Merge branch 'trunk' into HDFS-7240

hadoop-common-project/hadoop-auth/src/main/java/org/apache/hadoop/security/authentication/server/AuthenticationFilter.java

@@ -40,85 +40,26 @@ import java.text.SimpleDateFormat;
 import java.util.*;
 
 /**
- * <p>The {@link AuthenticationFilter} enables protecting web application
+ * The {@link AuthenticationFilter} enables protecting web application
  * resources with different (pluggable)
  * authentication mechanisms and signer secret providers.
- * </p>
  * <p>
- * Out of the box it provides 2 authentication mechanisms: Pseudo and Kerberos SPNEGO.
- * </p>
  * Additional authentication mechanisms are supported via the {@link AuthenticationHandler} interface.
  * <p>
  * This filter delegates to the configured authentication handler for authentication and once it obtains an
  * {@link AuthenticationToken} from it, sets a signed HTTP cookie with the token. For client requests
  * that provide the signed HTTP cookie, it verifies the validity of the cookie, extracts the user information
  * and lets the request proceed to the target resource.
- * </p>
- * The supported configuration properties are:
- * <ul>
- * <li>config.prefix: indicates the prefix to be used by all other configuration properties, the default value
- * is no prefix. See below for details on how/why this prefix is used.</li>
- * <li>[#PREFIX#.]type: simple|kerberos|#CLASS#, 'simple' is short for the
- * {@link PseudoAuthenticationHandler}, 'kerberos' is short for {@link KerberosAuthenticationHandler}, otherwise
- * the full class name of the {@link AuthenticationHandler} must be specified.</li>
- * <li>[#PREFIX#.]signature.secret.file: when signer.secret.provider is set to
- * "file" or not specified, this is the location of file including the secret
- *  used to sign the HTTP cookie.</li>
- * <li>[#PREFIX#.]token.validity: time -in seconds- that the generated token is
- * valid before a new authentication is triggered, default value is
- * <code>3600</code> seconds. This is also used for the rollover interval for
- * the "random" and "zookeeper" SignerSecretProviders.</li>
- * <li>[#PREFIX#.]cookie.domain: domain to use for the HTTP cookie that stores the authentication token.</li>
- * <li>[#PREFIX#.]cookie.path: path to use for the HTTP cookie that stores the authentication token.</li>
- * </ul>
  * <p>
  * The rest of the configuration properties are specific to the {@link AuthenticationHandler} implementation and the
  * {@link AuthenticationFilter} will take all the properties that start with the prefix #PREFIX#, it will remove
  * the prefix from it and it will pass them to the the authentication handler for initialization. Properties that do
  * not start with the prefix will not be passed to the authentication handler initialization.
- * </p>
  * <p>
- * Out of the box it provides 3 signer secret provider implementations:
- * "file", "random" and "zookeeper"
- * </p>
- * Additional signer secret providers are supported via the
- * {@link SignerSecretProvider} class.
- * <p>
- * For the HTTP cookies mentioned above, the SignerSecretProvider is used to
- * determine the secret to use for signing the cookies. Different
- * implementations can have different behaviors. The "file" implementation
- * loads the secret from a specified file. The "random" implementation uses a
- * randomly generated secret that rolls over at the interval specified by the
- * [#PREFIX#.]token.validity mentioned above.  The "zookeeper" implementation
- * is like the "random" one, except that it synchronizes the random secret
- * and rollovers between multiple servers; it's meant for HA services.
- * </p>
- * The relevant configuration properties are:
- * <ul>
- * <li>signer.secret.provider: indicates the name of the SignerSecretProvider
- * class to use. Possible values are: "file", "random", "zookeeper", or a
- * classname. If not specified, the "file" implementation will be used with
- * [#PREFIX#.]signature.secret.file; and if that's not specified, the "random"
- * implementation will be used.</li>
- * <li>[#PREFIX#.]signature.secret.file: When the "file" implementation is
- * specified, this content of this file is used as the secret.</li>
- * <li>[#PREFIX#.]token.validity: When the "random" or "zookeeper"
- * implementations are specified, this value is used as the rollover
- * interval.</li>
- * </ul>
+ * Details of the configurations are listed on <a href="../../../../../../../Configuration.html">Configuration Page</a>
  * <p>
  * The "zookeeper" implementation has additional configuration properties that
  * must be specified; see {@link ZKSignerSecretProvider} for details.
- * </p>
- * For subclasses of AuthenticationFilter that want additional control over the
- * SignerSecretProvider, they can use the following attribute set in the
- * ServletContext:
- * <ul>
- * <li>signer.secret.provider.object: A SignerSecretProvider implementation can
- * be passed here that will be used instead of the signer.secret.provider
- * configuration property. Note that the class should already be
- * initialized.</li>
- * </ul>
  */
 
 @InterfaceAudience.Private

hadoop-common-project/hadoop-auth/src/main/java/org/apache/hadoop/security/authentication/util/ZKSignerSecretProvider.java

@@ -57,34 +57,7 @@ import org.slf4j.LoggerFactory;
  * {@link org.apache.hadoop.security.authentication.server.AuthenticationFilter}
  * for more details.
  * <p>
- * The supported configuration properties are:
- * <ul>
- * <li>signer.secret.provider.zookeeper.connection.string: indicates the
- * ZooKeeper connection string to connect with.</li>
- * <li>signer.secret.provider.zookeeper.path: indicates the ZooKeeper path
- * to use for storing and retrieving the secrets.  All ZKSignerSecretProviders
- * that need to coordinate should point to the same path.</li>
- * <li>signer.secret.provider.zookeeper.auth.type: indicates the auth type to
- * use.  Supported values are "none" and "sasl".  The default value is "none"
- * </li>
- * <li>signer.secret.provider.zookeeper.kerberos.keytab: set this to the path
- * with the Kerberos keytab file.  This is only required if using Kerberos.</li>
- * <li>signer.secret.provider.zookeeper.kerberos.principal: set this to the
- * Kerberos principal to use.  This only required if using Kerberos.</li>
- * <li>signer.secret.provider.zookeeper.disconnect.on.close: when set to "true",
- * ZKSignerSecretProvider will close the ZooKeeper connection on shutdown.  The
- * default is "true". Only set this to "false" if a custom Curator client is
- * being provided and the disconnection is being handled elsewhere.</li>
- * </ul>
- *
- * The following attribute in the ServletContext can also be set if desired:
- * <ul>
- * <li>signer.secret.provider.zookeeper.curator.client: A CuratorFramework
- * client object can be passed here. If given, the "zookeeper" implementation
- * will use this Curator client instead of creating its own, which is useful if
- * you already have a Curator client or want more control over its
- * configuration.</li>
- * </ul>
+ * Details of the configurations are listed on <a href="../../../../../../../Configuration.html">Configuration Page</a>
  */
 @InterfaceStability.Unstable
 @InterfaceAudience.Private

hadoop-common-project/hadoop-auth/src/site/markdown/Configuration.md

@@ -34,12 +34,11 @@ Hadoop Auth uses SLF4J-API for logging. Auth Maven POM dependencies define the S
 *   `[PREFIX.]type`: the authentication type keyword (`simple` or \
     `kerberos`) or a Authentication handler implementation.
 
-*   `[PREFIX.]signature.secret`: When `signer.secret.provider` is set to
-    `string` or not specified, this is the value for the secret used to sign
-    the HTTP cookie.
+*   `[PREFIX.]signature.secret.file`: When `signer.secret.provider` is set to
+    `file`, this is the location of file including the secret used to sign the HTTP cookie.
 
 *   `[PREFIX.]token.validity`: The validity -in seconds- of the generated
-    authentication token. The default value is `3600` seconds. This is also
+    authentication token. The default value is `36000` seconds. This is also
     used for the rollover interval when `signer.secret.provider` is set to
     `random` or `zookeeper`.
 
@@ -50,10 +49,11 @@ Hadoop Auth uses SLF4J-API for logging. Auth Maven POM dependencies define the S
     authentication token.
 
 *   `signer.secret.provider`: indicates the name of the SignerSecretProvider
-    class to use. Possible values are: `string`, `random`,
-    `zookeeper`, or a classname. If not specified, the `string`
+    class to use. Possible values are: `file`, `random`,
+    `zookeeper`, or a classname. If not specified, the `file`
     implementation will be used; and failing that, the `random`
-    implementation will be used.
+    implementation will be used. If "file" is to be used, one need to specify
+    `signature.secret.file` and point to the secret file.
 
 ### Kerberos Configuration
 
@@ -232,24 +232,25 @@ The SignerSecretProvider is used to provide more advanced behaviors for the secr
 These are the relevant configuration properties:
 
 *   `signer.secret.provider`: indicates the name of the
-    SignerSecretProvider class to use. Possible values are: "string",
-    "random", "zookeeper", or a classname. If not specified, the "string"
+    SignerSecretProvider class to use. Possible values are: "file",
+    "random", "zookeeper", or a classname. If not specified, the "file"
     implementation will be used; and failing that, the "random" implementation
-    will be used.
+    will be used. If "file" is to be used, one need to specify `signature.secret.file`
+    and point to the secret file.
 
-*   `[PREFIX.]signature.secret`: When `signer.secret.provider` is set
-    to `string` or not specified, this is the value for the secret used to
+*   `[PREFIX.]signature.secret.file`: When `signer.secret.provider` is set
+    to `file` or not specified, this is the value for the secret used to
     sign the HTTP cookie.
 
 *   `[PREFIX.]token.validity`: The validity -in seconds- of the generated
-    authentication token. The default value is `3600` seconds. This is
+    authentication token. The default value is `36000` seconds. This is
     also used for the rollover interval when `signer.secret.provider` is
     set to `random` or `zookeeper`.
 
 The following configuration properties are specific to the `zookeeper` implementation:
 
 *   `signer.secret.provider.zookeeper.connection.string`: Indicates the
-    ZooKeeper connection string to connect with.
+    ZooKeeper connection string to connect with. The default value is `localhost:2181`
 
 *   `signer.secret.provider.zookeeper.path`: Indicates the ZooKeeper path
     to use for storing and retrieving the secrets. All servers
@@ -266,6 +267,17 @@ The following configuration properties are specific to the `zookeeper` implement
 *   `signer.secret.provider.zookeeper.kerberos.principal`: Set this to the
     Kerberos principal to use. This only required if using Kerberos.
 
