);
   }
 
+  /**
+   * @return the block size of the source file if we need to preserve either
+   *         the block size or the checksum type. Otherwise the default block
+   *         size of the target FS.
+   */
   private static long getBlockSize(
           EnumSet<FileAttribute> fileAttributes,
           FileStatus sourceFile, FileSystem targetFS, Path tmpTargetPath) {
-    return fileAttributes.contains(FileAttribute.BLOCKSIZE)?
+    boolean preserve = fileAttributes.contains(FileAttribute.BLOCKSIZE)
-            sourceFile.getBlockSize() : targetFS.getDefaultBlockSize(tmpTargetPath);
+        || fileAttributes.contains(FileAttribute.CHECKSUMTYPE);
+    return preserve ? sourceFile.getBlockSize() : targetFS
+        .getDefaultBlockSize(tmpTargetPath);
   }
 
   /**

hadoop-tools/hadoop-distcp/src/main/java/org/apache/hadoop/tools/util/DistCpUtils.java

@@ -280,6 +280,8 @@ public class DistCpUtils {
    *
    * @param sourceFS FileSystem for the source path.
    * @param source The source path.
+   * @param sourceChecksum The checksum of the source file. If it is null we
+   * still need to retrieve it through sourceFS.
    * @param targetFS FileSystem for the target path.
    * @param target The target path.
    * @return If either checksum couldn't be retrieved, the function returns
@@ -288,12 +290,12 @@ public class DistCpUtils {
    * @throws IOException if there's an exception while retrieving checksums.
    */
   public static boolean checksumsAreEqual(FileSystem sourceFS, Path source,
-                                   FileSystem targetFS, Path target)
+      FileChecksum sourceChecksum, FileSystem targetFS, Path target)
       throws IOException {
-    FileChecksum sourceChecksum = null;
     FileChecksum targetChecksum = null;
     try {
-      sourceChecksum = sourceFS.getFileChecksum(source);
+      sourceChecksum = sourceChecksum != null ? sourceChecksum : sourceFS
+          .getFileChecksum(source);
       targetChecksum = targetFS.getFileChecksum(target);
     } catch (IOException e) {
       LOG.error("Unable to retrieve checksum for " + source + " or " + target, e);

hadoop-tools/hadoop-distcp/src/test/java/org/apache/hadoop/tools/TestOptionsParser.java

@@ -397,6 +397,7 @@ public class TestOptionsParser {
     Assert.assertFalse(options.shouldPreserve(FileAttribute.PERMISSION));
     Assert.assertFalse(options.shouldPreserve(FileAttribute.USER));
     Assert.assertFalse(options.shouldPreserve(FileAttribute.GROUP));
+    Assert.assertFalse(options.shouldPreserve(FileAttribute.CHECKSUMTYPE));
 
     options = OptionsParser.parse(new String[] {
         "-p",
@@ -408,6 +409,7 @@ public class TestOptionsParser {
     Assert.assertTrue(options.shouldPreserve(FileAttribute.PERMISSION));
     Assert.assertTrue(options.shouldPreserve(FileAttribute.USER));
     Assert.assertTrue(options.shouldPreserve(FileAttribute.GROUP));
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.CHECKSUMTYPE));
 
     options = OptionsParser.parse(new String[] {
         "-p",
@@ -418,6 +420,7 @@ public class TestOptionsParser {
     Assert.assertTrue(options.shouldPreserve(FileAttribute.PERMISSION));
     Assert.assertTrue(options.shouldPreserve(FileAttribute.USER));
     Assert.assertTrue(options.shouldPreserve(FileAttribute.GROUP));
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.CHECKSUMTYPE));
 
     options = OptionsParser.parse(new String[] {
         "-pbr",
@@ -429,6 +432,7 @@ public class TestOptionsParser {
     Assert.assertFalse(options.shouldPreserve(FileAttribute.PERMISSION));
     Assert.assertFalse(options.shouldPreserve(FileAttribute.USER));
     Assert.assertFalse(options.shouldPreserve(FileAttribute.GROUP));
+    Assert.assertFalse(options.shouldPreserve(FileAttribute.CHECKSUMTYPE));
 
     options = OptionsParser.parse(new String[] {
         "-pbrgup",
@@ -440,6 +444,31 @@ public class TestOptionsParser {
     Assert.assertTrue(options.shouldPreserve(FileAttribute.PERMISSION));
     Assert.assertTrue(options.shouldPreserve(FileAttribute.USER));
     Assert.assertTrue(options.shouldPreserve(FileAttribute.GROUP));
+    Assert.assertFalse(options.shouldPreserve(FileAttribute.CHECKSUMTYPE));
+
+    options = OptionsParser.parse(new String[] {
+        "-pbrgupc",
+        "-f",
+        "hdfs://localhost:8020/source/first",
+        "hdfs://localhost:8020/target/"});
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.BLOCKSIZE));
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.REPLICATION));
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.PERMISSION));
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.USER));
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.GROUP));
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.CHECKSUMTYPE));
+
+    options = OptionsParser.parse(new String[] {
+        "-pc",
+        "-f",
+        "hdfs://localhost:8020/source/first",
+        "hdfs://localhost:8020/target/"});
+    Assert.assertFalse(options.shouldPreserve(FileAttribute.BLOCKSIZE));
+    Assert.assertFalse(options.shouldPreserve(FileAttribute.REPLICATION));
+    Assert.assertFalse(options.shouldPreserve(FileAttribute.PERMISSION));
+    Assert.assertFalse(options.shouldPreserve(FileAttribute.USER));
+    Assert.assertFalse(options.shouldPreserve(FileAttribute.GROUP));
+    Assert.assertTrue(options.shouldPreserve(FileAttribute.CHECKSUMTYPE));
 
     options = OptionsParser.parse(new String[] {
         "-p",
@@ -452,7 +481,7 @@ public class TestOptionsParser {
       attribIterator.next();
       i++;
     }
-    Assert.assertEquals(i, 5);
+    Assert.assertEquals(i, 6);
 
     try {
       OptionsParser.parse(new String[] {

hadoop-tools/hadoop-distcp/src/test/java/org/apache/hadoop/tools/mapred/TestCopyMapper.java

@@ -18,29 +18,6 @@
 
 package org.apache.hadoop.tools.mapred;
 
-import org.apache.commons.logging.Log;
-import org.apache.commons.logging.LogFactory;
-import org.apache.hadoop.conf.Configuration;
-import org.apache.hadoop.fs.FileStatus;
-import org.apache.hadoop.fs.FileSystem;
-import org.apache.hadoop.fs.Path;
-import org.apache.hadoop.fs.permission.FsAction;
-import org.apache.hadoop.fs.permission.FsPermission;
-import org.apache.hadoop.hdfs.MiniDFSCluster;
-import org.apache.hadoop.io.IOUtils;
-import org.apache.hadoop.io.Text;
-import org.apache.hadoop.mapreduce.*;
-import org.apache.hadoop.security.AccessControlException;
-import org.apache.hadoop.security.UserGroupInformation;
-import org.apache.hadoop.tools.DistCpConstants;
-import org.apache.hadoop.tools.DistCpOptionSwitch;
-import org.apache.hadoop.tools.DistCpOptions;
-import org.apache.hadoop.tools.StubContext;
-import org.apache.hadoop.tools.util.DistCpUtils;
-import org.junit.Assert;
-import org.junit.BeforeClass;
-import org.junit.Test;
-
 import java.io.DataOutputStream;
 import java.io.IOException;
 import java.io.OutputStream;
@@ -49,11 +26,38 @@ import java.util.ArrayList;
 import java.util.EnumSet;
 import java.util.List;
 