+*   `signer.secret.provider.zookeeper.disconnect.on.shutdown`: Whether to close the
+    ZooKeeper connection when the provider is shutdown. The default value is `true`.
+    Only set this to `false` if a custom Curator client is being provided and
+    the disconnection is being handled elsewhere.
+
+The following attribute in the ServletContext can also be set if desired:
+*   `signer.secret.provider.zookeeper.curator.client`: A CuratorFramework client
+    object can be passed here. If given, the "zookeeper" implementation will use
+    this Curator client instead of creating its own, which is useful if you already
+    have a Curator client or want more control over its configuration.
+
 **Example**:
 
 ```xml
@@ -276,11 +288,11 @@ The following configuration properties are specific to the `zookeeper` implement
             <!-- AuthenticationHandler configs not shown -->
             <init-param>
                 <param-name>signer.secret.provider</param-name>
-                <param-value>string</param-value>
+                <param-value>file</param-value>
             </init-param>
             <init-param>
-                <param-name>signature.secret</param-name>
-                <param-value>my_secret</param-value>
+                <param-name>signature.secret.file</param-name>
+                <param-value>/myapp/secret_file</param-value>
             </init-param>
         </filter>
 
@@ -334,10 +346,6 @@ The following configuration properties are specific to the `zookeeper` implement
                 <param-name>signer.secret.provider.zookeeper.path</param-name>
                 <param-value>/myapp/secrets</param-value>
             </init-param>
-            <init-param>
-                <param-name>signer.secret.provider.zookeeper.use.kerberos.acls</param-name>
-                <param-value>true</param-value>
-            </init-param>
             <init-param>
                 <param-name>signer.secret.provider.zookeeper.kerberos.keytab</param-name>
                 <param-value>/tmp/auth.keytab</param-value>

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/crypto/CryptoInputStream.java

@@ -17,6 +17,7 @@
  */
 package org.apache.hadoop.crypto;
 
+import java.io.EOFException;
 import java.io.FileDescriptor;
 import java.io.FileInputStream;
 import java.io.FilterInputStream;
@@ -34,6 +35,7 @@ import org.apache.hadoop.classification.InterfaceStability;
 import org.apache.hadoop.fs.ByteBufferReadable;
 import org.apache.hadoop.fs.CanSetDropBehind;
 import org.apache.hadoop.fs.CanSetReadahead;
+import org.apache.hadoop.fs.FSExceptionMessages;
 import org.apache.hadoop.fs.HasEnhancedByteBufferAccess;
 import org.apache.hadoop.fs.HasFileDescriptor;
 import org.apache.hadoop.fs.PositionedReadable;
@@ -395,7 +397,9 @@ public class CryptoInputStream extends FilterInputStream implements
   /** Seek to a position. */
   @Override
   public void seek(long pos) throws IOException {
-    Preconditions.checkArgument(pos >= 0, "Cannot seek to negative offset.");
+    if (pos < 0) {
+      throw new EOFException(FSExceptionMessages.NEGATIVE_SEEK);
+    }
     checkStream();
     try {
       /*

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

@@ -0,0 +1,168 @@
+/**
+ * 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
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * 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.hadoop.fs;
+
+import org.apache.hadoop.classification.InterfaceAudience;
+import org.apache.hadoop.classification.InterfaceStability;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.io.Closeable;
+import java.io.File;
+import java.io.IOException;
+import java.util.concurrent.atomic.AtomicBoolean;
+import java.util.concurrent.atomic.AtomicLong;
+
+/**
+ * Interface for class that can tell estimate much space
+ * is used in a directory.
+ * <p>
+ * The implementor is fee to cache space used. As such there
+ * are methods to update the cached value with any known changes.
+ */
+@InterfaceAudience.LimitedPrivate({"HDFS", "MapReduce"})
+@InterfaceStability.Evolving
+public abstract class CachingGetSpaceUsed implements Closeable, GetSpaceUsed {
+  static final Logger LOG = LoggerFactory.getLogger(CachingGetSpaceUsed.class);
+
+  protected final AtomicLong used = new AtomicLong();
+  private final AtomicBoolean running = new AtomicBoolean(true);
+  private final long refreshInterval;
+  private final String dirPath;
+  private Thread refreshUsed;
+
+  /**
+   * This is the constructor used by the builder.
+   * All overriding classes should implement this.
+   */
+  public CachingGetSpaceUsed(CachingGetSpaceUsed.Builder builder)
+      throws IOException {
+    this(builder.getPath(), builder.getInterval(), builder.getInitialUsed());
+  }
+
+  /**
+   * Keeps track of disk usage.
+   *
+   * @param path        the path to check disk usage in
+   * @param interval    refresh the disk usage at this interval
+   * @param initialUsed use this value until next refresh
+   * @throws IOException if we fail to refresh the disk usage
+   */
+  CachingGetSpaceUsed(File path,
+                      long interval,
+                      long initialUsed) throws IOException {
+    dirPath = path.getCanonicalPath();
+    refreshInterval = interval;
+    used.set(initialUsed);
+  }
+
+  void init() {
+    if (used.get() < 0) {
+      used.set(0);
+      refresh();
+    }
+
+    if (refreshInterval > 0) {
+      refreshUsed = new Thread(new RefreshThread(this),
+          "refreshUsed-" + dirPath);
+      refreshUsed.setDaemon(true);
+      refreshUsed.start();
+    } else {
+      running.set(false);
+      refreshUsed = null;
+    }
+  }
+
+  protected abstract void refresh();
+
+  /**
+   * @return an estimate of space used in the directory path.
+   */
+  @Override public long getUsed() throws IOException {
+    return Math.max(used.get(), 0);
+  }
+
+  /**
+   * @return The directory path being monitored.
+   */
+  public String getDirPath() {
+    return dirPath;
+  }
+
+  /**
+   * Increment the cached value of used space.
+   */
+  public void incDfsUsed(long value) {
+    used.addAndGet(value);
+  }
+
+  /**
+   * Is the background thread running.
+   */
+  boolean running() {
+    return running.get();
+  }
+
+  /**
+   * How long in between runs of the background refresh.
+   */
+  long getRefreshInterval() {
+    return refreshInterval;
+  }
+
+  /**
+   * Reset the current used data amount. This should be called
+   * when the cached value is re-computed.
+   *
+   * @param usedValue new value that should be the disk usage.
+   */
+  protected void setUsed(long usedValue) {
+    this.used.set(usedValue);
+  }
+
+  @Override
+  public void close() throws IOException {
+    running.set(false);
+    if (refreshUsed != null) {
+      refreshUsed.interrupt();
+    }
+  }
+
+  private static final class RefreshThread implements Runnable {
+
+    final CachingGetSpaceUsed spaceUsed;
+
+    RefreshThread(CachingGetSpaceUsed spaceUsed) {
+      this.spaceUsed = spaceUsed;
+    }
+
+    @Override
+    public void run() {
+      while (spaceUsed.running()) {
+        try {
+          Thread.sleep(spaceUsed.getRefreshInterval());
+          // update the used variable
+          spaceUsed.refresh();
+        } catch (InterruptedException e) {
+          LOG.warn("Thread Interrupted waiting to refresh disk information", e);
+          Thread.currentThread().interrupt();
+        }
+      }
+    }
+  }
+}

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

@@ -182,20 +182,18 @@ public abstract class ChecksumFileSystem extends FilterFileSystem {
     public int read(long position, byte[] b, int off, int len)
       throws IOException {
       // parameter check
-      if ((off | len | (off + len) | (b.length - (off + len))) < 0) {
-        throw new IndexOutOfBoundsException();
-      } else if (len == 0) {
+      validatePositionedReadArgs(position, b, off, len);
+      if (len == 0) {
         return 0;
       }
-      if( position<0 ) {
-        throw new IllegalArgumentException(
-            "Parameter position can not to be negative");
-      }
 
-      ChecksumFSInputChecker checker = new ChecksumFSInputChecker(fs, file);
-      checker.seek(position);
-      int nread = checker.read(b, off, len);
-      checker.close();
+      int nread;
+      try (ChecksumFSInputChecker checker =
+               new ChecksumFSInputChecker(fs, file)) {
+        checker.seek(position);
+        nread = checker.read(b, off, len);
+        checker.close();
+      }
       return nread;
     }
     

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

@@ -164,28 +164,26 @@ public abstract class ChecksumFs extends FilterFs {
     public int available() throws IOException {
       return datas.available() + super.available();
     }
-    
+
     @Override
     public int read(long position, byte[] b, int off, int len)
       throws IOException, UnresolvedLinkException {
       // parameter check
-      if ((off | len | (off + len) | (b.length - (off + len))) < 0) {
-        throw new IndexOutOfBoundsException();
-      } else if (len == 0) {
+      validatePositionedReadArgs(position, b, off, len);
+      if (len == 0) {
         return 0;
       }
-      if (position<0) {
-        throw new IllegalArgumentException(
-            "Parameter position can not to be negative");
-      }
 
-      ChecksumFSInputChecker checker = new ChecksumFSInputChecker(fs, file);
-      checker.seek(position);
-      int nread = checker.read(b, off, len);
-      checker.close();
+      int nread;
+      try (ChecksumFSInputChecker checker =
+               new ChecksumFSInputChecker(fs, file)) {
+        checker.seek(position);
+        nread = checker.read(b, off, len);
+        checker.close();
+      }
       return nread;
     }
-    
+
     @Override
     public void close() throws IOException {
       datas.close();

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

@@ -17,227 +17,73 @@
  */
 package org.apache.hadoop.fs;
 
+import com.google.common.annotations.VisibleForTesting;
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
 import org.apache.hadoop.conf.Configuration;
-import org.apache.hadoop.fs.CommonConfigurationKeys;
 import org.apache.hadoop.util.Shell;
 
 import java.io.BufferedReader;
 import java.io.File;
 import java.io.IOException;
-import java.util.concurrent.atomic.AtomicLong;
 
-/** Filesystem disk space usage statistics.  Uses the unix 'du' program*/
+/** Filesystem disk space usage statistics.  Uses the unix 'du' program */
 @InterfaceAudience.LimitedPrivate({"HDFS", "MapReduce"})
 @InterfaceStability.Evolving
-public class DU extends Shell {
-  private String  dirPath;
+public class DU extends CachingGetSpaceUsed {
+  private DUShell duShell;
 
-  private AtomicLong used = new AtomicLong();
-  private volatile boolean shouldRun = true;
-  private Thread refreshUsed;
-  private IOException duException = null;
-  private long refreshInterval;
-  
-  /**
-   * Keeps track of disk usage.
-   * @param path the path to check disk usage in
-   * @param interval refresh the disk usage at this interval
-   * @throws IOException if we fail to refresh the disk usage
-   */
-  public DU(File path, long interval) throws IOException {
-    this(path, interval, -1L);
+  @VisibleForTesting
+   public DU(File path, long interval, long initialUsed) throws IOException {
+    super(path, interval, initialUsed);
   }
-  
-  /**
-   * Keeps track of disk usage.
-   * @param path the path to check disk usage in
-   * @param interval refresh the disk usage at this interval
-   * @param initialUsed use this value until next refresh
-   * @throws IOException if we fail to refresh the disk usage
-   */
-  public DU(File path, long interval, long initialUsed) throws IOException { 
-    super(0);
 
-    //we set the Shell interval to 0 so it will always run our command
-    //and use this one to set the thread sleep interval
-    this.refreshInterval = interval;
-    this.dirPath = path.getCanonicalPath();
-
-    //populate the used variable if the initial value is not specified.
-    if (initialUsed < 0) {
-      run();
-    } else {
-      this.used.set(initialUsed);
-    }
-  }
-
-  /**
-   * Keeps track of disk usage.
-   * @param path the path to check disk usage in
-   * @param conf configuration object
-   * @throws IOException if we fail to refresh the disk usage
-   */
-  public DU(File path, Configuration conf) throws IOException {
-    this(path, conf, -1L);
+  public DU(CachingGetSpaceUsed.Builder builder) throws IOException {
+    this(builder.getPath(), builder.getInterval(), builder.getInitialUsed());
   }
 
-  /**
-   * Keeps track of disk usage.
-   * @param path the path to check disk usage in
-   * @param conf configuration object
-   * @param initialUsed use it until the next refresh.
-   * @throws IOException if we fail to refresh the disk usage
-   */
-  public DU(File path, Configuration conf, long initialUsed)
-      throws IOException {
-    this(path, conf.getLong(CommonConfigurationKeys.FS_DU_INTERVAL_KEY,
-                CommonConfigurationKeys.FS_DU_INTERVAL_DEFAULT), initialUsed);
+  @Override
+  protected synchronized void refresh() {
+    if (duShell == null) {
+      duShell = new DUShell();
+    }
+    try {
+      duShell.startRefresh();
+    } catch (IOException ioe) {
+      LOG.warn("Could not get disk usage information", ioe);
+    }
   }
-    
-  
 
-  /**
-   * This thread refreshes the "used" variable.
-   * 
-   * Future improvements could be to not permanently
-   * run this thread, instead run when getUsed is called.
-   **/
-  class DURefreshThread implements Runnable {
-    
+  private final class DUShell extends Shell  {
+    void startRefresh() throws IOException {
+      run();
+    }
     @Override
-    public void run() {
-      
-      while(shouldRun) {
+    public String toString() {
+      return
+          "du -sk " + getDirPath() + "\n" + used.get() + "\t" + getDirPath();
+    }
 
-        try {
-          Thread.sleep(refreshInterval);
-          
-          try {
-            //update the used variable
-            DU.this.run();
-          } catch (IOException e) {
-            synchronized (DU.this) {
-              //save the latest exception so we can return it in getUsed()
-              duException = e;
-            }
-            
-            LOG.warn("Could not get disk usage information", e);
-          }
-        } catch (InterruptedException e) {
-        }
-      }
+    @Override
+    protected String[] getExecString() {
+      return new String[]{"du", "-sk", getDirPath()};
     }
-  }
-  
-  /**
-   * Decrease how much disk space we use.
-   * @param value decrease by this value
-   */
-  public void decDfsUsed(long value) {
-    used.addAndGet(-value);
-  }
 
-  /**
-   * Increase how much disk space we use.
-   * @param value increase by this value
-   */
-  public void incDfsUsed(long value) {
-    used.addAndGet(value);
-  }
-  
-  /**
-   * @return disk space used 
-   * @throws IOException if the shell command fails
-   */
-  public long getUsed() throws IOException {
-    //if the updating thread isn't started, update on demand
-    if(refreshUsed == null) {
-      run();
-    } else {
-      synchronized (DU.this) {
-        //if an exception was thrown in the last run, rethrow
-        if(duException != null) {
-          IOException tmp = duException;
-          duException = null;
-          throw tmp;
-        }
+    @Override
+    protected void parseExecResult(BufferedReader lines) throws IOException {
+      String line = lines.readLine();
+      if (line == null) {
+        throw new IOException("Expecting a line not the end of stream");
+      }
+      String[] tokens = line.split("\t");
+      if (tokens.length == 0) {
+        throw new IOException("Illegal du output");
       }
+      setUsed(Long.parseLong(tokens[0]) * 1024);
     }
-    
-    return Math.max(used.longValue(), 0L);
-  }
-
-  /**
-   * @return the path of which we're keeping track of disk usage
-   */
-  public String getDirPath() {
-    return dirPath;
-  }
 
-
-  /**
-   * Override to hook in DUHelper class. Maybe this can be used more
-   * generally as well on Unix/Linux based systems
-   */
-  @Override
-  protected void run() throws IOException {
-    if (WINDOWS) {
-      used.set(DUHelper.getFolderUsage(dirPath));
-      return;
-    }
-    super.run();
-  }
-  
-  /**
-   * Start the disk usage checking thread.
-   */
-  public void start() {
-    //only start the thread if the interval is sane
-    if(refreshInterval > 0) {
-      refreshUsed = new Thread(new DURefreshThread(), 
-          "refreshUsed-"+dirPath);
-      refreshUsed.setDaemon(true);
-      refreshUsed.start();
-    }
-  }
-  
-  /**
-   * Shut down the refreshing thread.
-   */
-  public void shutdown() {
-    this.shouldRun = false;
-    
-    if(this.refreshUsed != null) {
-      this.refreshUsed.interrupt();
-    }
-  }
-  
-  @Override
-  public String toString() {
-    return
-      "du -sk " + dirPath +"\n" +
-      used + "\t" + dirPath;
   }
 
-  @Override
-  protected String[] getExecString() {
-    return new String[] {"du", "-sk", dirPath};
-  }
-  
-  @Override
-  protected void parseExecResult(BufferedReader lines) throws IOException {
-    String line = lines.readLine();
-    if (line == null) {
-      throw new IOException("Expecting a line not the end of stream");
-    }
-    String[] tokens = line.split("\t");
-    if(tokens.length == 0) {
-      throw new IOException("Illegal du output");
-    }
-    this.used.set(Long.parseLong(tokens[0])*1024);
-  }
 
   public static void main(String[] args) throws Exception {
     String path = ".";
@@ -245,6 +91,10 @@ public class DU extends Shell {
       path = args[0];
     }
 
-    System.out.println(new DU(new File(path), new Configuration()).toString());
+    GetSpaceUsed du = new GetSpaceUsed.Builder().setPath(new File(path))
+                                                .setConf(new Configuration())
+                                                .build();
+    String duResult = du.toString();
+    System.out.println(duResult);
   }
 }

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

@@ -18,18 +18,21 @@
  */
 package org.apache.hadoop.fs;
 
-import java.io.*;
+import java.io.DataInputStream;
+import java.io.FileDescriptor;
+import java.io.FileInputStream;
+import java.io.IOException;
+import java.io.InputStream;
 import java.nio.ByteBuffer;
 import java.util.EnumSet;
 
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
 import org.apache.hadoop.io.ByteBufferPool;
-import org.apache.hadoop.fs.ByteBufferUtil;
 import org.apache.hadoop.util.IdentityHashStore;
 
 /** Utility that wraps a {@link FSInputStream} in a {@link DataInputStream}
- * and buffers input through a {@link BufferedInputStream}. */
+ * and buffers input through a {@link java.io.BufferedInputStream}. */
 @InterfaceAudience.Public
 @InterfaceStability.Stable
 public class FSDataInputStream extends DataInputStream
@@ -97,6 +100,7 @@ public class FSDataInputStream extends DataInputStream
    * @param buffer    buffer into which data is read
    * @param offset    offset into the buffer in which data is written
    * @param length    the number of bytes to read
+   * @throws IOException IO problems
    * @throws EOFException If the end of stream is reached while reading.
    *                      If an exception is thrown an undetermined number
    *                      of bytes in the buffer may have been written. 

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

@@ -40,4 +40,10 @@ public class FSExceptionMessages {
    */
   public static final String CANNOT_SEEK_PAST_EOF =
       "Attempted to seek or read past the end of the file";
+
+  public static final String EOF_IN_READ_FULLY =
+      "End of file reached before reading fully.";
+
+  public static final String TOO_MANY_BYTES_FOR_DEST_BUFFER
+      = "Requested more bytes than destination buffer size";
 }

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

@@ -17,22 +17,28 @@
  */
 package org.apache.hadoop.fs;
 
-import java.io.*;
-import java.nio.ByteBuffer;
+import java.io.EOFException;
+import java.io.IOException;
+import java.io.InputStream;
 
+import com.google.common.base.Preconditions;
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
-import org.apache.hadoop.fs.ZeroCopyUnavailableException;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
 
 /****************************************************************
  * FSInputStream is a generic old InputStream with a little bit
  * of RAF-style seek ability.
  *
  *****************************************************************/
-@InterfaceAudience.LimitedPrivate({"HDFS"})
-@InterfaceStability.Unstable
+@InterfaceAudience.Public
+@InterfaceStability.Evolving
 public abstract class FSInputStream extends InputStream
     implements Seekable, PositionedReadable {
+  private static final Logger LOG =
+      LoggerFactory.getLogger(FSInputStream.class);
+
   /**
    * Seek to the given offset from the start of the file.
    * The next read() will be from that location.  Can't
@@ -57,32 +63,69 @@ public abstract class FSInputStream extends InputStream
   @Override
   public int read(long position, byte[] buffer, int offset, int length)
     throws IOException {
+    validatePositionedReadArgs(position, buffer, offset, length);
+    if (length == 0) {
+      return 0;
+    }
     synchronized (this) {
       long oldPos = getPos();
       int nread = -1;
       try {
         seek(position);
         nread = read(buffer, offset, length);
+      } catch (EOFException e) {
+        // end of file; this can be raised by some filesystems
+        // (often: object stores); it is swallowed here.
+        LOG.debug("Downgrading EOFException raised trying to" +
+            " read {} bytes at offset {}", length, offset, e);
       } finally {
         seek(oldPos);
       }
       return nread;
     }
   }
-    
+
+  /**
+   * Validation code, available for use in subclasses.
+   * @param position position: if negative an EOF exception is raised
+   * @param buffer destination buffer
+   * @param offset offset within the buffer
+   * @param length length of bytes to read
+   * @throws EOFException if the position is negative
+   * @throws IndexOutOfBoundsException if there isn't space for the amount of
+   * data requested.
+   * @throws IllegalArgumentException other arguments are invalid.
+   */
+  protected void validatePositionedReadArgs(long position,
+      byte[] buffer, int offset, int length) throws EOFException {
+    Preconditions.checkArgument(length >= 0, "length is negative");
+    if (position < 0) {
+      throw new EOFException("position is negative");
+    }
+    Preconditions.checkArgument(buffer != null, "Null buffer");
+    if (buffer.length - offset < length) {
+      throw new IndexOutOfBoundsException(
+          FSExceptionMessages.TOO_MANY_BYTES_FOR_DEST_BUFFER);
+    }
+  }
+
   @Override
   public void readFully(long position, byte[] buffer, int offset, int length)
     throws IOException {
+    validatePositionedReadArgs(position, buffer, offset, length);
     int nread = 0;
     while (nread < length) {
-      int nbytes = read(position+nread, buffer, offset+nread, length-nread);
+      int nbytes = read(position + nread,
+          buffer,
+          offset + nread,
+          length - nread);
       if (nbytes < 0) {
-        throw new EOFException("End of file reached before reading fully.");
+        throw new EOFException(FSExceptionMessages.EOF_IN_READ_FULLY);
       }
       nread += nbytes;
     }
   }
-    
+
   @Override
   public void readFully(long position, byte[] buffer)
     throws IOException {

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

@@ -0,0 +1,147 @@
+/**
+ * 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
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * 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.hadoop.fs;
+
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.util.Shell;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.io.File;
+import java.io.IOException;
+import java.lang.reflect.Constructor;
+import java.lang.reflect.InvocationTargetException;
+
+public interface GetSpaceUsed {
+  long getUsed() throws IOException;
+
+  /**
+   * The builder class
+   */
+  final class Builder {
+    static final Logger LOG = LoggerFactory.getLogger(Builder.class);
+
+    static final String CLASSNAME_KEY = "fs.getspaceused.classname";
+
+    private Configuration conf;
+    private Class<? extends GetSpaceUsed> klass = null;
+    private File path = null;
+    private Long interval = null;
+    private Long initialUsed = null;
+
+    public Configuration getConf() {
+      return conf;
+    }
+
+    public Builder setConf(Configuration conf) {
+      this.conf = conf;
+      return this;
+    }
+
+    public long getInterval() {
+      if (interval != null) {
+        return interval;
+      }
+      long result = CommonConfigurationKeys.FS_DU_INTERVAL_DEFAULT;
+      if (conf == null) {
+        return result;
+      }
+      return conf.getLong(CommonConfigurationKeys.FS_DU_INTERVAL_KEY, result);
+    }
+
+    public Builder setInterval(long interval) {
+      this.interval = interval;
+      return this;
+    }
+
+    public Class<? extends GetSpaceUsed> getKlass() {
+      if (klass != null) {
+        return klass;
+      }
+      Class<? extends GetSpaceUsed> result = null;
+      if (Shell.WINDOWS) {
+        result = WindowsGetSpaceUsed.class;
+      } else {
+        result = DU.class;
+      }
+      if (conf == null) {
+        return result;
+      }
+      return conf.getClass(CLASSNAME_KEY, result, GetSpaceUsed.class);
+    }
+
+    public Builder setKlass(Class<? extends GetSpaceUsed> klass) {
+      this.klass = klass;
+      return this;
+    }
+
+    public File getPath() {
+      return path;
+    }
+
+    public Builder setPath(File path) {
+      this.path = path;
+      return this;
+    }
+
+    public long getInitialUsed() {
+      if (initialUsed == null) {
+        return -1;
+      }
+      return initialUsed;
+    }
+
+    public Builder setInitialUsed(long initialUsed) {
+      this.initialUsed = initialUsed;
+      return this;
+    }
+
+    public GetSpaceUsed build() throws IOException {
+      GetSpaceUsed getSpaceUsed = null;
+      try {
+        Constructor<? extends GetSpaceUsed> cons =
+            getKlass().getConstructor(Builder.class);
+        getSpaceUsed = cons.newInstance(this);
+      } catch (InstantiationException e) {
+        LOG.warn("Error trying to create an instance of " + getKlass(), e);
+      } catch (IllegalAccessException e) {
+        LOG.warn("Error trying to create " + getKlass(), e);
+      } catch (InvocationTargetException e) {
+        LOG.warn("Error trying to create " + getKlass(), e);
+      } catch (NoSuchMethodException e) {
+        LOG.warn("Doesn't look like the class " + getKlass() +
+            " have the needed constructor", e);
+      }
+      // If there were any exceptions then du will be null.
+      // Construct our best guess fallback.
+      if (getSpaceUsed == null) {
+        if (Shell.WINDOWS) {
+          getSpaceUsed = new WindowsGetSpaceUsed(this);
+        } else {
+          getSpaceUsed = new DU(this);
+        }
+      }
+      // Call init after classes constructors have finished.
+      if (getSpaceUsed instanceof CachingGetSpaceUsed) {
+        ((CachingGetSpaceUsed) getSpaceUsed).init();
+      }
+      return getSpaceUsed;
+    }
+
+  }
+}

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

@@ -26,6 +26,7 @@ import org.apache.hadoop.io.Text;
 import org.apache.hadoop.util.LineReader;
 import org.apache.hadoop.util.Progressable;
 
+import java.io.EOFException;
 import java.io.FileNotFoundException;
 import java.io.IOException;
 import java.io.UnsupportedEncodingException;
@@ -1053,16 +1054,15 @@ public class HarFileSystem extends FileSystem {
       @Override
       public void readFully(long pos, byte[] b, int offset, int length) 
       throws IOException {
+        validatePositionedReadArgs(pos, b, offset, length);
+        if (length == 0) {
+          return;
+        }
         if (start + length + pos > end) {
-          throw new IOException("Not enough bytes to read.");
+          throw new EOFException("Not enough bytes to read.");
         }
         underLyingStream.readFully(pos + start, b, offset, length);
       }
-      
-      @Override
-      public void readFully(long pos, byte[] b) throws IOException {
-          readFully(pos, b, 0, b.length);
-      }
 
       @Override
       public void setReadahead(Long readahead) throws IOException {

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

@@ -22,30 +22,67 @@ import java.io.*;
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
 
-/** Stream that permits positional reading. */
+/**
+ * Stream that permits positional reading.
+ *
+ * Implementations are required to implement thread-safe operations; this may
+ * be supported by concurrent access to the data, or by using a synchronization
+ * mechanism to serialize access.
+ *
+ * Not all implementations meet this requirement. Those that do not cannot
+ * be used as a backing store for some applications, such as Apache HBase.
+ *
+ * Independent of whether or not they are thread safe, some implementations
+ * may make the intermediate state of the system, specifically the position
+ * obtained in {@code Seekable.getPos()} visible.
+ */
 @InterfaceAudience.Public
 @InterfaceStability.Evolving
 public interface PositionedReadable {
   /**
-   * Read upto the specified number of bytes, from a given
+   * Read up to the specified number of bytes, from a given
    * position within a file, and return the number of bytes read. This does not
    * change the current offset of a file, and is thread-safe.
+   *
+   * <i>Warning: Not all filesystems satisfy the thread-safety requirement.</i>
+   * @param position position within file
+   * @param buffer destination buffer
+   * @param offset offset in the buffer
+   * @param length number of bytes to read
+   * @return actual number of bytes read; -1 means "none"
+   * @throws IOException IO problems.
    */
-  public int read(long position, byte[] buffer, int offset, int length)
+  int read(long position, byte[] buffer, int offset, int length)
     throws IOException;
   
   /**
    * Read the specified number of bytes, from a given
    * position within a file. This does not
    * change the current offset of a file, and is thread-safe.
+   *
+   * <i>Warning: Not all filesystems satisfy the thread-safety requirement.</i>
+   * @param position position within file
+   * @param buffer destination buffer
+   * @param offset offset in the buffer
+   * @param length number of bytes to read
+   * @throws IOException IO problems.
+   * @throws EOFException the end of the data was reached before
+   * the read operation completed
    */
-  public void readFully(long position, byte[] buffer, int offset, int length)
+  void readFully(long position, byte[] buffer, int offset, int length)
     throws IOException;
   
   /**
    * Read number of bytes equal to the length of the buffer, from a given
    * position within a file. This does not
    * change the current offset of a file, and is thread-safe.
+   *
+   * <i>Warning: Not all filesystems satisfy the thread-safety requirement.</i>
+   * @param position position within file
+   * @param buffer destination buffer
+   * @throws IOException IO problems.
+   * @throws EOFException the end of the data was reached before
+   * the read operation completed
    */
-  public void readFully(long position, byte[] buffer) throws IOException;
+  void readFully(long position, byte[] buffer) throws IOException;
 }

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

@@ -160,6 +160,8 @@ public class RawLocalFileSystem extends FileSystem {
     
     @Override
     public int read(byte[] b, int off, int len) throws IOException {
+      // parameter check
+      validatePositionedReadArgs(position, b, off, len);
       try {
         int value = fis.read(b, off, len);
         if (value > 0) {
@@ -175,6 +177,12 @@ public class RawLocalFileSystem extends FileSystem {
     @Override
     public int read(long position, byte[] b, int off, int len)
       throws IOException {
+      // parameter check
+      validatePositionedReadArgs(position, b, off, len);
+      if (len == 0) {
+        return 0;
+      }
+
       ByteBuffer bb = ByteBuffer.wrap(b, off, len);
       try {
         int value = fis.getChannel().read(bb, position);

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

@@ -0,0 +1,46 @@
+/**
+ * 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
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * 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.hadoop.fs;
+
+import org.apache.hadoop.classification.InterfaceAudience;
+import org.apache.hadoop.classification.InterfaceStability;
+
+import java.io.IOException;
+
+/**
+ * Class to tell the size of a path on windows.
+ * Rather than shelling out, on windows this uses DUHelper.getFolderUsage
+ */
+@InterfaceAudience.LimitedPrivate({"HDFS", "MapReduce"})
+@InterfaceStability.Evolving
+public class WindowsGetSpaceUsed extends CachingGetSpaceUsed {
+
+
+  WindowsGetSpaceUsed(CachingGetSpaceUsed.Builder builder) throws IOException {
+    super(builder.getPath(), builder.getInterval(), builder.getInitialUsed());
+  }
+
+  /**
+   * Override to hook in DUHelper class.
+   */
+  @Override
+  protected void refresh() {
+    used.set(DUHelper.getFolderUsage(getDirPath()));
+  }
+}

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/http/HttpServer2.java

@@ -55,10 +55,7 @@ import org.apache.hadoop.conf.ConfServlet;
 import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.CommonConfigurationKeys;
 import org.apache.hadoop.security.AuthenticationFilterInitializer;
-import org.apache.hadoop.security.authentication.util.FileSignerSecretProvider;
-import org.apache.hadoop.security.authentication.util.RandomSignerSecretProvider;
 import org.apache.hadoop.security.authentication.util.SignerSecretProvider;
-import org.apache.hadoop.security.authentication.util.ZKSignerSecretProvider;
 import org.apache.hadoop.security.ssl.SslSocketConnectorSecure;
 import org.apache.hadoop.jmx.JMXJsonServlet;
 import org.apache.hadoop.log.LogLevel;
@@ -98,8 +95,6 @@ import com.google.common.base.Preconditions;
 import com.google.common.collect.Lists;
 import com.sun.jersey.spi.container.servlet.ServletContainer;
 
-import static org.apache.hadoop.security.authentication.server
-    .AuthenticationFilter.*;
 /**
  * Create a Jetty embedded server to answer http requests. The primary goal is
  * to serve up status information for the server. There are three contexts:
@@ -1124,9 +1119,11 @@ public final class HttpServer2 implements FilterContainer {
   /**
    * A Servlet input filter that quotes all HTML active characters in the
    * parameter names and values. The goal is to quote the characters to make
-   * all of the servlets resistant to cross-site scripting attacks.
+   * all of the servlets resistant to cross-site scripting attacks. It also
+   * sets X-FRAME-OPTIONS in the header to mitigate clickjacking attacks.
    */
   public static class QuotingInputFilter implements Filter {
+    private static final XFrameOption X_FRAME_OPTION = XFrameOption.SAMEORIGIN;
     private FilterConfig config;
 
     public static class RequestQuoter extends HttpServletRequestWrapper {
@@ -1246,6 +1243,7 @@ public final class HttpServer2 implements FilterContainer {
       } else if (mime.startsWith("application/xml")) {
         httpResponse.setContentType("text/xml; charset=utf-8");
       }
+      httpResponse.addHeader("X-FRAME-OPTIONS", X_FRAME_OPTION.toString());
       chain.doFilter(quoted, httpResponse);
     }
 
@@ -1262,4 +1260,23 @@ public final class HttpServer2 implements FilterContainer {
     }
 
   }
+
+  /**
+   * The X-FRAME-OPTIONS header in HTTP response to mitigate clickjacking
+   * attack.
+   */
+  public enum XFrameOption {
+    DENY("DENY") , SAMEORIGIN ("SAMEORIGIN"), ALLOWFROM ("ALLOW-FROM");
+
+    XFrameOption(String name) {
+      this.name = name;
+    }
+
+    private final String name;
+
+    @Override
+    public String toString() {
+      return this.name;
+    }
+  }
 }

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/AbstractMapWritable.java

@@ -181,20 +181,22 @@ public abstract class AbstractMapWritable implements Writable, Configurable {
   public void readFields(DataInput in) throws IOException {
     
     // Get the number of "unknown" classes
-    
     newClasses = in.readByte();
-    
+
+    // Use the classloader of the current thread to load classes instead of the
+    // system-classloader so as to support both client-only and inside-a-MR-job
+    // use-cases. The context-loader by default eventually falls back to the
+    // system one, so there should be no cases where changing this is an issue.
+    ClassLoader classLoader = Thread.currentThread().getContextClassLoader();
+
     // Then read in the class names and add them to our tables
-    
     for (int i = 0; i < newClasses; i++) {
       byte id = in.readByte();
       String className = in.readUTF();
       try {
-        addToMap(Class.forName(className), id);
-        
+        addToMap(classLoader.loadClass(className), id);
       } catch (ClassNotFoundException e) {
-        throw new IOException("can't find class: " + className + " because "+
-            e.getMessage());
+        throw new IOException(e);
       }
     }
   }    

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/metrics2/lib/Interns.java

@@ -18,15 +18,15 @@
 
 package org.apache.hadoop.metrics2.lib;
 
-import java.util.Map;
-import java.util.LinkedHashMap;
-
-import org.apache.commons.logging.Log;
-import org.apache.commons.logging.LogFactory;
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
 import org.apache.hadoop.metrics2.MetricsInfo;
 import org.apache.hadoop.metrics2.MetricsTag;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.util.LinkedHashMap;
+import java.util.Map;
 
 /**
  * Helpers to create interned metrics info
@@ -34,7 +34,7 @@ import org.apache.hadoop.metrics2.MetricsTag;
 @InterfaceAudience.Public
 @InterfaceStability.Evolving
 public class Interns {
-  private static final Log LOG = LogFactory.getLog(Interns.class);
+  private static final Logger LOG = LoggerFactory.getLogger(Interns.class);
 
   // A simple intern cache with two keys
   // (to avoid creating new (combined) key objects for lookup)
@@ -47,7 +47,7 @@ public class Interns {
       protected boolean removeEldestEntry(Map.Entry<K1, Map<K2, V>> e) {
         boolean overflow = expireKey1At(size());
         if (overflow && !gotOverflow) {
-          LOG.warn("Metrics intern cache overflow at "+ size() +" for "+ e);
+          LOG.info("Metrics intern cache overflow at {} for {}", size(), e);
           gotOverflow = true;
         }
         return overflow;
@@ -67,7 +67,7 @@ public class Interns {
           @Override protected boolean removeEldestEntry(Map.Entry<K2, V> e) {
             boolean overflow = expireKey2At(size());
             if (overflow && !gotOverflow) {
-              LOG.warn("Metrics intern cache overflow at "+ size() +" for "+ e);
+              LOG.info("Metrics intern cache overflow at {} for {}", size(), e);
               gotOverflow = true;
             }
             return overflow;

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/util/ShutdownHookManager.java

@@ -81,7 +81,7 @@ public class ShutdownHookManager {
                 LOG.error("ShutdownHookManger shutdown forcefully.");
                 EXECUTOR.shutdownNow();
               }
-              LOG.info("ShutdownHookManger complete shutdown.");
+              LOG.debug("ShutdownHookManger complete shutdown.");
             } catch (InterruptedException ex) {
               LOG.error("ShutdownHookManger interrupted while waiting for " +
                   "termination.", ex);

hadoop-common-project/hadoop-common/src/site/markdown/DeprecatedProperties.md

@@ -51,6 +51,7 @@ The following table lists the configuration property names that are deprecated i
 | dfs.secondary.http.address | dfs.namenode.secondary.http-address |
 | dfs.socket.timeout | dfs.client.socket-timeout |
 | dfs.umaskmode | fs.permissions.umask-mode |
+| dfs.web.ugi | hadoop.http.staticuser.user |
 | dfs.write.packet.size | dfs.client-write-packet-size |
 | fs.checkpoint.dir | dfs.namenode.checkpoint.dir |
 | fs.checkpoint.edits.dir | dfs.namenode.checkpoint.edits.dir |

hadoop-common-project/hadoop-common/src/site/markdown/filesystem/fsdatainputstream.md

@@ -120,7 +120,8 @@ Return the data at the current position.
 ### <a name="InputStream.read.buffer[]"></a> `InputStream.read(buffer[], offset, length)`
 
 Read `length` bytes of data into the destination buffer, starting at offset
-`offset`
+`offset`. The source of the data is the current position of the stream,
+as implicitly set in `pos`
 
 #### Preconditions
 
@@ -129,6 +130,7 @@ Read `length` bytes of data into the destination buffer, starting at offset
     length >= 0
     offset < len(buffer)
     length <= len(buffer) - offset
+    pos >= 0 else raise EOFException, IOException
 
 Exceptions that may be raised on precondition failure are
 
@@ -136,20 +138,39 @@ Exceptions that may be raised on precondition failure are
     ArrayIndexOutOfBoundsException
     RuntimeException
 
+Not all filesystems check the `isOpen` state.
+
 #### Postconditions
 
     if length == 0 :
       result = 0
 
-    elseif pos > len(data):
-      result -1
+    else if pos > len(data):
+      result = -1
 
     else
       let l = min(length, len(data)-length) :
-          buffer' = buffer where forall i in [0..l-1]:
-              buffer'[o+i] = data[pos+i]
-          FSDIS' = (pos+l, data, true)
-          result = l
+        buffer' = buffer where forall i in [0..l-1]:
+           buffer'[o+i] = data[pos+i]
+        FSDIS' = (pos+l, data, true)
+        result = l
+
+The `java.io` API states that if the amount of data to be read (i.e. `length`)
+then the call must block until the amount of data available is greater than
+zero —that is, until there is some data. The call is not required to return
+when the buffer is full, or indeed block until there is no data left in
+the stream.
+
+That is, rather than `l` being simply defined as `min(length, len(data)-length)`,
+it strictly is an integer in the range `1..min(length, len(data)-length)`.
+While the caller may expect for as much as the buffer as possible to be filled
+in, it is within the specification for an implementation to always return
+a smaller number, perhaps only ever 1 byte.
+
+What is critical is that unless the destination buffer size is 0, the call
+must block until at least one byte is returned. Thus, for any data source
+of length greater than zero, repeated invocations of this `read()` operation
+will eventually read all the data.
 
 ### <a name="Seekable.seek"></a>`Seekable.seek(s)`
 
@@ -279,6 +300,9 @@ on the underlying stream:
     read(dest3, ... len3) -> dest3[0..len3 - 1] =
       [data(FS, path, pos3), data(FS, path, pos3 + 1) ... data(FS, path, pos3 + len3 - 1]
 
+Note that implementations are not required to be atomic; the intermediate state
+of the operation (the change in the value of `getPos()`) may be visible.
+
 #### Implementation preconditions
 
 Not all `FSDataInputStream` implementations support these operations. Those that do
@@ -287,7 +311,7 @@ interface.
 
     supported(FSDIS, Seekable.seek) else raise [UnsupportedOperationException, IOException]
 
-This could be considered obvious: if a stream is not Seekable, a client
+This could be considered obvious: if a stream is not `Seekable`, a client
 cannot seek to a location. It is also a side effect of the
 base class implementation, which uses `Seekable.seek()`.
 
@@ -304,14 +328,14 @@ For any operations that fail, the contents of the destination
 `buffer` are undefined. Implementations may overwrite part
 or all of the buffer before reporting a failure.
 
-
-
 ### `int PositionedReadable.read(position, buffer, offset, length)`
 
+Read as much data as possible into the buffer space allocated for it.
+
 #### Preconditions
 
-    position > 0 else raise [IllegalArgumentException, RuntimeException]
-    len(buffer) + offset < len(data) else raise [IndexOutOfBoundException, RuntimeException]
+    position >= 0 else raise [EOFException, IOException, IllegalArgumentException, RuntimeException]
+    len(buffer) - offset >= length else raise [IndexOutOfBoundException, RuntimeException]
     length >= 0
     offset >= 0
 
@@ -324,23 +348,36 @@ of data available from the specified position:
     buffer'[offset..(offset+available-1)] = data[position..position+available -1]
     result = available
 
+1. A return value of -1 means that the stream had no more available data.
+1. An invocation with `length==0` implicitly does not read any data;
+implementations may short-cut the operation and omit any IO. In such instances,
+checks for the stream being at the end of the file may be omitted.
+1. If an IO exception occurs during the read operation(s),
+the final state of `buffer` is undefined.
 
 ### `void PositionedReadable.readFully(position, buffer, offset, length)`
 
+Read exactly `length` bytes of data into the buffer, failing if there is not
+enough data available.
+
 #### Preconditions
 
-    position > 0 else raise [IllegalArgumentException, RuntimeException]
+    position >= 0 else raise [EOFException, IOException, IllegalArgumentException, RuntimeException]
     length >= 0
     offset >= 0
+    len(buffer) - offset >= length else raise [IndexOutOfBoundException, RuntimeException]
     (position + length) <= len(data) else raise [EOFException, IOException]
-    len(buffer) + offset < len(data)
+
+If an IO exception occurs during the read operation(s),
+the final state of `buffer` is undefined.
+
+If there is not enough data in the input stream to satisfy the requests,
+the final state of `buffer` is undefined.
 
 #### Postconditions
 
-The amount of data read is the less of the length or the amount
-of data available from the specified position:
+The buffer from offset `offset` is filled with the data starting at `position`
 
-    let available = min(length, len(data)-position)
     buffer'[offset..(offset+length-1)] = data[position..(position + length -1)]
 
 ### `PositionedReadable.readFully(position, buffer)`
@@ -349,6 +386,9 @@ The semantics of this are exactly equivalent to
 
     readFully(position, buffer, 0, len(buffer))
 
+That is, the buffer is filled entirely with the contents of the input source
+from position `position`
+
 
 ## Consistency
 

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/crypto/CryptoStreamsTestBase.java

@@ -17,6 +17,7 @@
  */
 package org.apache.hadoop.crypto;
 
+import java.io.EOFException;
 import java.io.IOException;
 import java.io.InputStream;
 import java.io.OutputStream;
@@ -29,6 +30,7 @@ import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
 import org.apache.hadoop.fs.ByteBufferReadable;
 import org.apache.hadoop.fs.FSDataOutputStream;
+import org.apache.hadoop.fs.FSExceptionMessages;
 import org.apache.hadoop.fs.HasEnhancedByteBufferAccess;
 import org.apache.hadoop.fs.PositionedReadable;
 import org.apache.hadoop.fs.ReadOption;
@@ -339,7 +341,7 @@ public abstract class CryptoStreamsTestBase {
     try {
       ((PositionedReadable) in).readFully(pos, result);
       Assert.fail("Read fully exceeds maximum length should fail.");
-    } catch (IOException e) {
+    } catch (EOFException e) {
     }
   }
   
@@ -365,9 +367,9 @@ public abstract class CryptoStreamsTestBase {
     try {
       seekCheck(in, -3);
       Assert.fail("Seek to negative offset should fail.");
-    } catch (IllegalArgumentException e) {
-      GenericTestUtils.assertExceptionContains("Cannot seek to negative " +
-          "offset", e);
+    } catch (EOFException e) {
+      GenericTestUtils.assertExceptionContains(
+          FSExceptionMessages.NEGATIVE_SEEK, e);
     }
     Assert.assertEquals(pos, ((Seekable) in).getPos());
     

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/FileContextMainOperationsBaseTest.java

@@ -272,7 +272,8 @@ public abstract class FileContextMainOperationsBaseTest  {
       // expected
     }
   } 
-  
+
+  @Test
   public void testListStatusThrowsExceptionForNonExistentFile()
                                                     throws Exception {
     try {

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestDU.java

@@ -28,7 +28,7 @@ import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.CommonConfigurationKeys;
 import org.apache.hadoop.test.GenericTestUtils;
 
-/** This test makes sure that "DU" does not get to run on each call to getUsed */ 
+/** This test makes sure that "DU" does not get to run on each call to getUsed */
 public class TestDU extends TestCase {
   final static private File DU_DIR = GenericTestUtils.getTestDir("dutmp");
 
@@ -42,7 +42,7 @@ public class TestDU extends TestCase {
   public void tearDown() throws IOException {
       FileUtil.fullyDelete(DU_DIR);
   }
-    
+
   private void createFile(File newFile, int size) throws IOException {
     // write random data so that filesystems with compression enabled (e.g., ZFS)
     // can't compress the file
@@ -54,18 +54,18 @@ public class TestDU extends TestCase {
     RandomAccessFile file = new RandomAccessFile(newFile, "rws");
 
     file.write(data);
-      
+
     file.getFD().sync();
     file.close();
   }
 
   /**
    * Verify that du returns expected used space for a file.
-   * We assume here that if a file system crates a file of size 
+   * We assume here that if a file system crates a file of size
    * that is a multiple of the block size in this file system,
    * then the used size for the file will be exactly that size.
    * This is true for most file systems.
-   * 
+   *
    * @throws IOException
    * @throws InterruptedException
    */
@@ -78,28 +78,29 @@ public class TestDU extends TestCase {
     createFile(file, writtenSize);
 
     Thread.sleep(5000); // let the metadata updater catch up
-    
-    DU du = new DU(file, 10000);
-    du.start();
+
+    DU du = new DU(file, 10000, -1);
+    du.init();
     long duSize = du.getUsed();
-    du.shutdown();
+    du.close();
 
     assertTrue("Invalid on-disk size",
         duSize >= writtenSize &&
         writtenSize <= (duSize + slack));
-    
-    //test with 0 interval, will not launch thread 
-    du = new DU(file, 0);
-    du.start();
+
+    //test with 0 interval, will not launch thread
+    du = new DU(file, 0, -1);
+    du.init();
     duSize = du.getUsed();
-    du.shutdown();
-    
+    du.close();
+
     assertTrue("Invalid on-disk size",
         duSize >= writtenSize &&
         writtenSize <= (duSize + slack));
-    
-    //test without launching thread 
-    du = new DU(file, 10000);
+
+    //test without launching thread
+    du = new DU(file, 10000, -1);
+    du.init();
     duSize = du.getUsed();
 
     assertTrue("Invalid on-disk size",
@@ -111,8 +112,8 @@ public class TestDU extends TestCase {
     assertTrue(file.createNewFile());
     Configuration conf = new Configuration();
     conf.setLong(CommonConfigurationKeys.FS_DU_INTERVAL_KEY, 10000L);
-    DU du = new DU(file, conf);
-    du.decDfsUsed(Long.MAX_VALUE);
+    DU du = new DU(file, 10000L, -1);
+    du.incDfsUsed(-Long.MAX_VALUE);
     long duSize = du.getUsed();
     assertTrue(String.valueOf(duSize), duSize >= 0L);
   }
@@ -121,7 +122,7 @@ public class TestDU extends TestCase {
     File file = new File(DU_DIR, "dataX");
     createFile(file, 8192);
     DU du = new DU(file, 3000, 1024);
-    du.start();
+    du.init();
     assertTrue("Initial usage setting not honored", du.getUsed() == 1024);
 
     // wait until the first du runs.
@@ -131,4 +132,7 @@ public class TestDU extends TestCase {
 
     assertTrue("Usage didn't get updated", du.getUsed() == 8192);
   }
+
+
+
 }

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestGetSpaceUsed.java

@@ -0,0 +1,133 @@
+/**
+ * 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
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * 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.hadoop.fs;
+
+import org.apache.hadoop.conf.Configuration;
+import org.junit.After;
+import org.junit.Before;
+import org.junit.Test;
+
+import java.io.File;
+import java.io.IOException;
+
+import static org.junit.Assert.*;
+
+public class TestGetSpaceUsed {
+  final static private File DIR = new File(
+      System.getProperty("test.build.data", "/tmp"), "TestGetSpaceUsed");
+
+  @Before
+  public void setUp() {
+    FileUtil.fullyDelete(DIR);
+    assertTrue(DIR.mkdirs());
+  }
+
+  @After
+  public void tearDown() throws IOException {
+    FileUtil.fullyDelete(DIR);
+  }
+
+  /**
+   * Test that the builder can create a class specified through the class.
+   */
+  @Test
+  public void testBuilderConf() throws Exception {
+    File file = new File(DIR, "testBuilderConf");
+    assertTrue(file.createNewFile());
+    Configuration conf = new Configuration();
+    conf.set("fs.getspaceused.classname", DummyDU.class.getName());
+    CachingGetSpaceUsed instance =
+        (CachingGetSpaceUsed) new CachingGetSpaceUsed.Builder()
+            .setPath(file)
+            .setInterval(0)
+            .setConf(conf)
+            .build();
+    assertNotNull(instance);
+    assertTrue(instance instanceof DummyDU);
+    assertFalse(instance.running());
+    instance.close();
+  }
+
+  @Test
+  public void testBuildInitial() throws Exception {
+    File file = new File(DIR, "testBuildInitial");
+    assertTrue(file.createNewFile());
+    CachingGetSpaceUsed instance =
+        (CachingGetSpaceUsed) new CachingGetSpaceUsed.Builder()
+            .setPath(file)
+            .setInitialUsed(90210)
+            .setKlass(DummyDU.class)
+            .build();
+    assertEquals(90210, instance.getUsed());
+    instance.close();
+  }
+
+  @Test
+  public void testBuildInterval() throws Exception {
+    File file = new File(DIR, "testBuildInitial");
+    assertTrue(file.createNewFile());
+    CachingGetSpaceUsed instance =
+        (CachingGetSpaceUsed) new CachingGetSpaceUsed.Builder()
+            .setPath(file)
+            .setInitialUsed(90210)
+            .setInterval(50060)
+            .setKlass(DummyDU.class)
+            .build();
+    assertEquals(50060, instance.getRefreshInterval());
+    instance.close();
+  }
+
+  @Test
+  public void testBuildNonCaching() throws Exception {
+    File file = new File(DIR, "testBuildNonCaching");
+    assertTrue(file.createNewFile());
+    GetSpaceUsed instance =  new CachingGetSpaceUsed.Builder()
+            .setPath(file)
+            .setInitialUsed(90210)
+            .setInterval(50060)
+            .setKlass(DummyGetSpaceUsed.class)
+            .build();
+    assertEquals(300, instance.getUsed());
+    assertTrue(instance instanceof DummyGetSpaceUsed);
+  }
+
+  private static class DummyDU extends CachingGetSpaceUsed {
+
+    public DummyDU(Builder builder) throws IOException {
+      // Push to the base class.
+      // Most times that's all that will need to be done.
+      super(builder);
+    }
+
+    @Override
+    protected void refresh() {
+      // This is a test so don't du anything.
+    }
+  }
+
+  private static class DummyGetSpaceUsed implements GetSpaceUsed {
+
+    public DummyGetSpaceUsed(GetSpaceUsed.Builder builder) {
+
+    }
+
+    @Override public long getUsed() throws IOException {
+      return 300;
+    }
+  }
+}

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractAppendTest.java

@@ -52,11 +52,8 @@ public abstract class AbstractContractAppendTest extends AbstractFSContractTestB
   public void testAppendToEmptyFile() throws Throwable {
     touch(getFileSystem(), target);
     byte[] dataset = dataset(256, 'a', 'z');
-    FSDataOutputStream outputStream = getFileSystem().append(target);
-    try {
+    try (FSDataOutputStream outputStream = getFileSystem().append(target)) {
       outputStream.write(dataset);
-    } finally {
-      outputStream.close();
     }
     byte[] bytes = ContractTestUtils.readDataset(getFileSystem(), target,
                                                  dataset.length);

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractConcatTest.java

@@ -53,7 +53,7 @@ public abstract class AbstractContractConcatTest extends AbstractFSContractTestB
     target = new Path(testPath, "target");
 
     byte[] block = dataset(TEST_FILE_LEN, 0, 255);
-    createFile(getFileSystem(), srcFile, false, block);
+    createFile(getFileSystem(), srcFile, true, block);
     touch(getFileSystem(), zeroByteFile);
   }
 

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractCreateTest.java

@@ -123,7 +123,7 @@ public abstract class AbstractContractCreateTest extends
     } catch (AssertionError failure) {
       if (isSupported(IS_BLOBSTORE)) {
         // file/directory hack surfaces here
-        throw new AssumptionViolatedException(failure.toString()).initCause(failure);
+        throw new AssumptionViolatedException(failure.toString(), failure);
       }
       // else: rethrow
       throw failure;
@@ -163,13 +163,11 @@ public abstract class AbstractContractCreateTest extends
   public void testCreatedFileIsImmediatelyVisible() throws Throwable {
     describe("verify that a newly created file exists as soon as open returns");
     Path path = path("testCreatedFileIsImmediatelyVisible");
-    FSDataOutputStream out = null;
-    try {
-      out = getFileSystem().create(path,
+    try(FSDataOutputStream out = getFileSystem().create(path,
                                    false,
                                    4096,
                                    (short) 1,
-                                   1024);
+                                   1024)) {
       if (!getFileSystem().exists(path)) {
 
         if (isSupported(IS_BLOBSTORE)) {
@@ -180,8 +178,6 @@ public abstract class AbstractContractCreateTest extends
         assertPathExists("expected path to be visible before anything written",
                          path);
       }
-    } finally {
-      IOUtils.closeStream(out);
     }
   }
 }

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractDeleteTest.java

@@ -47,7 +47,7 @@ public abstract class AbstractContractDeleteTest extends
   @Test
   public void testDeleteNonexistentPathRecursive() throws Throwable {
     Path path = path("testDeleteNonexistentPathRecursive");
-    ContractTestUtils.assertPathDoesNotExist(getFileSystem(), "leftover", path);
+    assertPathDoesNotExist("leftover", path);
     ContractTestUtils.rejectRootOperation(path);
     assertFalse("Returned true attempting to delete"
                 + " a nonexistent path " + path,
@@ -58,7 +58,7 @@ public abstract class AbstractContractDeleteTest extends
   @Test
   public void testDeleteNonexistentPathNonRecursive() throws Throwable {
     Path path = path("testDeleteNonexistentPathNonRecursive");
-    ContractTestUtils.assertPathDoesNotExist(getFileSystem(), "leftover", path);
+    assertPathDoesNotExist("leftover", path);
     ContractTestUtils.rejectRootOperation(path);
     assertFalse("Returned true attempting to recursively delete"
                 + " a nonexistent path " + path,
@@ -81,7 +81,7 @@ public abstract class AbstractContractDeleteTest extends
       //expected
       handleExpectedException(expected);
     }
-    ContractTestUtils.assertIsDirectory(getFileSystem(), path);
+    assertIsDirectory(path);
   }
 
   @Test
@@ -92,7 +92,7 @@ public abstract class AbstractContractDeleteTest extends
     ContractTestUtils.writeTextFile(getFileSystem(), file, "goodbye, world",
                                     true);
     assertDeleted(path, true);
-    ContractTestUtils.assertPathDoesNotExist(getFileSystem(), "not deleted", file);
+    assertPathDoesNotExist("not deleted", file);
   }
 
   @Test
@@ -100,12 +100,11 @@ public abstract class AbstractContractDeleteTest extends
     mkdirs(path("testDeleteDeepEmptyDir/d1/d2/d3/d4"));
     assertDeleted(path("testDeleteDeepEmptyDir/d1/d2/d3"), true);
 
-    FileSystem fs = getFileSystem();
-    ContractTestUtils.assertPathDoesNotExist(fs,
+    assertPathDoesNotExist(
         "not deleted", path("testDeleteDeepEmptyDir/d1/d2/d3/d4"));
-    ContractTestUtils.assertPathDoesNotExist(fs,
+    assertPathDoesNotExist(
         "not deleted", path("testDeleteDeepEmptyDir/d1/d2/d3"));
-    ContractTestUtils.assertPathExists(fs, "parent dir is deleted",
+    assertPathExists( "parent dir is deleted",
         path("testDeleteDeepEmptyDir/d1/d2"));
   }
 
@@ -117,8 +116,7 @@ public abstract class AbstractContractDeleteTest extends
     Path file = new Path(path, "childfile");
     ContractTestUtils.writeTextFile(getFileSystem(), file,
         "single file to be deleted.", true);
-    ContractTestUtils.assertPathExists(getFileSystem(),
-        "single file not created", file);
+    assertPathExists("single file not created", file);
     assertDeleted(file, false);
   }
 }

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractMkdirTest.java

@@ -67,12 +67,9 @@ public abstract class AbstractContractMkdirTest extends AbstractFSContractTestBa