+import org.apache.commons.logging.Log;
+import org.apache.commons.logging.LogFactory;
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.CreateFlag;
+import org.apache.hadoop.fs.FileStatus;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Options.ChecksumOpt;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.fs.permission.FsAction;
+import org.apache.hadoop.fs.permission.FsPermission;
+import org.apache.hadoop.hdfs.MiniDFSCluster;
+import org.apache.hadoop.io.IOUtils;
+import org.apache.hadoop.io.Text;
+import org.apache.hadoop.mapreduce.Mapper;
+import org.apache.hadoop.security.AccessControlException;
+import org.apache.hadoop.security.UserGroupInformation;
+import org.apache.hadoop.tools.DistCpConstants;
+import org.apache.hadoop.tools.DistCpOptionSwitch;
+import org.apache.hadoop.tools.DistCpOptions;
+import org.apache.hadoop.tools.StubContext;
+import org.apache.hadoop.tools.util.DistCpUtils;
+import org.apache.hadoop.util.DataChecksum;
+import org.junit.Assert;
+import org.junit.BeforeClass;
+import org.junit.Test;
+
 public class TestCopyMapper {
   private static final Log LOG = LogFactory.getLog(TestCopyMapper.class);
   private static List<Path> pathList = new ArrayList<Path>();
   private static int nFiles = 0;
   private static final int DEFAULT_FILE_SIZE = 1024;
+  private static final long NON_DEFAULT_BLOCK_SIZE = 4096;
 
   private static MiniDFSCluster cluster;
 
@@ -119,12 +123,27 @@ public class TestCopyMapper {
     mkdirs(SOURCE_PATH + "/2/3/4");
     mkdirs(SOURCE_PATH + "/2/3");
     mkdirs(SOURCE_PATH + "/5");
-    touchFile(SOURCE_PATH + "/5/6", true);
+    touchFile(SOURCE_PATH + "/5/6", true, null);
     mkdirs(SOURCE_PATH + "/7");
     mkdirs(SOURCE_PATH + "/7/8");
     touchFile(SOURCE_PATH + "/7/8/9");
   }
 
+  private static void createSourceDataWithDifferentChecksumType()
+      throws Exception {
+    mkdirs(SOURCE_PATH + "/1");
+    mkdirs(SOURCE_PATH + "/2");
+    mkdirs(SOURCE_PATH + "/2/3/4");
+    mkdirs(SOURCE_PATH + "/2/3");
+    mkdirs(SOURCE_PATH + "/5");
+    touchFile(SOURCE_PATH + "/5/6", new ChecksumOpt(DataChecksum.Type.CRC32,
+        512));
+    mkdirs(SOURCE_PATH + "/7");
+    mkdirs(SOURCE_PATH + "/7/8");
+    touchFile(SOURCE_PATH + "/7/8/9", new ChecksumOpt(DataChecksum.Type.CRC32C,
+        512));
+  }
+
   private static void mkdirs(String path) throws Exception {
     FileSystem fileSystem = cluster.getFileSystem();
     final Path qualifiedPath = new Path(path).makeQualified(fileSystem.getUri(),
@@ -134,21 +153,31 @@ public class TestCopyMapper {
   }
 
   private static void touchFile(String path) throws Exception {
-    touchFile(path, false);
+    touchFile(path, false, null);
   }
 
-  private static void touchFile(String path, boolean createMultipleBlocks) throws Exception {
+  private static void touchFile(String path, ChecksumOpt checksumOpt)
-    final long NON_DEFAULT_BLOCK_SIZE = 4096;
+      throws Exception {
+    // create files with specific checksum opt and non-default block size
+    touchFile(path, true, checksumOpt);
+  }
+
+  private static void touchFile(String path, boolean createMultipleBlocks,
+      ChecksumOpt checksumOpt) throws Exception {
     FileSystem fs;
     DataOutputStream outputStream = null;
     try {
       fs = cluster.getFileSystem();
       final Path qualifiedPath = new Path(path).makeQualified(fs.getUri(),
           fs.getWorkingDirectory());
-      final long blockSize = createMultipleBlocks? NON_DEFAULT_BLOCK_SIZE : fs.getDefaultBlockSize(qualifiedPath) * 2;
+      final long blockSize = createMultipleBlocks ? NON_DEFAULT_BLOCK_SIZE : fs
-      outputStream = fs.create(qualifiedPath, true, 0,
+          .getDefaultBlockSize(qualifiedPath) * 2;
-              (short)(fs.getDefaultReplication(qualifiedPath)*2),
+      FsPermission permission = FsPermission.getFileDefault().applyUMask(
-              blockSize);
+          FsPermission.getUMask(fs.getConf()));
+      outputStream = fs.create(qualifiedPath, permission,
+          EnumSet.of(CreateFlag.CREATE, CreateFlag.OVERWRITE), 0,
+          (short) (fs.getDefaultReplication(qualifiedPath) * 2), blockSize,
+          null, checksumOpt);
       byte[] bytes = new byte[DEFAULT_FILE_SIZE];
       outputStream.write(bytes);
       long fileSize = DEFAULT_FILE_SIZE;
@@ -171,17 +200,40 @@ public class TestCopyMapper {
     }
   }
 