       boolean made = fs.mkdirs(path);
       fail("mkdirs did not fail over a file but returned " + made
             + "; " + ls(path));
-    } catch (ParentNotDirectoryException e) {
+    } catch (ParentNotDirectoryException | FileAlreadyExistsException e) {
       //parent is a directory
       handleExpectedException(e);
-    } catch (FileAlreadyExistsException e) {
-      //also allowed as an exception (HDFS)
-      handleExpectedException(e);;
     } catch (IOException e) {
       //here the FS says "no create"
       handleRelaxedException("mkdirs", "FileAlreadyExistsException", e);
@@ -97,11 +94,9 @@ public abstract class AbstractContractMkdirTest extends AbstractFSContractTestBa
       boolean made = fs.mkdirs(child);
       fail("mkdirs did not fail over a file but returned " + made
            + "; " + ls(path));
-    } catch (ParentNotDirectoryException e) {
+    } catch (ParentNotDirectoryException | FileAlreadyExistsException e) {
       //parent is a directory
       handleExpectedException(e);
-    } catch (FileAlreadyExistsException e) {
-      handleExpectedException(e);
     } catch (IOException e) {
       handleRelaxedException("mkdirs", "ParentNotDirectoryException", e);
     }

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractOpenTest.java

@@ -125,10 +125,10 @@ public abstract class AbstractContractOpenTest extends AbstractFSContractTestBas
     createFile(getFileSystem(), path, false, block);
     //open first
     FSDataInputStream instream1 = getFileSystem().open(path);
-    int c = instream1.read();
-    assertEquals(0,c);
     FSDataInputStream instream2 = null;
     try {
+      int c = instream1.read();
+      assertEquals(0,c);
       instream2 = getFileSystem().open(path);
       assertEquals("first read of instream 2", 0, instream2.read());
       assertEquals("second read of instream 1", 1, instream1.read());

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractRenameTest.java

@@ -26,8 +26,7 @@ import org.junit.Test;
 import java.io.FileNotFoundException;
 import java.io.IOException;
 
-import static org.apache.hadoop.fs.contract.ContractTestUtils.dataset;
-import static org.apache.hadoop.fs.contract.ContractTestUtils.writeDataset;
+import static org.apache.hadoop.fs.contract.ContractTestUtils.*;
 
 /**
  * Test creating files, overwrite options &c
@@ -46,9 +45,9 @@ public abstract class AbstractContractRenameTest extends
     boolean rename = rename(renameSrc, renameTarget);
     assertTrue("rename("+renameSrc+", "+ renameTarget+") returned false",
         rename);
-    ContractTestUtils.assertListStatusFinds(getFileSystem(),
+    assertListStatusFinds(getFileSystem(),
         renameTarget.getParent(), renameTarget);
-    ContractTestUtils.verifyFileContents(getFileSystem(), renameTarget, data);
+    verifyFileContents(getFileSystem(), renameTarget, data);
   }
 
   @Test
@@ -129,7 +128,7 @@ public abstract class AbstractContractRenameTest extends
     }
     // verify that the destination file is as expected based on the expected
     // outcome
-    ContractTestUtils.verifyFileContents(getFileSystem(), destFile,
+    verifyFileContents(getFileSystem(), destFile,
         destUnchanged? destData: srcData);
   }
 
@@ -154,7 +153,7 @@ public abstract class AbstractContractRenameTest extends
     Path renamedSrc = new Path(destDir, sourceSubdir);
     assertIsFile(destFilePath);
     assertIsDirectory(renamedSrc);
-    ContractTestUtils.verifyFileContents(fs, destFilePath, destDateset);
+    verifyFileContents(fs, destFilePath, destDateset);
     assertTrue("rename returned false though the contents were copied", rename);
   }
 
@@ -172,10 +171,10 @@ public abstract class AbstractContractRenameTest extends
       boolean rename = rename(renameSrc, renameTarget);
       if (renameCreatesDestDirs) {
         assertTrue(rename);
-        ContractTestUtils.verifyFileContents(getFileSystem(), renameTarget, data);
+        verifyFileContents(getFileSystem(), renameTarget, data);
       } else {
         assertFalse(rename);
-        ContractTestUtils.verifyFileContents(getFileSystem(), renameSrc, data);
+        verifyFileContents(getFileSystem(), renameSrc, data);
       }
     } catch (FileNotFoundException e) {
        // allowed unless that rename flag is set
@@ -191,36 +190,36 @@ public abstract class AbstractContractRenameTest extends
     final Path finalDir = new Path(renameTestDir, "dest");
     FileSystem fs = getFileSystem();
     boolean renameRemoveEmptyDest = isSupported(RENAME_REMOVE_DEST_IF_EMPTY_DIR);
-    ContractTestUtils.rm(fs, renameTestDir, true, false);
+    rm(fs, renameTestDir, true, false);
 
     fs.mkdirs(srcDir);
     fs.mkdirs(finalDir);
-    ContractTestUtils.writeTextFile(fs, new Path(srcDir, "source.txt"),
+    writeTextFile(fs, new Path(srcDir, "source.txt"),
         "this is the file in src dir", false);
-    ContractTestUtils.writeTextFile(fs, new Path(srcSubDir, "subfile.txt"),
+    writeTextFile(fs, new Path(srcSubDir, "subfile.txt"),
         "this is the file in src/sub dir", false);
 
-    ContractTestUtils.assertPathExists(fs, "not created in src dir",
+    assertPathExists("not created in src dir",
         new Path(srcDir, "source.txt"));
-    ContractTestUtils.assertPathExists(fs, "not created in src/sub dir",
+    assertPathExists("not created in src/sub dir",
         new Path(srcSubDir, "subfile.txt"));
 
     fs.rename(srcDir, finalDir);
     // Accept both POSIX rename behavior and CLI rename behavior
     if (renameRemoveEmptyDest) {
       // POSIX rename behavior
-      ContractTestUtils.assertPathExists(fs, "not renamed into dest dir",
+      assertPathExists("not renamed into dest dir",
           new Path(finalDir, "source.txt"));
-      ContractTestUtils.assertPathExists(fs, "not renamed into dest/sub dir",
+      assertPathExists("not renamed into dest/sub dir",
           new Path(finalDir, "sub/subfile.txt"));
     } else {
       // CLI rename behavior
-      ContractTestUtils.assertPathExists(fs, "not renamed into dest dir",
+      assertPathExists("not renamed into dest dir",
           new Path(finalDir, "src1/source.txt"));
-      ContractTestUtils.assertPathExists(fs, "not renamed into dest/sub dir",
+      assertPathExists("not renamed into dest/sub dir",
           new Path(finalDir, "src1/sub/subfile.txt"));
     }
-    ContractTestUtils.assertPathDoesNotExist(fs, "not deleted",
+    assertPathDoesNotExist("not deleted",
         new Path(srcDir, "source.txt"));
   }
 }

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractRootDirectoryTest.java

@@ -51,7 +51,7 @@ public abstract class AbstractContractRootDirectoryTest extends AbstractFSContra
     Path dir = new Path("/testmkdirdepth1");
     assertPathDoesNotExist("directory already exists", dir);
     fs.mkdirs(dir);
-    ContractTestUtils.assertIsDirectory(getFileSystem(), dir);
+    assertIsDirectory(dir);
     assertPathExists("directory already exists", dir);
     assertDeleted(dir, true);
   }
@@ -61,10 +61,10 @@ public abstract class AbstractContractRootDirectoryTest extends AbstractFSContra
     //extra sanity checks here to avoid support calls about complete loss of data
     skipIfUnsupported(TEST_ROOT_TESTS_ENABLED);
     Path root = new Path("/");
-    ContractTestUtils.assertIsDirectory(getFileSystem(), root);
+    assertIsDirectory(root);
     boolean deleted = getFileSystem().delete(root, true);
     LOG.info("rm / of empty dir result is {}", deleted);
-    ContractTestUtils.assertIsDirectory(getFileSystem(), root);
+    assertIsDirectory(root);
   }
 
   @Test
@@ -75,7 +75,7 @@ public abstract class AbstractContractRootDirectoryTest extends AbstractFSContra
     String touchfile = "/testRmNonEmptyRootDirNonRecursive";
     Path file = new Path(touchfile);
     ContractTestUtils.touch(getFileSystem(), file);
-    ContractTestUtils.assertIsDirectory(getFileSystem(), root);
+    assertIsDirectory(root);
     try {
       boolean deleted = getFileSystem().delete(root, false);
       fail("non recursive delete should have raised an exception," +
@@ -86,7 +86,7 @@ public abstract class AbstractContractRootDirectoryTest extends AbstractFSContra
     } finally {
       getFileSystem().delete(file, false);
     }
-    ContractTestUtils.assertIsDirectory(getFileSystem(), root);
+    assertIsDirectory(root);
   }
 
   @Test
@@ -94,11 +94,11 @@ public abstract class AbstractContractRootDirectoryTest extends AbstractFSContra
     //extra sanity checks here to avoid support calls about complete loss of data
     skipIfUnsupported(TEST_ROOT_TESTS_ENABLED);
     Path root = new Path("/");
-    ContractTestUtils.assertIsDirectory(getFileSystem(), root);
+    assertIsDirectory(root);
     Path file = new Path("/testRmRootRecursive");
     ContractTestUtils.touch(getFileSystem(), file);
     boolean deleted = getFileSystem().delete(root, true);
-    ContractTestUtils.assertIsDirectory(getFileSystem(), root);
+    assertIsDirectory(root);
     LOG.info("rm -rf / result is {}", deleted);
     if (deleted) {
       assertPathDoesNotExist("expected file to be deleted", file);

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractContractSeekTest.java

@@ -21,6 +21,7 @@ package org.apache.hadoop.fs.contract;
 import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.CommonConfigurationKeysPublic;
 import org.apache.hadoop.fs.FSDataInputStream;
+import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.io.IOUtils;
 import org.junit.Test;
@@ -31,9 +32,9 @@ import java.io.EOFException;
 import java.io.IOException;
 import java.util.Random;
 
-import static org.apache.hadoop.fs.contract.ContractTestUtils.cleanup;
 import static org.apache.hadoop.fs.contract.ContractTestUtils.createFile;
 import static org.apache.hadoop.fs.contract.ContractTestUtils.dataset;
+import static org.apache.hadoop.fs.contract.ContractTestUtils.skip;
 import static org.apache.hadoop.fs.contract.ContractTestUtils.touch;
 import static org.apache.hadoop.fs.contract.ContractTestUtils.verifyRead;
 
@@ -46,7 +47,6 @@ public abstract class AbstractContractSeekTest extends AbstractFSContractTestBas
 
   public static final int DEFAULT_RANDOM_SEEK_COUNT = 100;
 
-  private Path testPath;
   private Path smallSeekFile;
   private Path zeroByteFile;
   private FSDataInputStream instream;
@@ -56,13 +56,13 @@ public abstract class AbstractContractSeekTest extends AbstractFSContractTestBas
     super.setup();
     skipIfUnsupported(SUPPORTS_SEEK);
     //delete the test directory
-    testPath = getContract().getTestPath();
     smallSeekFile = path("seekfile.txt");
     zeroByteFile = path("zero.txt");
     byte[] block = dataset(TEST_FILE_LEN, 0, 255);
     //this file now has a simple rule: offset => value
-    createFile(getFileSystem(), smallSeekFile, false, block);
-    touch(getFileSystem(), zeroByteFile);
+    FileSystem fs = getFileSystem();
+    createFile(fs, smallSeekFile, true, block);
+    touch(fs, zeroByteFile);
   }
 
   @Override
@@ -79,6 +79,21 @@ public abstract class AbstractContractSeekTest extends AbstractFSContractTestBas
     super.teardown();
   }
 
+  /**
+   * Skip a test case if the FS doesn't support positioned readable.
+   * This should hold automatically if the FS supports seek, even
+   * if it doesn't support seeking past the EOF.
+   * And, because this test suite requires seek to be supported, the
+   * feature is automatically assumed to be true unless stated otherwise.
+   */
+  protected void assumeSupportsPositionedReadable() throws IOException {
+    // because this ,
+    if (!getContract().isSupported(SUPPORTS_POSITIONED_READABLE, true)) {
+      skip("Skipping as unsupported feature: "
+          + SUPPORTS_POSITIONED_READABLE);
+    }
+  }
+
   @Test
   public void testSeekZeroByteFile() throws Throwable {
     describe("seek and read a 0 byte file");
@@ -282,6 +297,7 @@ public abstract class AbstractContractSeekTest extends AbstractFSContractTestBas
   public void testPositionedBulkReadDoesntChangePosition() throws Throwable {
     describe(
       "verify that a positioned read does not change the getPos() value");
+    assumeSupportsPositionedReadable();
     Path testSeekFile = path("bigseekfile.txt");
     byte[] block = dataset(65536, 0, 255);
     createFile(getFileSystem(), testSeekFile, false, block);
@@ -290,8 +306,9 @@ public abstract class AbstractContractSeekTest extends AbstractFSContractTestBas
     assertTrue(-1 != instream.read());
     assertEquals(40000, instream.getPos());
 
-    byte[] readBuffer = new byte[256];
-    instream.read(128, readBuffer, 0, readBuffer.length);
+    int v = 256;
+    byte[] readBuffer = new byte[v];
+    assertEquals(v, instream.read(128, readBuffer, 0, v));
     //have gone back
     assertEquals(40000, instream.getPos());
     //content is the same too
@@ -317,12 +334,11 @@ public abstract class AbstractContractSeekTest extends AbstractFSContractTestBas
     Path randomSeekFile = path("testrandomseeks.bin");
     createFile(getFileSystem(), randomSeekFile, false, buf);
     Random r = new Random();
-    FSDataInputStream stm = getFileSystem().open(randomSeekFile);
 
     // Record the sequence of seeks and reads which trigger a failure.
     int[] seeks = new int[10];
     int[] reads = new int[10];
-    try {
+    try (FSDataInputStream stm = getFileSystem().open(randomSeekFile)) {
       for (int i = 0; i < limit; i++) {
         int seekOff = r.nextInt(buf.length);
         int toRead = r.nextInt(Math.min(buf.length - seekOff, 32000));
@@ -336,13 +352,232 @@ public abstract class AbstractContractSeekTest extends AbstractFSContractTestBas
       sb.append("Sequence of actions:\n");
       for (int j = 0; j < seeks.length; j++) {
         sb.append("seek @ ").append(seeks[j]).append("  ")
-          .append("read ").append(reads[j]).append("\n");
+            .append("read ").append(reads[j]).append("\n");
       }
       LOG.error(sb.toString());
       throw afe;
-    } finally {
-      stm.close();
     }
   }
 
+  @Test
+  public void testReadFullyZeroByteFile() throws Throwable {
+    describe("readFully against a 0 byte file");
+    assumeSupportsPositionedReadable();
+    instream = getFileSystem().open(zeroByteFile);
+    assertEquals(0, instream.getPos());
+    byte[] buffer = new byte[1];
+    instream.readFully(0, buffer, 0, 0);
+    assertEquals(0, instream.getPos());
+    // seek to 0 read 0 bytes from it
+    instream.seek(0);
+    assertEquals(0, instream.read(buffer, 0, 0));
+  }
+
+  @Test
+  public void testReadFullyPastEOFZeroByteFile() throws Throwable {
+    assumeSupportsPositionedReadable();
+    describe("readFully past the EOF of a 0 byte file");
+    instream = getFileSystem().open(zeroByteFile);
+    byte[] buffer = new byte[1];
+    // try to read past end of file
+    try {
+      instream.readFully(0, buffer, 0, 16);
+      fail("Expected an exception");
+    } catch (IllegalArgumentException | IndexOutOfBoundsException
+        | EOFException e) {
+      // expected
+    }
+  }
+
+  @Test
+  public void testReadFullySmallFile() throws Throwable {
+    describe("readFully operations");
+    assumeSupportsPositionedReadable();
+    instream = getFileSystem().open(smallSeekFile);
+    byte[] buffer = new byte[256];
+    // expect negative length to fail
+    try {
+      instream.readFully(0, buffer, 0, -16);
+      fail("Expected an exception");
+    } catch (IllegalArgumentException | IndexOutOfBoundsException e) {
+      // expected
+    }
+    // negative offset into buffer
+    try {
+      instream.readFully(0, buffer, -1, 16);
+      fail("Expected an exception");
+    } catch (IllegalArgumentException | IndexOutOfBoundsException e) {
+      // expected
+    }
+    // expect negative position to fail, ideally with EOF
+    try {
+      instream.readFully(-1, buffer);
+      fail("Expected an exception");
+    } catch (EOFException e) {
+      handleExpectedException(e);
+    } catch (IOException |IllegalArgumentException | IndexOutOfBoundsException e) {
+      handleRelaxedException("readFully with a negative position ",
+          "EOFException",
+          e);
+    }
+
+    // read more than the offset allows
+    try {
+      instream.readFully(0, buffer, buffer.length - 8, 16);
+      fail("Expected an exception");
+    } catch (IllegalArgumentException | IndexOutOfBoundsException e) {
+      // expected
+    }
+
+    // read properly
+    assertEquals(0, instream.getPos());
+    instream.readFully(0, buffer);
+    assertEquals(0, instream.getPos());
+
+    // now read the entire file in one go
+    byte[] fullFile = new byte[TEST_FILE_LEN];
+    instream.readFully(0, fullFile);
+    assertEquals(0, instream.getPos());
+
+    try {
+      instream.readFully(16, fullFile);
+      fail("Expected an exception");
+    } catch (EOFException e) {
+      handleExpectedException(e);
+    } catch (IOException e) {
+      handleRelaxedException("readFully which reads past EOF ",
+          "EOFException",
+          e);
+    }
+  }
+
+  @Test
+  public void testReadFullyPastEOF() throws Throwable {
+    describe("readFully past the EOF of a file");
+    assumeSupportsPositionedReadable();
+    instream = getFileSystem().open(smallSeekFile);
+    byte[] buffer = new byte[256];
+
+    // now read past the end of the file
+    try {
+      instream.readFully(TEST_FILE_LEN + 1, buffer);
+      fail("Expected an exception");
+    } catch (EOFException e) {
+      handleExpectedException(e);
+    } catch (IOException e) {
+      handleRelaxedException("readFully with an offset past EOF ",
+          "EOFException",
+          e);
+    }
+    // read zero bytes from an offset past EOF.
+    try {
+      instream.readFully(TEST_FILE_LEN + 1, buffer, 0, 0);
+      // a zero byte read may fail-fast
+      LOG.info("Filesystem short-circuits 0-byte reads");
+    } catch (EOFException e) {
+      handleExpectedException(e);
+    } catch (IOException e) {
+      handleRelaxedException("readFully(0 bytes) with an offset past EOF ",
+          "EOFException",
+          e);
+    }
+  }
+
+  @Test
+  public void testReadFullyZeroBytebufferPastEOF() throws Throwable {
+    describe("readFully zero bytes from an offset past EOF");
+    assumeSupportsPositionedReadable();
+    instream = getFileSystem().open(smallSeekFile);
+    byte[] buffer = new byte[256];
+    try {
+      instream.readFully(TEST_FILE_LEN + 1, buffer, 0, 0);
+      // a zero byte read may fail-fast
+      LOG.info("Filesystem short-circuits 0-byte reads");
+    } catch (EOFException e) {
+      handleExpectedException(e);
+    } catch (IOException e) {
+      handleRelaxedException("readFully(0 bytes) with an offset past EOF ",
+          "EOFException",
+          e);
+    }
+  }
+
+  @Test
+  public void testReadNullBuffer() throws Throwable {
+    describe("try to read a null buffer ");
+    assumeSupportsPositionedReadable();
+    try (FSDataInputStream in = getFileSystem().open(smallSeekFile)) {
+      // Null buffer
+      int r = in.read(0, null, 0, 16);
+      fail("Expected an exception from a read into a null buffer, got " + r);
+    } catch (IllegalArgumentException e) {
+      // expected
+    }
+  }
+
+  @Test
+  public void testReadSmallFile() throws Throwable {
+    describe("PositionedRead.read operations");
+    assumeSupportsPositionedReadable();
+    instream = getFileSystem().open(smallSeekFile);
+    byte[] buffer = new byte[256];
+    int r;
+    // expect negative length to fail
+    try {
+      r = instream.read(0, buffer, 0, -16);
+      fail("Expected an exception, got " + r);
+    } catch (IllegalArgumentException | IndexOutOfBoundsException e) {
+      // expected
+    }
+    // negative offset into buffer
+    try {
+      r = instream.read(0, buffer, -1, 16);
+      fail("Expected an exception, got " + r);
+    } catch (IllegalArgumentException | IndexOutOfBoundsException e) {
+      // expected
+    }
+    // negative position
+    try {
+      r = instream.read(-1, buffer, 0, 16);
+      fail("Expected an exception, got " + r);
+    } catch (EOFException e) {
+      handleExpectedException(e);
+    } catch (IOException | IllegalArgumentException | IndexOutOfBoundsException e) {
+      handleRelaxedException("read() with a negative position ",
+          "EOFException",
+          e);
+    }
+
+    // read more than the offset allows
+    try {
+      r = instream.read(0, buffer, buffer.length - 8, 16);
+      fail("Expected an exception, got " + r);
+    } catch (IllegalArgumentException | IndexOutOfBoundsException e) {
+      // expected
+    }
+
+    // read properly
+    assertEquals(0, instream.getPos());
+    instream.readFully(0, buffer);
+    assertEquals(0, instream.getPos());
+
+    // now read the entire file in one go
+    byte[] fullFile = new byte[TEST_FILE_LEN];
+    assertEquals(TEST_FILE_LEN,
+        instream.read(0, fullFile, 0, fullFile.length));
+    assertEquals(0, instream.getPos());
+
+    // now read past the end of the file
+    assertEquals(-1,
+        instream.read(TEST_FILE_LEN + 16, buffer, 0, 1));
+  }
+
+  @Test
+  public void testReadAtExactEOF() throws Throwable {
+    describe("read at the end of the file");
+    instream = getFileSystem().open(smallSeekFile);
+    instream.seek(TEST_FILE_LEN -1);
+    assertTrue("read at last byte", instream.read() > 0);
+    assertEquals("read just past EOF", -1, instream.read());
+  }
 }

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/AbstractFSContractTestBase.java

@@ -57,7 +57,7 @@ public abstract class AbstractFSContractTestBase extends Assert
   public static final int DEFAULT_TEST_TIMEOUT = 180 * 1000;
 
   /**
-   * The FS contract used for these tets
+   * The FS contract used for these tests
    */
   private AbstractFSContract contract;
 

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/ContractOptions.java

@@ -53,20 +53,20 @@ public interface ContractOptions {
   /**
    * Flag to indicate that the FS can rename into directories that
    * don't exist, creating them as needed.
-   * @{value}
+   * {@value}
    */
   String RENAME_CREATES_DEST_DIRS = "rename-creates-dest-dirs";
 
   /**
    * Flag to indicate that the FS does not follow the rename contract -and
    * instead only returns false on a failure.
-   * @{value}
+   * {@value}
    */
   String RENAME_OVERWRITES_DEST = "rename-overwrites-dest";
 
   /**
    * Flag to indicate that the FS returns false if the destination exists
-   * @{value}
+   * {@value}
    */
   String RENAME_RETURNS_FALSE_IF_DEST_EXISTS =
       "rename-returns-false-if-dest-exists";
@@ -74,7 +74,7 @@ public interface ContractOptions {
   /**
    * Flag to indicate that the FS returns false on a rename
    * if the source is missing
-   * @{value}
+   * {@value}
    */
   String RENAME_RETURNS_FALSE_IF_SOURCE_MISSING =
       "rename-returns-false-if-source-missing";
@@ -82,74 +82,74 @@ public interface ContractOptions {
   /**
    * Flag to indicate that the FS remove dest first if it is an empty directory
    * mean the FS honors POSIX rename behavior.
-   * @{value}
+   * {@value}
    */
   String RENAME_REMOVE_DEST_IF_EMPTY_DIR = "rename-remove-dest-if-empty-dir";
 
   /**
    * Flag to indicate that append is supported
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_APPEND = "supports-append";
 
   /**
    * Flag to indicate that setTimes is supported.
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_SETTIMES = "supports-settimes";
 
   /**
    * Flag to indicate that getFileStatus is supported.
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_GETFILESTATUS = "supports-getfilestatus";
 
   /**
    * Flag to indicate that renames are atomic
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_ATOMIC_RENAME = "supports-atomic-rename";
 
   /**
    * Flag to indicate that directory deletes are atomic
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_ATOMIC_DIRECTORY_DELETE = "supports-atomic-directory-delete";
 
   /**
    * Does the FS support multiple block locations?
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_BLOCK_LOCALITY = "supports-block-locality";
 
   /**
    * Does the FS support the concat() operation?
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_CONCAT = "supports-concat";
 
   /**
    * Is seeking supported at all?
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_SEEK = "supports-seek";
 
   /**
    * Is seeking past the EOF allowed?
-   * @{value}
+   * {@value}
    */
   String REJECTS_SEEK_PAST_EOF = "rejects-seek-past-eof";
 
   /**
    * Is seeking on a closed file supported? Some filesystems only raise an
    * exception later, when trying to read.
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_SEEK_ON_CLOSED_FILE = "supports-seek-on-closed-file";
 
   /**
    * Is available() on a closed InputStream supported?
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_AVAILABLE_ON_CLOSED_FILE = "supports-available-on-closed-file";
 
@@ -157,32 +157,39 @@ public interface ContractOptions {
    * Flag to indicate that this FS expects to throw the strictest
    * exceptions it can, not generic IOEs, which, if returned,
    * must be rejected.
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_STRICT_EXCEPTIONS = "supports-strict-exceptions";
 
   /**
    * Are unix permissions
-   * @{value}
+   * {@value}
    */
   String SUPPORTS_UNIX_PERMISSIONS = "supports-unix-permissions";
 
+  /**
+   * Is positioned readable supported? Supporting seek should be sufficient
+   * for this.
+   * {@value}
+   */
+  String SUPPORTS_POSITIONED_READABLE = "supports-positioned-readable";
+
   /**
    * Maximum path length
-   * @{value}
+   * {@value}
    */
   String MAX_PATH_ = "max-path";
 
   /**
    * Maximum filesize: 0 or -1 for no limit
-   * @{value}
+   * {@value}
    */
   String MAX_FILESIZE = "max-filesize";
 
   /**
    * Flag to indicate that tests on the root directories of a filesystem/
    * object store are permitted
-   * @{value}
+   * {@value}
    */
   String TEST_ROOT_TESTS_ENABLED = "test.root-tests-enabled";
 

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/contract/ContractTestUtils.java

@@ -23,6 +23,7 @@ import org.apache.hadoop.fs.FSDataOutputStream;
 import org.apache.hadoop.fs.FileStatus;
 import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.io.IOUtils;
 import org.junit.Assert;
 import org.junit.internal.AssumptionViolatedException;
 import org.slf4j.Logger;
@@ -432,9 +433,7 @@ public class ContractTestUtils extends Assert {
    * @throws AssertionError with the text and throwable -always
    */
   public static void fail(String text, Throwable thrown) {
-    AssertionError e = new AssertionError(text);
-    e.initCause(thrown);
-    throw e;
+    throw new AssertionError(text, thrown);
   }
 
   /**
@@ -509,10 +508,14 @@ public class ContractTestUtils extends Assert {
                                  boolean overwrite,
                                  byte[] data) throws IOException {
     FSDataOutputStream stream = fs.create(path, overwrite);
-    if (data != null && data.length > 0) {
-      stream.write(data);
+    try {
+      if (data != null && data.length > 0) {
+        stream.write(data);
+      }
+      stream.close();
+    } finally {
+      IOUtils.closeStream(stream);
     }
-    stream.close();
   }
 
   /**
@@ -574,13 +577,10 @@ public class ContractTestUtils extends Assert {
   public static String readBytesToString(FileSystem fs,
                                   Path path,
                                   int length) throws IOException {
-    FSDataInputStream in = fs.open(path);
-    try {
+    try (FSDataInputStream in = fs.open(path)) {
       byte[] buf = new byte[length];
       in.readFully(0, buf);
       return toChar(buf);
-    } finally {
-      in.close();
     }
   }
 
@@ -786,8 +786,7 @@ public class ContractTestUtils extends Assert {
 
     long totalBytesRead = 0;
     int nextExpectedNumber = 0;
-    final InputStream inputStream = fs.open(path);
-    try {
+    try (InputStream inputStream = fs.open(path)) {
       while (true) {
         final int bytesRead = inputStream.read(testBuffer);
         if (bytesRead < 0) {
@@ -814,8 +813,6 @@ public class ContractTestUtils extends Assert {
         throw new IOException("Expected to read " + expectedSize +
             " bytes but only received " + totalBytesRead);
       }
-    } finally {
-      inputStream.close();
     }
   }
 

hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/http/TestHttpServer.java

@@ -235,6 +235,16 @@ public class TestHttpServer extends HttpServerFunctionalTest {
     assertEquals("text/html; charset=utf-8", conn.getContentType());
   }
 
+  @Test
+  public void testHttpResonseContainsXFrameOptions() throws IOException {
+    URL url = new URL(baseUrl, "");
+    HttpURLConnection conn = (HttpURLConnection) url.openConnection();
+    conn.connect();
+
+    String xfoHeader = conn.getHeaderField("X-FRAME-OPTIONS");
+    assertTrue("X-FRAME-OPTIONS is absent in the header", xfoHeader != null);
+  }
+
   /**
    * Dummy filter that mimics as an authentication filter. Obtains user identity
    * from the request parameter user.name. Wraps around the request so that

hadoop-common-project/hadoop-common/src/test/resources/contract/localfs.xml

@@ -100,7 +100,7 @@ case sensitivity and permission options are determined at run time from OS type
     <value>true</value>
   </property>
 
-  <!-- checksum FS doesn't allow seeing past EOF -->
+  <!-- checksum FS doesn't allow seeking past EOF -->
   <property>
     <name>fs.contract.rejects-seek-past-eof</name>
     <value>true</value>
@@ -121,4 +121,4 @@ case sensitivity and permission options are determined at run time from OS type
     <value>true</value>
   </property>
 
-</configuration>
+</configuration>

hadoop-common-project/hadoop-kms/src/site/markdown/index.md.vm

@@ -284,7 +284,15 @@ The answer to "What is your first and last name?" (i.e. "CN") must be the hostna
 
 NOTE: You need to restart the KMS for the configuration changes to take effect.
 
-$H4 KMS Access Control
+$H4 ACLs (Access Control Lists)
+
+KMS supports ACLs (Access Control Lists) for fine-grained permission control.
+
+Two levels of ACLs exist in KMS: KMS ACLs and Key ACLs. KMS ACLs control access at KMS operation level, and precede Key ACLs. In particular, only if permission is granted at KMS ACLs level, shall the permission check against Key ACLs be performed.
+
+The configuration and usage of KMS ACLs and Key ACLs are described in the sections below.
+
+$H5 KMS ACLs
 
 KMS ACLs configuration are defined in the KMS `etc/hadoop/kms-acls.xml` configuration file. This file is hot-reloaded when it changes.
 
@@ -452,7 +460,7 @@ A user accessing KMS is first checked for inclusion in the Access Control List f
 </configuration>
 ```
 
-$H4 Key Access Control
+$H5 Key ACLs
 
 KMS supports access control for all non-read operations at the Key level. All Key Access operations are classified as :
 
@@ -466,9 +474,9 @@ These can be defined in the KMS `etc/hadoop/kms-acls.xml` as follows
 
 For all keys for which a key access has not been explicitly configured, It is possible to configure a default key access control for a subset of the operation types.
 
-It is also possible to configure a "whitelist" key ACL for a subset of the operation types. The whitelist key ACL is a whitelist in addition to the explicit or default per-key ACL. That is, if no per-key ACL is explicitly set, a user will be granted access if they are present in the default per-key ACL or the whitelist key ACL. If a per-key ACL is explicitly set, a user will be granted access if they are present in the per-key ACL or the whitelist key ACL.
+It is also possible to configure a "whitelist" key ACL for a subset of the operation types. The whitelist key ACL grants access to the key, in addition to the explicit or default per-key ACL. That is, if no per-key ACL is explicitly set, a user will be granted access if they are present in the default per-key ACL or the whitelist key ACL. If a per-key ACL is explicitly set, a user will be granted access if they are present in the per-key ACL or the whitelist key ACL.
 
-If no ACL is configured for a specific key AND no default ACL is configured AND no root key ACL is configured for the requested operation, then access will be DENIED.
+If no ACL is configured for a specific key AND no default ACL is configured AND no whitelist key ACL is configured for the requested operation, then access will be DENIED.
 
 **NOTE:** The default and whitelist key ACL does not support `ALL` operation qualifier.
 
@@ -575,7 +583,11 @@ If no ACL is configured for a specific key AND no default ACL is configured AND
 