+  @Test
+  public void testCopyWithDifferentChecksumType() throws Exception {
+    testCopy(true);
+  }
+
   @Test(timeout=40000)
   public void testRun() {
+    testCopy(false);
+  }
+
+  private void testCopy(boolean preserveChecksum) {
     try {
       deleteState();
+      if (preserveChecksum) {
+        createSourceDataWithDifferentChecksumType();
+      } else {
         createSourceData();
+      }
 
       FileSystem fs = cluster.getFileSystem();
       CopyMapper copyMapper = new CopyMapper();
       StubContext stubContext = new StubContext(getConfiguration(), null, 0);
       Mapper<Text, FileStatus, Text, Text>.Context context
               = stubContext.getContext();
+
+      Configuration configuration = context.getConfiguration();
+      EnumSet<DistCpOptions.FileAttribute> fileAttributes
+              = EnumSet.of(DistCpOptions.FileAttribute.REPLICATION);
+      if (preserveChecksum) {
+        fileAttributes.add(DistCpOptions.FileAttribute.CHECKSUMTYPE);
+      }
+      configuration.set(DistCpOptionSwitch.PRESERVE_STATUS.getConfigLabel(),
+              DistCpUtils.packAttributes(fileAttributes));
+
       copyMapper.setup(context);
 
       for (Path path: pathList) {
@@ -195,19 +247,29 @@ public class TestCopyMapper {
                 .replaceAll(SOURCE_PATH, TARGET_PATH));
         Assert.assertTrue(fs.exists(targetPath));
         Assert.assertTrue(fs.isFile(targetPath) == fs.isFile(path));
-        Assert.assertEquals(fs.getFileStatus(path).getReplication(),
+        FileStatus sourceStatus = fs.getFileStatus(path);
-                fs.getFileStatus(targetPath).getReplication());
+        FileStatus targetStatus = fs.getFileStatus(targetPath);
-        Assert.assertEquals(fs.getFileStatus(path).getBlockSize(),
+        Assert.assertEquals(sourceStatus.getReplication(),
-                fs.getFileStatus(targetPath).getBlockSize());
+            targetStatus.getReplication());
-        Assert.assertTrue(!fs.isFile(targetPath) ||
+        if (preserveChecksum) {
-                fs.getFileChecksum(targetPath).equals(
+          Assert.assertEquals(sourceStatus.getBlockSize(),
-                        fs.getFileChecksum(path)));
+              targetStatus.getBlockSize());
+        }
+        Assert.assertTrue(!fs.isFile(targetPath)
+            || fs.getFileChecksum(targetPath).equals(fs.getFileChecksum(path)));
       }
 
       Assert.assertEquals(pathList.size(),
               stubContext.getReporter().getCounter(CopyMapper.Counter.COPY).getValue());
-      Assert.assertEquals(nFiles * DEFAULT_FILE_SIZE,
+      if (!preserveChecksum) {
-              stubContext.getReporter().getCounter(CopyMapper.Counter.BYTESCOPIED).getValue());
+        Assert.assertEquals(nFiles * DEFAULT_FILE_SIZE, stubContext
+            .getReporter().getCounter(CopyMapper.Counter.BYTESCOPIED)
+            .getValue());
+      } else {
+        Assert.assertEquals(nFiles * NON_DEFAULT_BLOCK_SIZE * 2, stubContext
+            .getReporter().getCounter(CopyMapper.Counter.BYTESCOPIED)
+            .getValue());
+      }
 
       testCopyingExistingFiles(fs, copyMapper, context);
       for (Text value : stubContext.getWriter().values()) {