 $H3 KMS Delegation Token Configuration
 
-KMS delegation token secret manager can be configured with the following properties:
+KMS supports delegation tokens to authenticate to the key providers from processes without Kerberos credentials.
+
+KMS delegation token authentication extends the default Hadoop authentication. See [Hadoop Auth](../hadoop-auth/index.html) page for more details.
+
+Additionally, KMS delegation token secret manager can be configured with the following properties:
 
 ```xml
   <property>
@@ -590,7 +602,7 @@ KMS delegation token secret manager can be configured with the following propert
     <name>hadoop.kms.authentication.delegation-token.max-lifetime.sec</name>
     <value>604800</value>
     <description>
-      Maximum lifetime of a delagation token, in seconds. Default value 7 days.
+      Maximum lifetime of a delegation token, in seconds. Default value 7 days.
     </description>
   </property>
 
@@ -598,7 +610,7 @@ KMS delegation token secret manager can be configured with the following propert
     <name>hadoop.kms.authentication.delegation-token.renew-interval.sec</name>
     <value>86400</value>
     <description>
-      Renewal interval of a delagation token, in seconds. Default value 1 day.
+      Renewal interval of a delegation token, in seconds. Default value 1 day.
     </description>
   </property>
 
@@ -640,7 +652,7 @@ $H4 HTTP Authentication Signature
 
 KMS uses Hadoop Authentication for HTTP authentication. Hadoop Authentication issues a signed HTTP Cookie once the client has authenticated successfully. This HTTP Cookie has an expiration time, after which it will trigger a new authentication sequence. This is done to avoid triggering the authentication on every HTTP request of a client.
 
-A KMS instance must verify the HTTP Cookie signatures signed by other KMS instances. To do this all KMS instances must share the signing secret.
+A KMS instance must verify the HTTP Cookie signatures signed by other KMS instances. To do this, all KMS instances must share the signing secret. Please see [SignerSecretProvider Configuration](../hadoop-auth/Configuration.html#SignerSecretProvider_Configuration) for detailed description and configuration examples. Note that KMS configurations need to be prefixed with `hadoop.kms.authentication`, as shown in the example below.
 
 This secret sharing can be done using a Zookeeper service which is configured in KMS with the following properties in the `kms-site.xml`:
 
@@ -650,8 +662,9 @@ This secret sharing can be done using a Zookeeper service which is configured in
     <value>zookeeper</value>
     <description>
       Indicates how the secret to sign the authentication cookies will be
-      stored. Options are 'random' (default), 'string' and 'zookeeper'.
+      stored. Options are 'random' (default), 'file' and 'zookeeper'.
       If using a setup with multiple KMS instances, 'zookeeper' should be used.
+      If using file, signature.secret.file should be configured and point to the secret file.
     </description>
   </property>
   <property>
@@ -659,7 +672,7 @@ This secret sharing can be done using a Zookeeper service which is configured in
     <value>/hadoop-kms/hadoop-auth-signature-secret</value>
     <description>
       The Zookeeper ZNode path where the KMS instances will store and retrieve
-      the secret from.
+      the secret from. All KMS instances that need to coordinate should point to the same path.
     </description>
   </property>
   <property>
@@ -696,7 +709,11 @@ This secret sharing can be done using a Zookeeper service which is configured in
 
 $H4 Delegation Tokens
 
-TBD
+Similar to HTTP authentication, KMS uses Hadoop Authentication for delegation tokens too.
+
+Under HA, A KMS instance must verify the delegation token given by another KMS instance, by checking the shared secret used to sign the delegation token. To do this, all KMS instances must be able to retrieve the shared secret from ZooKeeper.
+
+Please see the examples given in HTTP Authentication section to configure ZooKeeper for secret sharing.
 
 $H3 KMS HTTP REST API
 

hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/DFSInputStream.java

@@ -972,6 +972,10 @@ public class DFSInputStream extends FSInputStream
   @Override
   public synchronized int read(@Nonnull final byte buf[], int off, int len)
       throws IOException {
+    validatePositionedReadArgs(pos, buf, off, len);
+    if (len == 0) {
+      return 0;
+    }
     ReaderStrategy byteArrayReader = new ByteArrayStrategy(buf);
     try (TraceScope scope =
              dfsClient.newReaderTraceScope("DFSInputStream#byteArrayRead",
@@ -1423,6 +1427,10 @@ public class DFSInputStream extends FSInputStream
   @Override
   public int read(long position, byte[] buffer, int offset, int length)
       throws IOException {
+    validatePositionedReadArgs(position, buffer, offset, length);
+    if (length == 0) {
+      return 0;
+    }
     try (TraceScope scope = dfsClient.
         newReaderTraceScope("DFSInputStream#byteArrayPread",
             src, position, length)) {

hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/HdfsConfiguration.java

@@ -139,9 +139,6 @@ public class HdfsConfiguration extends Configuration {
             HdfsClientConfigKeys.DFS_NAMESERVICES),
         new DeprecationDelta("dfs.federation.nameservice.id",
             DeprecatedKeys.DFS_NAMESERVICE_ID),
-        new DeprecationDelta("dfs.client.file-block-storage-locations.timeout",
-            HdfsClientConfigKeys.
-                DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_MS),
     });
   }
 

hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/client/HdfsClientConfigKeys.java

@@ -97,12 +97,6 @@ public interface HdfsClientConfigKeys {
   int     DFS_CLIENT_CACHED_CONN_RETRY_DEFAULT = 3;
   String  DFS_CLIENT_CONTEXT = "dfs.client.context";
   String  DFS_CLIENT_CONTEXT_DEFAULT = "default";
-  String  DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS =
-      "dfs.client.file-block-storage-locations.num-threads";
-  int     DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS_DEFAULT = 10;
-  String  DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_MS =
-      "dfs.client.file-block-storage-locations.timeout.millis";
-  int     DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_MS_DEFAULT = 1000;
   String  DFS_CLIENT_USE_LEGACY_BLOCKREADER =
       "dfs.client.use.legacy.blockreader";
   boolean DFS_CLIENT_USE_LEGACY_BLOCKREADER_DEFAULT = false;

hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/web/ByteRangeInputStream.java

@@ -28,6 +28,7 @@ import java.util.Map;
 import java.util.StringTokenizer;
 
 import org.apache.commons.io.input.BoundedInputStream;
+import org.apache.hadoop.fs.FSExceptionMessages;
 import org.apache.hadoop.fs.FSInputStream;
 
 import com.google.common.annotations.VisibleForTesting;
@@ -128,6 +129,9 @@ public abstract class ByteRangeInputStream extends FSInputStream {
   @VisibleForTesting
   protected InputStreamAndFileLength openInputStream(long startOffset)
       throws IOException {
+    if (startOffset < 0) {
+      throw new EOFException("Negative Position");
+    }
     // Use the original url if no resolved url exists, eg. if
     // it's the first time a request is made.
     final boolean resolved = resolvedURL.getURL() != null;
@@ -224,6 +228,10 @@ public abstract class ByteRangeInputStream extends FSInputStream {
   @Override
   public int read(long position, byte[] buffer, int offset, int length)
       throws IOException {
+    validatePositionedReadArgs(position, buffer, offset, length);
+    if (length == 0) {
+      return 0;
+    }
     try (InputStream in = openInputStream(position).in) {
       return in.read(buffer, offset, length);
     }
@@ -232,17 +240,21 @@ public abstract class ByteRangeInputStream extends FSInputStream {
   @Override
   public void readFully(long position, byte[] buffer, int offset, int length)
       throws IOException {
-    final InputStreamAndFileLength fin = openInputStream(position);
-    if (fin.length != null && length + position > fin.length) {
-      throw new EOFException("The length to read " + length
-          + " exceeds the file length " + fin.length);
+    validatePositionedReadArgs(position, buffer, offset, length);
+    if (length == 0) {
+      return;
     }
+    final InputStreamAndFileLength fin = openInputStream(position);
     try {
+      if (fin.length != null && length + position > fin.length) {
+        throw new EOFException("The length to read " + length
+            + " exceeds the file length " + fin.length);
+      }
       int nread = 0;
       while (nread < length) {
         int nbytes = fin.in.read(buffer, offset + nread, length - nread);
         if (nbytes < 0) {
-          throw new EOFException("End of file reached before reading fully.");
+          throw new EOFException(FSExceptionMessages.EOF_IN_READ_FULLY);
         }
         nread += nbytes;
       }

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSConfigKeys.java

@@ -1197,24 +1197,6 @@ public class DFSConfigKeys extends CommonConfigurationKeys {
   @Deprecated
   public static final String  DFS_CLIENT_CONTEXT_DEFAULT =
       HdfsClientConfigKeys.DFS_CLIENT_CONTEXT_DEFAULT;
-  @Deprecated
-  public static final String
-      DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS =
-      HdfsClientConfigKeys.DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS;
-  @Deprecated
-  public static final int
-      DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS_DEFAULT =
-      HdfsClientConfigKeys
-          .DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS_DEFAULT;
-  @Deprecated
-  public static final String
-      DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_MS =
-      HdfsClientConfigKeys.DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_MS;
-  @Deprecated
-  public static final int
-      DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_MS_DEFAULT =
-      HdfsClientConfigKeys
-          .DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_MS_DEFAULT;
 
   @Deprecated
   public static final String  DFS_CLIENT_DATANODE_RESTART_TIMEOUT_KEY =

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java

@@ -35,6 +35,7 @@ import org.apache.hadoop.hdfs.DFSUtil;
 import org.apache.hadoop.hdfs.HdfsConfiguration;
 import org.apache.hadoop.hdfs.protocol.*;
 import org.apache.hadoop.hdfs.protocol.HdfsConstants.DatanodeReportType;
+import org.apache.hadoop.hdfs.security.token.block.BlockTokenIdentifier;
 import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeDescriptor.BlockTargetPair;
 import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeDescriptor.CachedBlocksList;
 import org.apache.hadoop.hdfs.server.namenode.CachedBlock;
@@ -47,6 +48,7 @@ import org.apache.hadoop.hdfs.server.protocol.BlockRecoveryCommand.RecoveringStr
 import org.apache.hadoop.ipc.Server;
 import org.apache.hadoop.net.*;
 import org.apache.hadoop.net.NetworkTopology.InvalidTopologyException;
+import org.apache.hadoop.security.token.Token;
 import org.apache.hadoop.util.ReflectionUtils;
 
 import java.io.IOException;
@@ -368,49 +370,110 @@ public class DatanodeManager {
 
   }
   
-  /** Sort the located blocks by the distance to the target host. */
-  public void sortLocatedBlocks(final String targethost,
-      final List<LocatedBlock> locatedblocks) {
-    //sort the blocks
+  /**
+   * Sort the non-striped located blocks by the distance to the target host.
+   *
+   * For striped blocks, it will only move decommissioned/stale nodes to the
+   * bottom. For example, assume we have storage list:
+   * d0, d1, d2, d3, d4, d5, d6, d7, d8, d9
+   * mapping to block indices:
+   * 0, 1, 2, 3, 4, 5, 6, 7, 8, 2
+   *
+   * Here the internal block b2 is duplicated, locating in d2 and d9. If d2 is
+   * a decommissioning node then should switch d2 and d9 in the storage list.
+   * After sorting locations, will update corresponding block indices
+   * and block tokens.
+   */
+  public void sortLocatedBlocks(final String targetHost,
+      final List<LocatedBlock> locatedBlocks) {
+    Comparator<DatanodeInfo> comparator = avoidStaleDataNodesForRead ?
+        new DFSUtil.DecomStaleComparator(staleInterval) :
+        DFSUtil.DECOM_COMPARATOR;
+    // sort located block
+    for (LocatedBlock lb : locatedBlocks) {
+      if (lb.isStriped()) {
+        sortLocatedStripedBlock(lb, comparator);
+      } else {
+        sortLocatedBlock(lb, targetHost, comparator);
+      }
+    }
+  }
+
+  /**
+   * Move decommissioned/stale datanodes to the bottom. After sorting it will
+   * update block indices and block tokens respectively.
+   *
+   * @param lb located striped block
+   * @param comparator dn comparator
+   */
+  private void sortLocatedStripedBlock(final LocatedBlock lb,
+      Comparator<DatanodeInfo> comparator) {
+    DatanodeInfo[] di = lb.getLocations();
+    HashMap<DatanodeInfo, Byte> locToIndex = new HashMap<>();
+    HashMap<DatanodeInfo, Token<BlockTokenIdentifier>> locToToken =
+        new HashMap<>();
+    LocatedStripedBlock lsb = (LocatedStripedBlock) lb;
+    for (int i = 0; i < di.length; i++) {
+      locToIndex.put(di[i], lsb.getBlockIndices()[i]);
+      locToToken.put(di[i], lsb.getBlockTokens()[i]);
+    }
+    // Move decommissioned/stale datanodes to the bottom
+    Arrays.sort(di, comparator);
+
+    // must update cache since we modified locations array
+    lb.updateCachedStorageInfo();
+
+    // must update block indices and block tokens respectively
+    for (int i = 0; i < di.length; i++) {
+      lsb.getBlockIndices()[i] = locToIndex.get(di[i]);
+      lsb.getBlockTokens()[i] = locToToken.get(di[i]);
+    }
+  }
+
+  /**
+   * Move decommissioned/stale datanodes to the bottom. Also, sort nodes by
+   * network distance.
+   *
+   * @param lb located block
+   * @param targetHost target host
+   * @param comparator dn comparator
+   */
+  private void sortLocatedBlock(final LocatedBlock lb, String targetHost,
+      Comparator<DatanodeInfo> comparator) {
     // As it is possible for the separation of node manager and datanode, 
     // here we should get node but not datanode only .
-    Node client = getDatanodeByHost(targethost);
+    Node client = getDatanodeByHost(targetHost);
     if (client == null) {
       List<String> hosts = new ArrayList<> (1);
-      hosts.add(targethost);
+      hosts.add(targetHost);
       List<String> resolvedHosts = dnsToSwitchMapping.resolve(hosts);
       if (resolvedHosts != null && !resolvedHosts.isEmpty()) {
         String rName = resolvedHosts.get(0);
         if (rName != null) {
           client = new NodeBase(rName + NodeBase.PATH_SEPARATOR_STR +
-            targethost);
+            targetHost);
         }
       } else {
         LOG.error("Node Resolution failed. Please make sure that rack " +
           "awareness scripts are functional.");
       }
     }
-    
-    Comparator<DatanodeInfo> comparator = avoidStaleDataNodesForRead ?
-        new DFSUtil.DecomStaleComparator(staleInterval) : 
-        DFSUtil.DECOM_COMPARATOR;
-        
-    for (LocatedBlock b : locatedblocks) {
-      DatanodeInfo[] di = b.getLocations();
-      // Move decommissioned/stale datanodes to the bottom
-      Arrays.sort(di, comparator);
-      
-      int lastActiveIndex = di.length - 1;
-      while (lastActiveIndex > 0 && isInactive(di[lastActiveIndex])) {
-          --lastActiveIndex;
-      }
-      int activeLen = lastActiveIndex + 1;      
-      networktopology.sortByDistance(client, b.getLocations(), activeLen);
-      // must update cache since we modified locations array
-      b.updateCachedStorageInfo();
+
+    DatanodeInfo[] di = lb.getLocations();
+    // Move decommissioned/stale datanodes to the bottom
+    Arrays.sort(di, comparator);
+
+    // Sort nodes by network distance only for located blocks
+    int lastActiveIndex = di.length - 1;
+    while (lastActiveIndex > 0 && isInactive(di[lastActiveIndex])) {
+      --lastActiveIndex;
     }
+    int activeLen = lastActiveIndex + 1;
+    networktopology.sortByDistance(client, lb.getLocations(), activeLen);
+
+    // must update cache since we modified locations array
+    lb.updateCachedStorageInfo();
   }
-  
 
   /** @return the datanode descriptor for the host. */
   public DatanodeDescriptor getDatanodeByHost(final String host) {

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/DNConf.java

@@ -114,6 +114,9 @@ public class DNConf {
   // Allow LAZY_PERSIST writes from non-local clients?
   private final boolean allowNonLocalLazyPersist;
 
+  private final int volFailuresTolerated;
+  private final int volsConfigured;
+
   public DNConf(Configuration conf) {
     this.conf = conf;
     socketTimeout = conf.getInt(DFS_CLIENT_SOCKET_TIMEOUT_KEY,
@@ -238,6 +241,13 @@ public class DNConf {
     this.bpReadyTimeout = conf.getLong(
         DFS_DATANODE_BP_READY_TIMEOUT_KEY,
         DFS_DATANODE_BP_READY_TIMEOUT_DEFAULT);
+
+    this.volFailuresTolerated =
+        conf.getInt(DFSConfigKeys.DFS_DATANODE_FAILED_VOLUMES_TOLERATED_KEY,
+            DFSConfigKeys.DFS_DATANODE_FAILED_VOLUMES_TOLERATED_DEFAULT);
+    String[] dataDirs =
+        conf.getTrimmedStrings(DFSConfigKeys.DFS_DATANODE_DATA_DIR_KEY);
+    this.volsConfigured = (dataDirs == null) ? 0 : dataDirs.length;
   }
 
   // We get minimumNameNodeVersion via a method so it can be mocked out in tests.
@@ -363,4 +373,12 @@ public class DNConf {
   public long getLifelineIntervalMs() {
     return lifelineIntervalMs;
   }
+
+  public int getVolFailuresTolerated() {
+    return volFailuresTolerated;
+  }
+
+  public int getVolsConfigured() {
+    return volsConfigured;
+  }
 }

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/DataNode.java

@@ -1280,6 +1280,15 @@ public class DataNode extends ReconfigurableBase
     LOG.info("Starting DataNode with maxLockedMemory = " +
         dnConf.maxLockedMemory);
 
+    int volFailuresTolerated = dnConf.getVolFailuresTolerated();
+    int volsConfigured = dnConf.getVolsConfigured();
+    if (volFailuresTolerated < 0 || volFailuresTolerated >= volsConfigured) {
+      throw new DiskErrorException("Invalid value configured for "
+          + "dfs.datanode.failed.volumes.tolerated - " + volFailuresTolerated
+          + ". Value configured is either less than 0 or >= "
+          + "to the number of configured volumes (" + volsConfigured + ").");
+    }
+
     storage = new DataStorage();
     
     // global DN settings

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/BlockPoolSlice.java

@@ -36,8 +36,9 @@ import org.apache.commons.io.FileUtils;
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
 import org.apache.hadoop.conf.Configuration;
-import org.apache.hadoop.fs.DU;
+import org.apache.hadoop.fs.CachingGetSpaceUsed;
 import org.apache.hadoop.fs.FileUtil;
+import org.apache.hadoop.fs.GetSpaceUsed;
 import org.apache.hadoop.hdfs.DFSConfigKeys;
 import org.apache.hadoop.hdfs.DFSUtilClient;
 import org.apache.hadoop.hdfs.protocol.Block;
@@ -62,10 +63,10 @@ import com.google.common.annotations.VisibleForTesting;
 import com.google.common.io.Files;
 
 /**
- * A block pool slice represents a portion of a block pool stored on a volume.  
- * Taken together, all BlockPoolSlices sharing a block pool ID across a 
+ * A block pool slice represents a portion of a block pool stored on a volume.
+ * Taken together, all BlockPoolSlices sharing a block pool ID across a
  * cluster represent a single block pool.
- * 
+ *
  * This class is synchronized by {@link FsVolumeImpl}.
  */
 class BlockPoolSlice {
@@ -92,10 +93,10 @@ class BlockPoolSlice {
   private final Timer timer;
 
   // TODO:FEDERATION scalability issue - a thread per DU is needed
-  private final DU dfsUsage;
+  private final GetSpaceUsed dfsUsage;
 
   /**
-   * Create a blook pool slice 
+   * Create a blook pool slice
    * @param bpid Block pool Id
    * @param volume {@link FsVolumeImpl} to which this BlockPool belongs to
    * @param bpDir directory corresponding to the BlockPool
@@ -107,7 +108,7 @@ class BlockPoolSlice {
       Configuration conf, Timer timer) throws IOException {
     this.bpid = bpid;
     this.volume = volume;
-    this.currentDir = new File(bpDir, DataStorage.STORAGE_DIR_CURRENT); 
+    this.currentDir = new File(bpDir, DataStorage.STORAGE_DIR_CURRENT);
     this.finalizedDir = new File(
         currentDir, DataStorage.STORAGE_DIR_FINALIZED);
     this.lazypersistDir = new File(currentDir, DataStorage.STORAGE_DIR_LAZY_PERSIST);
@@ -151,8 +152,10 @@ class BlockPoolSlice {
     }
     // Use cached value initially if available. Or the following call will
     // block until the initial du command completes.
-    this.dfsUsage = new DU(bpDir, conf, loadDfsUsed());
-    this.dfsUsage.start();
+    this.dfsUsage = new CachingGetSpaceUsed.Builder().setPath(bpDir)
+                                                     .setConf(conf)
+                                                     .setInitialUsed(loadDfsUsed())
+                                                     .build();
 
     // Make the dfs usage to be saved during shutdown.
     ShutdownHookManager.get().addShutdownHook(
@@ -173,7 +176,7 @@ class BlockPoolSlice {
   File getFinalizedDir() {
     return finalizedDir;
   }
-  
+
   File getLazypersistDir() {
     return lazypersistDir;
   }
@@ -188,17 +191,21 @@ class BlockPoolSlice {
 
   /** Run DU on local drives.  It must be synchronized from caller. */
   void decDfsUsed(long value) {
-    dfsUsage.decDfsUsed(value);
+    if (dfsUsage instanceof CachingGetSpaceUsed) {
+      ((CachingGetSpaceUsed)dfsUsage).incDfsUsed(-value);
+    }
   }
-  
+
   long getDfsUsed() throws IOException {
     return dfsUsage.getUsed();
   }
 
   void incDfsUsed(long value) {
-    dfsUsage.incDfsUsed(value);
+    if (dfsUsage instanceof CachingGetSpaceUsed) {
+      ((CachingGetSpaceUsed)dfsUsage).incDfsUsed(value);
+    }
   }
-  
+
   /**
    * Read in the cached DU value and return it if it is less than
    * cachedDfsUsedCheckTime which is set by
@@ -304,7 +311,10 @@ class BlockPoolSlice {
     }
     File blockFile = FsDatasetImpl.moveBlockFiles(b, f, blockDir);
     File metaFile = FsDatasetUtil.getMetaFile(blockFile, b.getGenerationStamp());
-    dfsUsage.incDfsUsed(b.getNumBytes()+metaFile.length());
+    if (dfsUsage instanceof CachingGetSpaceUsed) {
+      ((CachingGetSpaceUsed) dfsUsage).incDfsUsed(
+          b.getNumBytes() + metaFile.length());
+    }
     return blockFile;
   }
 
@@ -331,7 +341,7 @@ class BlockPoolSlice {
   }
 
 
-    
+
   void getVolumeMap(ReplicaMap volumeMap,
                     final RamDiskReplicaTracker lazyWriteReplicaMap)
       throws IOException {
@@ -342,7 +352,7 @@ class BlockPoolSlice {
       FsDatasetImpl.LOG.info(
           "Recovered " + numRecovered + " replicas from " + lazypersistDir);
     }
-    
+
     boolean  success = readReplicasFromCache(volumeMap, lazyWriteReplicaMap);
     if (!success) {
       // add finalized replicas
@@ -436,7 +446,7 @@ class BlockPoolSlice {
     FileUtil.fullyDelete(source);
     return numRecovered;
   }
-  
+
   private void addReplicaToReplicasMap(Block block, ReplicaMap volumeMap,
       final RamDiskReplicaTracker lazyWriteReplicaMap,boolean isFinalized)
       throws IOException {
@@ -444,7 +454,7 @@ class BlockPoolSlice {
     long blockId = block.getBlockId();
     long genStamp = block.getGenerationStamp();
     if (isFinalized) {
-      newReplica = new FinalizedReplica(blockId, 
+      newReplica = new FinalizedReplica(blockId,
           block.getNumBytes(), genStamp, volume, DatanodeUtil
           .idToBlockDir(finalizedDir, blockId));
     } else {
@@ -461,7 +471,7 @@ class BlockPoolSlice {
           // We don't know the expected block length, so just use 0
           // and don't reserve any more space for writes.
           newReplica = new ReplicaBeingWritten(blockId,
-              validateIntegrityAndSetLength(file, genStamp), 
+              validateIntegrityAndSetLength(file, genStamp),
               genStamp, volume, file.getParentFile(), null, 0);
           loadRwr = false;
         }
@@ -507,7 +517,7 @@ class BlockPoolSlice {
       incrNumBlocks();
     }
   }
-  
+
 
   /**
    * Add replicas under the given directory to the volume map
@@ -537,12 +547,12 @@ class BlockPoolSlice {
       }
       if (!Block.isBlockFilename(file))
         continue;
-      
+
       long genStamp = FsDatasetUtil.getGenerationStampFromFile(
           files, file);
       long blockId = Block.filename2id(file.getName());
-      Block block = new Block(blockId, file.length(), genStamp); 
-      addReplicaToReplicasMap(block, volumeMap, lazyWriteReplicaMap, 
+      Block block = new Block(blockId, file.length(), genStamp);
+      addReplicaToReplicasMap(block, volumeMap, lazyWriteReplicaMap,
           isFinalized);
     }
   }
@@ -636,11 +646,11 @@ class BlockPoolSlice {
 
   /**
    * Find out the number of bytes in the block that match its crc.
-   * 
-   * This algorithm assumes that data corruption caused by unexpected 
+   *
+   * This algorithm assumes that data corruption caused by unexpected
    * datanode shutdown occurs only in the last crc chunk. So it checks
    * only the last chunk.
-   * 
+   *
    * @param blockFile the block file
    * @param genStamp generation stamp of the block
    * @return the number of valid bytes
@@ -667,7 +677,7 @@ class BlockPoolSlice {
       int bytesPerChecksum = checksum.getBytesPerChecksum();
       int checksumSize = checksum.getChecksumSize();
       long numChunks = Math.min(
-          (blockFileLen + bytesPerChecksum - 1)/bytesPerChecksum, 
+          (blockFileLen + bytesPerChecksum - 1)/bytesPerChecksum,
           (metaFileLen - crcHeaderLen)/checksumSize);
       if (numChunks == 0) {
         return 0;
@@ -710,17 +720,20 @@ class BlockPoolSlice {
       IOUtils.closeStream(blockIn);
     }
   }
-    
+
   @Override
   public String toString() {
     return currentDir.getAbsolutePath();
   }
-  
+
   void shutdown(BlockListAsLongs blocksListToPersist) {
     saveReplicas(blocksListToPersist);
     saveDfsUsed();
     dfsUsedSaved = true;
-    dfsUsage.shutdown();
+
+    if (dfsUsage instanceof CachingGetSpaceUsed) {
+      IOUtils.cleanup(LOG, ((CachingGetSpaceUsed) dfsUsage));
+    }
   }
 
   private boolean readReplicasFromCache(ReplicaMap volumeMap,
@@ -729,17 +742,17 @@ class BlockPoolSlice {
     File replicaFile = new File(currentDir, REPLICA_CACHE_FILE);
     // Check whether the file exists or not.
     if (!replicaFile.exists()) {
-      LOG.info("Replica Cache file: "+  replicaFile.getPath() + 
+      LOG.info("Replica Cache file: "+  replicaFile.getPath() +
           " doesn't exist ");
       return false;
     }
     long fileLastModifiedTime = replicaFile.lastModified();
     if (System.currentTimeMillis() > fileLastModifiedTime + replicaCacheExpiry) {
-      LOG.info("Replica Cache file: " + replicaFile.getPath() + 
+      LOG.info("Replica Cache file: " + replicaFile.getPath() +
           " has gone stale");
       // Just to make findbugs happy
       if (!replicaFile.delete()) {
-        LOG.info("Replica Cache file: " + replicaFile.getPath() + 
+        LOG.info("Replica Cache file: " + replicaFile.getPath() +
             " cannot be deleted");
       }
       return false;
@@ -776,7 +789,7 @@ class BlockPoolSlice {
         iter.remove();
         volumeMap.add(bpid, info);
       }
-      LOG.info("Successfully read replica from cache file : " 
+      LOG.info("Successfully read replica from cache file : "
           + replicaFile.getPath());
       return true;
     } catch (Exception e) {
@@ -794,10 +807,10 @@ class BlockPoolSlice {
       // close the inputStream
       IOUtils.closeStream(inputStream);
     }
-  } 
-  
+  }
+
   private void saveReplicas(BlockListAsLongs blocksListToPersist) {
-    if (blocksListToPersist == null || 
+    if (blocksListToPersist == null ||
         blocksListToPersist.getNumberOfBlocks()== 0) {
       return;
     }
@@ -813,7 +826,7 @@ class BlockPoolSlice {
           replicaCacheFile.getPath());
       return;
     }
-    
+
     FileOutputStream out = null;
     try {
       out = new FileOutputStream(tmpFile);
@@ -827,7 +840,7 @@ class BlockPoolSlice {
       // and continue.
       LOG.warn("Failed to write replicas to cache ", e);
       if (replicaCacheFile.exists() && !replicaCacheFile.delete()) {
-        LOG.warn("Failed to delete replicas file: " + 
+        LOG.warn("Failed to delete replicas file: " +
             replicaCacheFile.getPath());
       }
     } finally {

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/FsDatasetImpl.java

@@ -268,24 +268,15 @@ class FsDatasetImpl implements FsDatasetSpi<FsVolumeImpl> {
     this.smallBufferSize = DFSUtilClient.getSmallBufferSize(conf);
     // The number of volumes required for operation is the total number
     // of volumes minus the number of failed volumes we can tolerate.
-    volFailuresTolerated =
-      conf.getInt(DFSConfigKeys.DFS_DATANODE_FAILED_VOLUMES_TOLERATED_KEY,
-                  DFSConfigKeys.DFS_DATANODE_FAILED_VOLUMES_TOLERATED_DEFAULT);
+    volFailuresTolerated = datanode.getDnConf().getVolFailuresTolerated();
 
-    String[] dataDirs = conf.getTrimmedStrings(DFSConfigKeys.DFS_DATANODE_DATA_DIR_KEY);
     Collection<StorageLocation> dataLocations = DataNode.getStorageLocations(conf);
     List<VolumeFailureInfo> volumeFailureInfos = getInitialVolumeFailureInfos(
         dataLocations, storage);
 
-    int volsConfigured = (dataDirs == null) ? 0 : dataDirs.length;
+    int volsConfigured = datanode.getDnConf().getVolsConfigured();
     int volsFailed = volumeFailureInfos.size();
 
-    if (volFailuresTolerated < 0 || volFailuresTolerated >= volsConfigured) {
-      throw new DiskErrorException("Invalid value configured for "
-          + "dfs.datanode.failed.volumes.tolerated - " + volFailuresTolerated
-          + ". Value configured is either less than 0 or >= "
-          + "to the number of configured volumes (" + volsConfigured + ").");
-    }
     if (volsFailed > volFailuresTolerated) {
       throw new DiskErrorException("Too many failed volumes - "
           + "current valid volumes: " + storage.getNumStorageDirs() 
@@ -1159,7 +1150,8 @@ class FsDatasetImpl implements FsDatasetSpi<FsVolumeImpl> {
     // construct a RBW replica with the new GS
     File blkfile = replicaInfo.getBlockFile();
     FsVolumeImpl v = (FsVolumeImpl)replicaInfo.getVolume();
-    if (v.getAvailable() < estimateBlockLen - replicaInfo.getNumBytes()) {
+    long bytesReserved = estimateBlockLen - replicaInfo.getNumBytes();
+    if (v.getAvailable() < bytesReserved) {
       throw new DiskOutOfSpaceException("Insufficient space for appending to "
           + replicaInfo);
     }
@@ -1167,7 +1159,7 @@ class FsDatasetImpl implements FsDatasetSpi<FsVolumeImpl> {
     File oldmeta = replicaInfo.getMetaFile();
     ReplicaBeingWritten newReplicaInfo = new ReplicaBeingWritten(
         replicaInfo.getBlockId(), replicaInfo.getNumBytes(), newGS,
-        v, newBlkFile.getParentFile(), Thread.currentThread(), estimateBlockLen);
+        v, newBlkFile.getParentFile(), Thread.currentThread(), bytesReserved);
     File newmeta = newReplicaInfo.getMetaFile();
 
     // rename meta file to rbw directory
@@ -1203,7 +1195,7 @@ class FsDatasetImpl implements FsDatasetSpi<FsVolumeImpl> {
     
     // Replace finalized replica by a RBW replica in replicas map
     volumeMap.add(bpid, newReplicaInfo);
-    v.reserveSpaceForReplica(estimateBlockLen - replicaInfo.getNumBytes());
+    v.reserveSpaceForReplica(bytesReserved);
     return newReplicaInfo;
   }
 

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java

@@ -184,7 +184,6 @@ import org.apache.hadoop.hdfs.protocol.HdfsFileStatus;
 import org.apache.hadoop.hdfs.protocol.LastBlockWithStatus;
 import org.apache.hadoop.hdfs.protocol.LocatedBlock;
 import org.apache.hadoop.hdfs.protocol.LocatedBlocks;
-import org.apache.hadoop.hdfs.protocol.LocatedStripedBlock;
 import org.apache.hadoop.hdfs.protocol.RecoveryInProgressException;
 import org.apache.hadoop.hdfs.protocol.RollingUpgradeException;
 import org.apache.hadoop.hdfs.protocol.RollingUpgradeInfo;
@@ -903,6 +902,36 @@ public class FSNamesystem implements Namesystem, FSNamesystemMBean,
     return null;
   }
 
+  /**
+   * Locate DefaultAuditLogger, if any, to enable/disable CallerContext.
+   *
+   * @param value
+   *          true, enable CallerContext, otherwise false to disable it.
+   */
+  void setCallerContextEnabled(final boolean value) {
+    for (AuditLogger logger : auditLoggers) {
+      if (logger instanceof DefaultAuditLogger) {
+        ((DefaultAuditLogger) logger).setCallerContextEnabled(value);
+        break;
+      }
+    }
+  }
+
+  /**
+   * Get the value indicating if CallerContext is enabled.
+   *
+   * @return true, if CallerContext is enabled, otherwise false, if it's
+   *         disabled.
+   */
+  boolean getCallerContextEnabled() {
+    for (AuditLogger logger : auditLoggers) {
+      if (logger instanceof DefaultAuditLogger) {
+        return ((DefaultAuditLogger) logger).getCallerContextEnabled();
+      }
+    }
+    return false;
+  }
+
   private List<AuditLogger> initAuditLoggers(Configuration conf) {
     // Initialize the custom access loggers if configured.
     Collection<String> alClasses =
@@ -1779,25 +1808,28 @@ public class FSNamesystem implements Namesystem, FSNamesystemMBean,
     }
 
     LocatedBlocks blocks = res.blocks;
+    sortLocatedBlocks(clientMachine, blocks);
+    return blocks;
+  }
+
+  private void sortLocatedBlocks(String clientMachine, LocatedBlocks blocks) {
     if (blocks != null) {
       List<LocatedBlock> blkList = blocks.getLocatedBlocks();
-      if (blkList == null || blkList.size() == 0 ||
-          blkList.get(0) instanceof LocatedStripedBlock) {
-        // no need to sort locations for striped blocks
-        return blocks;
+      if (blkList == null || blkList.size() == 0) {
+        // simply return, block list is empty
+        return;
       }
-      blockManager.getDatanodeManager().sortLocatedBlocks(
-          clientMachine, blkList);
+      blockManager.getDatanodeManager().sortLocatedBlocks(clientMachine,
+          blkList);
 
       // lastBlock is not part of getLocatedBlocks(), might need to sort it too
       LocatedBlock lastBlock = blocks.getLastLocatedBlock();
       if (lastBlock != null) {
         ArrayList<LocatedBlock> lastBlockList = Lists.newArrayList(lastBlock);
-        blockManager.getDatanodeManager().sortLocatedBlocks(
-            clientMachine, lastBlockList);
+        blockManager.getDatanodeManager().sortLocatedBlocks(clientMachine,
+            lastBlockList);
       }
     }
-    return blocks;
   }
 
   /**
@@ -4279,10 +4311,6 @@ public class FSNamesystem implements Namesystem, FSNamesystemMBean,
       setManualAndResourceLowSafeMode(!resourcesLow, resourcesLow);
       NameNode.stateChangeLog.info("STATE* Safe mode is ON.\n" +
           getSafeModeTip());
-      if (isEditlogOpenForWrite) {
-        getEditLog().logSyncAll();
-      }
-      NameNode.stateChangeLog.info("STATE* Safe mode is ON" + getSafeModeTip());
     } finally {
       writeUnlock();
     }
@@ -6850,13 +6878,33 @@ public class FSNamesystem implements Namesystem, FSNamesystemMBean,
           }
         };
 
-    private boolean isCallerContextEnabled;
+    private volatile boolean isCallerContextEnabled;
     private int callerContextMaxLen;
     private int callerSignatureMaxLen;
 
     private boolean logTokenTrackingId;
     private Set<String> debugCmdSet = new HashSet<String>();
 
+    /**
+     * Enable or disable CallerContext.
+     *
+     * @param value
+     *          true, enable CallerContext, otherwise false to disable it.
+     */
+    void setCallerContextEnabled(final boolean value) {
+      isCallerContextEnabled = value;
+    }
+
+    /**
+     * Get the value indicating if CallerContext is enabled.
+     *
+     * @return true, if CallerContext is enabled, otherwise false, if it's
+     *         disabled.
+     */
+    boolean getCallerContextEnabled() {
+      return isCallerContextEnabled;
+    }
+
     @Override
     public void initialize(Configuration conf) {
       isCallerContextEnabled = conf.getBoolean(

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/NameNode.java

@@ -107,6 +107,8 @@ import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.FS_DEFAULT_NAME
 import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.FS_TRASH_INTERVAL_DEFAULT;
 import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.FS_TRASH_INTERVAL_KEY;
 import static org.apache.hadoop.hdfs.client.HdfsClientConfigKeys.DFS_NAMENODE_RPC_PORT_DEFAULT;
+import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.HADOOP_CALLER_CONTEXT_ENABLED_KEY;
+import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.HADOOP_CALLER_CONTEXT_ENABLED_DEFAULT;
 import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_HA_AUTO_FAILOVER_ENABLED_DEFAULT;
 import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_HA_AUTO_FAILOVER_ENABLED_KEY;
 import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_HA_FENCE_METHODS_KEY;
@@ -277,7 +279,8 @@ public class NameNode extends ReconfigurableBase implements
       .unmodifiableList(Arrays
           .asList(DFS_HEARTBEAT_INTERVAL_KEY,
               DFS_NAMENODE_HEARTBEAT_RECHECK_INTERVAL_KEY,
-              FS_PROTECTED_DIRECTORIES));
+              FS_PROTECTED_DIRECTORIES,
+              HADOOP_CALLER_CONTEXT_ENABLED_KEY));
 
   private static final String USAGE = "Usage: hdfs namenode ["
       + StartupOption.BACKUP.getName() + "] | \n\t["
@@ -2008,7 +2011,9 @@ public class NameNode extends ReconfigurableBase implements
             + datanodeManager.getHeartbeatRecheckInterval());
       }
     case FS_PROTECTED_DIRECTORIES:
-      return getNamesystem().getFSDirectory().setProtectedDirectories(newVal);
+      return reconfProtectedDirectories(newVal);
+    case HADOOP_CALLER_CONTEXT_ENABLED_KEY:
+      return reconfCallerContextEnabled(newVal);
     default:
       break;
     }
@@ -2016,6 +2021,21 @@ public class NameNode extends ReconfigurableBase implements
         .get(property));
   }
 
+  private String reconfProtectedDirectories(String newVal) {
+    return getNamesystem().getFSDirectory().setProtectedDirectories(newVal);
+  }
+
+  private String reconfCallerContextEnabled(String newVal) {
+    Boolean callerContextEnabled;
+    if (newVal == null) {
+      callerContextEnabled = HADOOP_CALLER_CONTEXT_ENABLED_DEFAULT;
+    } else {
+      callerContextEnabled = Boolean.parseBoolean(newVal);
+    }
+    namesystem.setCallerContextEnabled(callerContextEnabled);
+    return Boolean.toString(callerContextEnabled);
+  }
+
   @Override  // ReconfigurableBase
   protected Configuration getNewConf() {
     return new HdfsConfiguration();

hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml

@@ -264,7 +264,7 @@
      dfs.datanode.dns.interface.
    </description>
  </property>
- 
+
 <property>
   <name>dfs.datanode.dns.nameserver</name>
   <value>default</value>
@@ -276,7 +276,7 @@
     dfs.datanode.dns.nameserver.
   </description>
  </property>
- 
+
  <property>
   <name>dfs.namenode.backup.address</name>
   <value>0.0.0.0:50100</value>
@@ -285,7 +285,7 @@
     If the port is 0 then the server will start on a free port.
   </description>
 </property>
- 
+
  <property>
   <name>dfs.namenode.backup.http-address</name>
   <value>0.0.0.0:50105</value>
@@ -1441,6 +1441,13 @@
   <description>
     The prefix for a given nameservice, contains a comma-separated
     list of namenodes for a given nameservice (eg EXAMPLENAMESERVICE).
+
+    Unique identifiers for each NameNode in the nameservice, delimited by
+    commas. This will be used by DataNodes to determine all the NameNodes
+    in the cluster. For example, if you used “mycluster” as the nameservice
+    ID previously, and you wanted to use “nn1” and “nn2” as the individual
+    IDs of the NameNodes, you would configure a property
+    dfs.ha.namenodes.mycluster, and its value "nn1,nn2".
   </description>
 </property>
 
@@ -3036,4 +3043,1024 @@
     refreshes the configuration files used by the class.
   </description>
 </property>
+
+<property>
+  <name>datanode.https.port</name>
+  <value>50475</value>
+  <description>
+    HTTPS port for DataNode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.balancer.dispatcherThreads</name>
+  <value>200</value>
+  <description>
+    Size of the thread pool for the HDFS balancer block mover.
+    dispatchExecutor
+  </description>
+</property>
+
+<property>
+  <name>dfs.balancer.movedWinWidth</name>
+  <value>5400000</value>
+  <description>
+    Window of time in ms for the HDFS balancer tracking blocks and its
+    locations.
+  </description>
+</property>
+
+<property>
+  <name>dfs.balancer.moverThreads</name>
+  <value>1000</value>
+  <description>
+    Thread pool size for executing block moves.
+    moverThreadAllocator
+  </description>
+</property>
+
+<property>
+  <name>dfs.balancer.max-size-to-move</name>
+  <value>10737418240</value>
+  <description>
+    Maximum number of bytes that can be moved by the balancer in a single
+    thread.
+  </description>
+</property>
+
+<property>
+  <name>dfs.balancer.getBlocks.min-block-size</name>
+  <value>10485760</value>
+  <description>
+    Minimum block threshold size in bytes to ignore when fetching a source's
+    block list.
+  </description>
+</property>
+
+<property>
+  <name>dfs.balancer.getBlocks.size</name>
+  <value>2147483648</value>
+  <description>
+    Total size in bytes of Datanode blocks to get when fetching a source's
+    block list.
+  </description>
+</property>
+
+<property>
+  <name>dfs.block.invalidate.limit</name>
+  <value>1000</value>
+  <description>
+    Limit on the list of invalidated block list kept by the Namenode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.block.misreplication.processing.limit</name>
+  <value>10000</value>
+  <description>
+    Maximum number of blocks to process for initializing replication queues.
+  </description>
+</property>
+
+<property>
+  <name>dfs.block.placement.ec.classname</name>
+  <value>org.apache.hadoop.hdfs.server.blockmanagement.BlockPlacementPolicyRackFaultTolerant</value>
+  <description>
+    Placement policy class for striped files.
+    Defaults to BlockPlacementPolicyRackFaultTolerant.class
+  </description>
+</property>
+
+<property>
+  <name>dfs.block.replicator.classname</name>
+  <value>org.apache.hadoop.hdfs.server.blockmanagement.BlockPlacementPolicyDefault</value>
+  <description>
+    Class representing block placement policy for non-striped files.
+  </description>
+</property>
+
+<property>
+  <name>dfs.blockreport.incremental.intervalMsec</name>
+  <value>0</value>
+  <description>
+    If set to a positive integer, the value in ms to wait between sending
+    incremental block reports from the Datanode to the Namenode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.checksum.type</name>
+  <value>CRC32C</value>
+  <description>
+    Checksum type
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.block.write.locateFollowingBlock.retries</name>
+  <value>5</value>
+  <description>
+    Number of retries to use when finding the next block during HDFS writes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.failover.proxy.provider</name>
+  <value></value>
+  <description>
+    The prefix (plus a required nameservice ID) for the class name of the
+    configured Failover proxy provider for the host.  For more detailed
+    information, please consult the "Configuration Details" section of
+    the HDFS High Availability documentation.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.key.provider.cache.expiry</name>
+  <value>864000000</value>
+  <description>
+    DFS client security key cache expiration in milliseconds.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.max.block.acquire.failures</name>
+  <value>3</value>
+  <description>
+    Maximum failures allowed when trying to get block information from a specific datanode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.read.prefetch.size</name>
+  <value></value>
+  <description>
+    The number of bytes for the DFSClient will fetch from the Namenode
+    during a read operation.  Defaults to 10 * ${dfs.blocksize}.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.read.short.circuit.replica.stale.threshold.ms</name>
+  <value>1800000</value>
+  <description>
+    Threshold in milliseconds for read entries during short-circuit local reads.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.read.shortcircuit.buffer.size</name>
+  <value>1048576</value>
+  <description>
+    Buffer size in bytes for short-circuit local reads.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.replica.accessor.builder.classes</name>
+  <value></value>
+  <description>
+    Comma-separated classes for building ReplicaAccessor.  If the classes
+    are specified, client will use external BlockReader that uses the
+    ReplicaAccessor built by the builder.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.retry.interval-ms.get-last-block-length</name>
+  <value>4000</value>
+  <description>
+    Retry interval in milliseconds to wait between retries in getting
+    block lengths from the datanodes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.retry.max.attempts</name>
+  <value>10</value>
+  <description>
+    Max retry attempts for DFSClient talking to namenodes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.retry.policy.enabled</name>
+  <value>false</value>
+  <description>
+    If true, turns on DFSClient retry policy.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.retry.policy.spec</name>
+  <value>10000,6,60000,10</value>
+  <description>
+    Set to pairs of timeouts and retries for DFSClient.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.retry.times.get-last-block-length</name>
+  <value>3</value>
+  <description>
+    Number of retries for calls to fetchLocatedBlocksAndGetLastBlockLength().
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.retry.window.base</name>
+  <value>3000</value>
+  <description>
+    Base time window in ms for DFSClient retries.  For each retry attempt,
+    this value is extended linearly (e.g. 3000 ms for first attempt and
+    first retry, 6000 ms for second retry, 9000 ms for third retry, etc.).
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.socket-timeout</name>
+  <value>60000</value>
+  <description>
+    Default timeout value in milliseconds for all sockets.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.socketcache.capacity</name>
+  <value>16</value>
+  <description>
+    Socket cache capacity (in entries) for short-circuit reads.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.socketcache.expiryMsec</name>
+  <value>3000</value>
+  <description>
+    Socket cache expiration for short-circuit reads in msec.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.test.drop.namenode.response.number</name>
+  <value>0</value>
+  <description>
+    The number of Namenode responses dropped by DFSClient for each RPC call.  Used
+    for testing the NN retry cache.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.hedged.read.threadpool.size</name>
+  <value>0</value>
+  <description>
+    Support 'hedged' reads in DFSClient. To enable this feature, set the parameter
+    to a positive number. The threadpool size is how many threads to dedicate
+    to the running of these 'hedged', concurrent reads in your client.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.hedged.read.threshold.millis</name>
+  <value>500</value>
+  <description>
+    Configure 'hedged' reads in DFSClient. This is the number of milliseconds
+    to wait before starting up a 'hedged' read.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.use.legacy.blockreader</name>
+  <value>false</value>
+  <description>
+    If true, use the RemoteBlockReader class for local read short circuit.  If false, use
+    the newer RemoteBlockReader2 class.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.write.byte-array-manager.count-limit</name>
+  <value>2048</value>
+  <description>
+    The maximum number of arrays allowed for each array length.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.write.byte-array-manager.count-reset-time-period-ms</name>
+  <value>10000</value>
+  <description>
+    The time period in milliseconds that the allocation count for each array length is
+    reset to zero if there is no increment.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.write.byte-array-manager.count-threshold</name>
+  <value>128</value>
+  <description>
+    The count threshold for each array length so that a manager is created only after the
+    allocation count exceeds the threshold. In other words, the particular array length
+    is not managed until the allocation count exceeds the threshold.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.write.byte-array-manager.enabled</name>
+  <value>false</value>
+  <description>
+    If true, enables byte array manager used by DFSOutputStream.
+  </description>
+</property>
+
+<property>
+  <name>dfs.client.write.max-packets-in-flight</name>
+  <value>80</value>
+  <description>
+    The maximum number of DFSPackets allowed in flight.
+  </description>
+</property>
+
+<property>
+  <name>dfs.content-summary.limit</name>
+  <value>5000</value>
+  <description>
+    The maximum content summary counts allowed in one locking period. 0 or a negative number
+    means no limit (i.e. no yielding).
+  </description>
+</property>
+
+<property>
+  <name>dfs.content-summary.sleep-microsec</name>
+  <value>500</value>
+  <description>
+    The length of time in microseconds to put the thread to sleep, between reaquiring the locks
+    in content summary computation.
+  </description>
+</property>
+
+<property>
+  <name>dfs.data.transfer.client.tcpnodelay</name>
+  <value>true</value>
+  <description>
+    If true, set TCP_NODELAY to sockets for transferring data from DFS client.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.balance.max.concurrent.moves</name>
+  <value>5</value>
+  <description>
+    Maximum number of threads for Datanode balancer pending moves.  This
+    value is reconfigurable via the "dfsadmin -reconfig" command.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.fsdataset.factory</name>
+  <value></value>
+  <description>
+    The class name for the underlying storage that stores replicas for a
+    Datanode.  Defaults to
+    org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetFactory.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.fsdataset.volume.choosing.policy</name>
+  <value></value>
+  <description>
+    The class name of the policy for choosing volumes in the list of
+    directories.  Defaults to
+    org.apache.hadoop.hdfs.server.datanode.fsdataset.RoundRobinVolumeChoosingPolicy.
+    If you would like to take into account available disk space, set the
+    value to
+    "org.apache.hadoop.hdfs.server.datanode.fsdataset.AvailableSpaceVolumeChoosingPolicy".
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.hostname</name>
+  <value></value>
+  <description>
+    Optional.  The hostname for the Datanode containing this
+    configuration file.  Will be different for each machine.
+    Defaults to current hostname.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.lazywriter.interval.sec</name>
+  <value>60</value>
+  <description>
+    Interval in seconds for Datanodes for lazy persist writes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.network.counts.cache.max.size</name>
+  <value>2147483647</value>
+  <description>
+    The maximum number of entries the datanode per-host network error
+    count cache may contain.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.oob.timeout-ms</name>
+  <value>1500,0,0,0</value>
+  <description>
+    Timeout value when sending OOB response for each OOB type, which are
+    OOB_RESTART, OOB_RESERVED1, OOB_RESERVED2, and OOB_RESERVED3,
+    respectively.  Currently, only OOB_RESTART is used.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.parallel.volumes.load.threads.num</name>
+  <value></value>
+  <description>
+    Maximum number of threads to use for upgrading data directories.
+    The default value is the number of storage directories in the
+    DataNode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.ram.disk.replica.tracker</name>
+  <value></value>
+  <description>
+    Name of the class implementing the RamDiskReplicaTracker interface.
+    Defaults to
+    org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.RamDiskReplicaLruTracker.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.restart.replica.expiration</name>
+  <value>50</value>
+  <description>
+    During shutdown for restart, the amount of time in seconds budgeted for
+    datanode restart.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.socket.reuse.keepalive</name>
+  <value>4000</value>
+  <description>
+    The window of time in ms before the DataXceiver closes a socket for a
+    single request.  If a second request occurs within that window, the
+    socket can be reused.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.socket.write.timeout</name>
+  <value>480000</value>
+  <description>
+    Timeout in ms for clients socket writes to DataNodes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.sync.behind.writes.in.background</name>
+  <value>false</value>
+  <description>
+    If set to true, then sync_file_range() system call will occur
+    asynchronously.  This property is only valid when the property
+    dfs.datanode.sync.behind.writes is true.
+  </description>
+</property>
+
+<property>
+  <name>dfs.datanode.transferTo.allowed</name>
+  <value>true</value>
+  <description>
+    If false, break block tranfers on 32-bit machines greater than
+    or equal to 2GB into smaller chunks.
+  </description>
+</property>
+
+<property>
+  <name>dfs.ha.fencing.methods</name>
+  <value></value>
+  <description>
+    A list of scripts or Java classes which will be used to fence
+    the Active NameNode during a failover.  See the HDFS High
+    Availability documentation for details on automatic HA
+    configuration.
+  </description>
+</property>
+
+<property>
+  <name>dfs.ha.standby.checkpoints</name>
+  <value>true</value>
+  <description>
+    If true, a NameNode in Standby state periodically takes a checkpoint
+    of the namespace, saves it to its local storage and then upload to
+    the remote NameNode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.ha.zkfc.port</name>
+  <value>8019</value>
+  <description>
+    The port number that the zookeeper failover controller RPC
+    server binds to.
+  </description>
+</property>
+
+<property>
+  <name>dfs.journalnode.edits.dir</name>
+  <value>/tmp/hadoop/dfs/journalnode/</value>
+  <description>
+    The directory where the journal edit files are stored.
+  </description>
+</property>
+
+<property>
+  <name>dfs.journalnode.kerberos.internal.spnego.principal</name>
+  <value></value>
+  <description>
+    Kerberos SPNEGO principal name used by the journal node.
+  </description>
+</property>
+
+<property>
+  <name>dfs.journalnode.kerberos.principal</name>
+  <value></value>
+  <description>
+    Kerberos principal name for the journal node.
+  </description>
+</property>
+
+<property>
+  <name>dfs.journalnode.keytab.file</name>
+  <value></value>
+  <description>
+    Kerberos keytab file for the journal node.
+  </description>
+</property>
+
+<property>
+  <name>dfs.ls.limit</name>
+  <value>1000</value>
+  <description>
+    Limit the number of files printed by ls. If less or equal to
+    zero, at most DFS_LIST_LIMIT_DEFAULT (= 1000) will be printed.
+  </description>
+</property>
+
+<property>
+  <name>dfs.mover.movedWinWidth</name>
+  <value>5400000</value>
+  <description>
+    The minimum time interval, in milliseconds, that a block can be
+    moved to another location again.
+  </description>
+</property>
+
+<property>
+  <name>dfs.mover.moverThreads</name>
+  <value>1000</value>
+  <description>
+    Configure the balancer's mover thread pool size.
+  </description>
+</property>
+
+<property>
+  <name>dfs.mover.retry.max.attempts</name>
+  <value>10</value>
+  <description>
+    The maximum number of retries before the mover consider the
+    move failed.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.audit.log.async</name>
+  <value>false</value>
+  <description>
+    If true, enables asynchronous audit log.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.audit.log.token.tracking.id</name>
+  <value>false</value>
+  <description>
+    If true, adds a tracking ID for all audit log events.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.available-space-block-placement-policy.balanced-space-preference-fraction</name>
+  <value>0.6</value>
+  <description>
+    Only used when the dfs.block.replicator.classname is set to
+    org.apache.hadoop.hdfs.server.blockmanagement.AvailableSpaceBlockPlacementPolicy.
+    Special value between 0 and 1, noninclusive.  Increases chance of
+    placing blocks on Datanodes with less disk space used.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.backup.dnrpc-address</name>
+  <value></value>
+  <description>
+    Service RPC address for the backup Namenode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.delegation.token.always-use</name>
+  <value>false</value>
+  <description>
+    For testing.  Setting to true always allows the DT secret manager
+    to be used, even if security is disabled.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.edits.asynclogging</name>
+  <value>false</value>
+  <description>
+    If set to true, enables asynchronous edit logs in the Namenode.  If set
+    to false, the Namenode uses the traditional synchronous edit logs.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.edits.dir.minimum</name>
+  <value>1</value>
+  <description>
+    dfs.namenode.edits.dir includes both required directories
+    (specified by dfs.namenode.edits.dir.required) and optional directories.
+
+    The number of usable optional directories must be greater than or equal
+    to this property.  If the number of usable optional directories falls
+    below dfs.namenode.edits.dir.minimum, HDFS will issue an error.
+
+    This property defaults to 1.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.edits.journal-plugin</name>
+  <value></value>
+  <description>
+    When FSEditLog is creating JournalManagers from dfs.namenode.edits.dir,
+    and it encounters a URI with a schema different to "file" it loads the
+    name of the implementing class from
+    "dfs.namenode.edits.journal-plugin.[schema]". This class must implement
+    JournalManager and have a constructor which takes (Configuration, URI).
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.file.close.num-committed-allowed</name>
+  <value>0</value>
+  <description>
+    Normally a file can only be closed with all its blocks are committed.
+    When this value is set to a positive integer N, a file can be closed
+    when N blocks are committed and the rest complete.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.inode.attributes.provider.class</name>
+  <value></value>
+  <description>
+    Name of class to use for delegating HDFS authorization.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.max-num-blocks-to-log</name>
+  <value>1000</value>
+  <description>
+    Puts a limit on the number of blocks printed to the log by the Namenode
+    after a block report.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.max.op.size</name>
+  <value>52428800</value>
+  <description>
+    Maximum opcode size in bytes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.missing.checkpoint.periods.before.shutdown</name>
+  <value>3</value>
+  <description>
+    The number of checkpoint period windows (as defined by the property
+    dfs.namenode.checkpoint.period) allowed by the Namenode to perform
+    saving the namespace before shutdown.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.name.cache.threshold</name>
+  <value>10</value>
+  <description>
+    Frequently accessed files that are accessed more times than this
+    threshold are cached in the FSDirectory nameCache.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.replication.max-streams</name>
+  <value>2</value>
+  <description>
+    Hard limit for the number of highest-priority replication streams.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.replication.max-streams-hard-limit</name>
+  <value>4</value>
+  <description>
+    Hard limit for all replication streams.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.replication.pending.timeout-sec</name>
+  <value>-1</value>
+  <description>
+    Timeout in seconds for block replication.  If this value is 0 or less,
+    then it will default to 5 minutes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.stale.datanode.minimum.interval</name>
+  <value>3</value>
+  <description>
+    Minimum number of missed heartbeats intervals for a datanode to
+    be marked stale by the Namenode.  The actual interval is calculated as
+    (dfs.namenode.stale.datanode.minimum.interval * dfs.heartbeat.interval)
+    in seconds.  If this value is greater than the property
+    dfs.namenode.stale.datanode.interval, then the calculated value above
+    is used.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.storageinfo.defragment.timeout.ms</name>
+  <value>4</value>
+  <description>
+    Timeout value in ms for the StorageInfo compaction run.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.storageinfo.defragment.interval.ms</name>
+  <value>600000</value>
+  <description>
+    The thread for checking the StorageInfo for defragmentation will
+    run periodically.  The time between runs is determined by this
+    property.
+  </description>
+</property>
+
+<property>
+  <name>dfs.namenode.storageinfo.defragment.ratio</name>
+  <value>0.75</value>
+  <description>
+    The defragmentation threshold for the StorageInfo.
+  </description>
+</property>
+
+<property>
+  <name>dfs.pipeline.ecn</name>
+  <value>false</value>
+  <description>
+    If true, allows ECN (explicit congestion notification) from the
+    Datanode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.accept-recovery.timeout.ms</name>
+  <value>120000</value>
+  <description>
+    Quorum timeout in milliseconds during accept phase of
+    recovery/synchronization for a specific segment.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.finalize-segment.timeout.ms</name>
+  <value>120000</value>
+  <description>
+    Quorum timeout in milliseconds during finalizing for a specific
+    segment.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.get-journal-state.timeout.ms</name>
+  <value>120000</value>
+  <description>
+    Timeout in milliseconds when calling getJournalState().
+    JournalNodes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.new-epoch.timeout.ms</name>
+  <value>120000</value>
+  <description>
+    Timeout in milliseconds when getting an epoch number for write
+    access to JournalNodes.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.prepare-recovery.timeout.ms</name>
+  <value>120000</value>
+  <description>
+    Quorum timeout in milliseconds during preparation phase of
+    recovery/synchronization for a specific segment.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.queued-edits.limit.mb</name>
+  <value>10</value>
+  <description>
+    Queue size in MB for quorum journal edits.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.select-input-streams.timeout.ms</name>
+  <value>20000</value>
+  <description>
+    Timeout in milliseconds for accepting streams from JournalManagers.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.start-segment.timeout.ms</name>
+  <value>20000</value>
+  <description>
+    Quorum timeout in milliseconds for starting a log segment.
+  </description>
+</property>
+
+<property>
+  <name>dfs.qjournal.write-txns.timeout.ms</name>
+  <value>20000</value>
+  <description>
+    Write timeout in milliseconds when writing to a quorum of remote
+    journals.
+  </description>
+</property>
+
+<property>
+  <name>dfs.quota.by.storage.type.enabled</name>
+  <value>true</value>
+  <description>
+    If true, enables quotas based on storage type.
+  </description>
+</property>
+
+<property>
+  <name>dfs.secondary.namenode.kerberos.principal</name>
+  <value></value>
+  <description>
+    Kerberos principal name for the Secondary NameNode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.secondary.namenode.keytab.file</name>
+  <value></value>
+  <description>
+    Kerberos keytab file for the Secondary NameNode.
+  </description>
+</property>
+
+<property>
+  <name>dfs.web.authentication.filter</name>
+  <value>org.apache.hadoop.hdfs.web.AuthFilter</value>
+  <description>
+    Authentication filter class used for WebHDFS.
+  </description>
+</property>
+
+<property>
+  <name>dfs.web.authentication.simple.anonymous.allowed</name>
+  <value></value>
+  <description>
+    If true, allow anonymous user to access WebHDFS. Set to
+    false to disable anonymous authentication.
+  </description>
+</property>
+
+<property>
+  <name>dfs.web.ugi</name>
+  <value></value>
+  <description>
+    dfs.web.ugi is deprecated. Use hadoop.http.staticuser.user instead.
+  </description>
+</property>
+
+<property>
+  <name>dfs.webhdfs.netty.high.watermark</name>
+  <value>65535</value>
+  <description>
+    High watermark configuration to Netty for Datanode WebHdfs.
+  </description>
+</property>
+
+<property>
+  <name>dfs.webhdfs.netty.low.watermark</name>
+  <value>32768</value>
+  <description>
+    Low watermark configuration to Netty for Datanode WebHdfs.
+  </description>
+</property>
+
+<property>
+  <name>dfs.webhdfs.oauth2.access.token.provider</name>
+  <value></value>
+  <description>
+    Access token provider class for WebHDFS using OAuth2.
+    Defaults to org.apache.hadoop.hdfs.web.oauth2.ConfCredentialBasedAccessTokenProvider.
+  </description>
+</property>
+
+<property>
+  <name>dfs.webhdfs.oauth2.client.id</name>
+  <value></value>
+  <description>
+    Client id used to obtain access token with either credential or
+    refresh token.
+  </description>
+</property>
+
+<property>
+  <name>dfs.webhdfs.oauth2.enabled</name>
+  <value>false</value>
+  <description>
+    If true, enables OAuth2 in WebHDFS
+  </description>
+</property>
+
+<property>
+  <name>dfs.webhdfs.oauth2.refresh.url</name>
+  <value></value>
+  <description>
+    URL against which to post for obtaining bearer token with
+    either credential or refresh token.
+  </description>
+</property>
+
+<property>
+  <name>ssl.server.keystore.keypassword</name>
+  <value></value>
+  <description>
+    Keystore key password for HTTPS SSL configuration
+  </description>
+</property>
+
+<property>
+  <name>ssl.server.keystore.location</name>
+  <value></value>
+  <description>
+    Keystore location for HTTPS SSL configuration
+  </description>
+</property>
+
+<property>
+  <name>ssl.server.keystore.password</name>
+  <value></value>
+  <description>
+    Keystore password for HTTPS SSL configuration
+  </description>
+</property>
+
+<property>
+  <name>ssl.server.truststore.location</name>
+  <value></value>
+  <description>
+    Truststore location for HTTPS SSL configuration
+  </description>
+</property>
+
+<property>
+  <name>ssl.server.truststore.password</name>
+  <value></value>
+  <description>
+    Truststore password for HTTPS SSL configuration
+  </description>
+</property>
 </configuration>

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestDecommissionWithStriped.java

@@ -27,6 +27,7 @@ import static org.junit.Assert.assertTrue;
 import java.io.IOException;
 import java.util.ArrayList;
 import java.util.Collection;
+import java.util.HashMap;
 import java.util.Iterator;
 import java.util.List;
 import java.util.concurrent.CountDownLatch;
@@ -49,12 +50,15 @@ import org.apache.hadoop.hdfs.client.HdfsDataInputStream;
 import org.apache.hadoop.hdfs.protocol.DatanodeInfo;
 import org.apache.hadoop.hdfs.protocol.DatanodeInfo.AdminStates;
 import org.apache.hadoop.hdfs.protocol.HdfsConstants.DatanodeReportType;
+import org.apache.hadoop.hdfs.security.token.block.BlockTokenIdentifier;
 import org.apache.hadoop.hdfs.protocol.LocatedBlock;
+import org.apache.hadoop.hdfs.protocol.LocatedStripedBlock;
 import org.apache.hadoop.hdfs.server.blockmanagement.BlockManager;
 import org.apache.hadoop.hdfs.server.datanode.DataNode;
 import org.apache.hadoop.hdfs.server.namenode.FSNamesystem;
 import org.apache.hadoop.hdfs.server.namenode.NameNode;
 import org.apache.hadoop.hdfs.server.namenode.NameNodeAdapter;
+import org.apache.hadoop.security.token.Token;
 import org.apache.hadoop.test.PathUtils;
 import org.junit.After;
 import org.junit.Assert;
@@ -159,6 +163,13 @@ public class TestDecommissionWithStriped {
     testDecommission(blockSize * dataBlocks, 9, 1, "testFileFullBlockGroup");
   }
 
+  @Test(timeout = 120000)
+  public void testFileMultipleBlockGroups() throws Exception {
+    LOG.info("Starting test testFileMultipleBlockGroups");
+    int writeBytes = 2 * blockSize * dataBlocks;
+    testDecommission(writeBytes, 9, 1, "testFileMultipleBlockGroups");
+  }
+
   @Test(timeout = 120000)
   public void testFileSmallerThanOneCell() throws Exception {
     LOG.info("Starting test testFileSmallerThanOneCell");
@@ -274,7 +285,15 @@ public class TestDecommissionWithStriped {
 
     int deadDecomissioned = fsn.getNumDecomDeadDataNodes();
     int liveDecomissioned = fsn.getNumDecomLiveDataNodes();
-    ((HdfsDataInputStream) dfs.open(ecFile)).getAllBlocks();
+    List<LocatedBlock> lbs = ((HdfsDataInputStream) dfs.open(ecFile))
+        .getAllBlocks();
+
+    // prepare expected block index and token list.
+    List<HashMap<DatanodeInfo, Byte>> locToIndexList = new ArrayList<>();
+    List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList =
+        new ArrayList<>();
+    prepareBlockIndexAndTokenList(lbs, locToIndexList, locToTokenList);
+
     // Decommission node. Verify that node is decommissioned.
     decommissionNode(0, decommisionNodes, AdminStates.DECOMMISSIONED);
 
@@ -290,9 +309,55 @@ public class TestDecommissionWithStriped {
     assertNull(checkFile(dfs, ecFile, storageCount, decommisionNodes, numDNs));
     StripedFileTestUtil.checkData(dfs, ecFile, writeBytes, decommisionNodes,
         null);
+
+    assertBlockIndexAndTokenPosition(lbs, locToIndexList, locToTokenList);
+
     cleanupFile(dfs, ecFile);
   }
 
+  private void prepareBlockIndexAndTokenList(List<LocatedBlock> lbs,
+      List<HashMap<DatanodeInfo, Byte>> locToIndexList,
+      List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList) {
+    for (LocatedBlock lb : lbs) {
+      HashMap<DatanodeInfo, Byte> locToIndex = new HashMap<DatanodeInfo, Byte>();
+      locToIndexList.add(locToIndex);
+
+      HashMap<DatanodeInfo, Token<BlockTokenIdentifier>> locToToken =
+          new HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>();
+      locToTokenList.add(locToToken);
+
+      DatanodeInfo[] di = lb.getLocations();
+      LocatedStripedBlock stripedBlk = (LocatedStripedBlock) lb;
+      for (int i = 0; i < di.length; i++) {
+        locToIndex.put(di[i], stripedBlk.getBlockIndices()[i]);
+        locToToken.put(di[i], stripedBlk.getBlockTokens()[i]);
+      }
+    }
+  }
+
+  /**
+   * Verify block index and token values. Must update block indices and block
+   * tokens after sorting.
+   */
+  private void assertBlockIndexAndTokenPosition(List<LocatedBlock> lbs,
+      List<HashMap<DatanodeInfo, Byte>> locToIndexList,
+      List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList) {
+    for (int i = 0; i < lbs.size(); i++) {
+      LocatedBlock lb = lbs.get(i);
+      LocatedStripedBlock stripedBlk = (LocatedStripedBlock) lb;
+      HashMap<DatanodeInfo, Byte> locToIndex = locToIndexList.get(i);
+      HashMap<DatanodeInfo, Token<BlockTokenIdentifier>> locToToken =
+          locToTokenList.get(i);
+      DatanodeInfo[] di = lb.getLocations();
+      for (int j = 0; j < di.length; j++) {
+        Assert.assertEquals("Block index value mismatches after sorting",
+            (byte) locToIndex.get(di[j]), stripedBlk.getBlockIndices()[j]);
+        Assert.assertEquals("Block token value mismatches after sorting",
+            locToToken.get(di[j]), stripedBlk.getBlockTokens()[j]);
+      }
+    }
+  }
+
   private List<DatanodeInfo> getDecommissionDatanode(DistributedFileSystem dfs,
       Path ecFile, int writeBytes, int decomNodeCount) throws IOException {
     ArrayList<DatanodeInfo> decommissionedNodes = new ArrayList<>();
@@ -447,7 +512,12 @@ public class TestDecommissionWithStriped {
               return "For block " + blk.getBlock() + " replica on " + nodes[j]
                   + " is given as downnode, " + "but is not decommissioned";
             }
-            // TODO: Add check to verify that the Decommissioned node (if any)
+            // Decommissioned node (if any) should only be last node in list.
+            if (j < repl) {
+              return "For block " + blk.getBlock() + " decommissioned node "
+                  + nodes[j] + " was not last node in list: " + (j + 1) + " of "
+                  + nodes.length;
+            }
             // should only be last node in list.
             LOG.info("Block " + blk.getBlock() + " replica on " + nodes[j]
                 + " is decommissioned.");
@@ -470,4 +540,4 @@ public class TestDecommissionWithStriped {
     }
     return null;
   }
-}
+}

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestSortLocatedStripedBlock.java

@@ -0,0 +1,557 @@
+/**
+ * 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.hadoop.hdfs.server.blockmanagement;
+
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.StorageType;
+import org.apache.hadoop.hdfs.DFSConfigKeys;
+import org.apache.hadoop.hdfs.DFSTestUtil;
+import org.apache.hadoop.hdfs.StripedFileTestUtil;
+import org.apache.hadoop.hdfs.protocol.DatanodeInfo;
+import org.apache.hadoop.hdfs.protocol.DatanodeInfo.AdminStates;
+import org.apache.hadoop.hdfs.security.token.block.BlockTokenIdentifier;
+import org.apache.hadoop.hdfs.protocol.ExtendedBlock;
+import org.apache.hadoop.hdfs.protocol.LocatedBlock;
+import org.apache.hadoop.hdfs.protocol.LocatedStripedBlock;
+import org.apache.hadoop.hdfs.server.namenode.FSNamesystem;
+import org.apache.hadoop.security.token.Token;
+import org.apache.hadoop.util.Time;
+import org.junit.Assert;
+import org.junit.BeforeClass;
+import org.junit.Test;
+import org.mockito.Mockito;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+/**
+ * This class tests the sorting of located striped blocks based on
+ * decommissioned states.
+ */
+public class TestSortLocatedStripedBlock {
+  static final Logger LOG = LoggerFactory
+      .getLogger(TestSortLocatedStripedBlock.class);
+  static final int BLK_GROUP_WIDTH = StripedFileTestUtil.NUM_DATA_BLOCKS
+      + StripedFileTestUtil.NUM_PARITY_BLOCKS;
+  static final int NUM_DATA_BLOCKS = StripedFileTestUtil.NUM_DATA_BLOCKS;
+  static final int NUM_PARITY_BLOCKS = StripedFileTestUtil.NUM_PARITY_BLOCKS;
+  static DatanodeManager dm;
+  static final long STALE_INTERVAL = 30 * 1000 * 60;
+
+  @BeforeClass
+  public static void setup() throws IOException {
+    dm = mockDatanodeManager();
+  }
+
+  /**
+   * Test to verify sorting with multiple decommissioned datanodes exists in
+   * storage lists.
+   *
+   * We have storage list, marked decommissioned internal blocks with a '
+   * d0, d1, d2, d3, d4, d5, d6, d7, d8, d9, d10, d11, d12
+   * mapping to indices
+   * 0', 1', 2, 3, 4, 5, 6, 7', 8', 0, 1, 7, 8
+   *
+   * Decommissioned node indices: 0, 1, 7, 8
+   *
+   * So in the original list nodes d0, d1, d7, d8 are decommissioned state.
+   *
+   * After sorting the expected block indices list should be,
+   * 0, 1, 2, 3, 4, 5, 6, 7, 8, 0', 1', 7', 8'
+   *
+   * After sorting the expected storage list will be,
+   * d9, d10, d2, d3, d4, d5, d6, d11, d12, d0, d1, d7, d8.
+   *
+   * Note: after sorting block indices will not be in ascending order.
+   */
+  @Test(timeout = 10000)
+  public void testWithMultipleDecommnDatanodes() {
+    LOG.info("Starting test testSortWithMultipleDecommnDatanodes");
+    int lbsCount = 2; // two located block groups
+    List<Integer> decommnNodeIndices = new ArrayList<>();
+    decommnNodeIndices.add(0);
+    decommnNodeIndices.add(1);
+    decommnNodeIndices.add(7);
+    decommnNodeIndices.add(8);
+    List<Integer> targetNodeIndices = new ArrayList<>();
+    targetNodeIndices.addAll(decommnNodeIndices);
+    // map contains decommissioned node details in each located strip block
+    // which will be used for assertions
+    HashMap<Integer, List<String>> decommissionedNodes = new HashMap<>(
+        lbsCount * decommnNodeIndices.size());
+    List<LocatedBlock> lbs = createLocatedStripedBlocks(lbsCount,
+        NUM_DATA_BLOCKS, NUM_PARITY_BLOCKS, decommnNodeIndices,
+        targetNodeIndices, decommissionedNodes);
+
+    // prepare expected block index and token list.
+    List<HashMap<DatanodeInfo, Byte>> locToIndexList = new ArrayList<>();
+    List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList =
+        new ArrayList<>();
+    prepareBlockIndexAndTokenList(lbs, locToIndexList, locToTokenList);
+
+    dm.sortLocatedBlocks(null, lbs);
+
+    assertDecommnNodePosition(BLK_GROUP_WIDTH, decommissionedNodes, lbs);
+    assertBlockIndexAndTokenPosition(lbs, locToIndexList, locToTokenList);
+  }
+
+  /**
+   * Test to verify sorting with two decommissioned datanodes exists in
+   * storage lists for the same block index.
+   *
+   * We have storage list, marked decommissioned internal blocks with a '
+   * d0, d1, d2, d3, d4, d5, d6, d7, d8, d9, d10, d11, d12, d13
+   * mapping to indices
+   * 0', 1', 2, 3, 4', 5', 6, 7, 8, 0, 1', 4, 5, 1
+   *
+   * Decommissioned node indices: 0', 1', 4', 5', 1'
+   *
+   * Here decommissioned has done twice to the datanode block index 1.
+   * So in the original list nodes d0, d1, d4, d5, d10 are decommissioned state.
+   *
+   * After sorting the expected block indices list will be,
+   * 0, 1, 2, 3, 4, 5, 6, 7, 8, 0', 1', 1', 4', 5'
+   *
+   * After sorting the expected storage list will be,
+   * d9, d13, d2, d3, d11, d12, d6, d7, d8, d0, d1, d10, d4, d5.
+   *
+   * Note: after sorting block indices will not be in ascending order.
+   */
+  @Test(timeout = 10000)
+  public void testTwoDatanodesWithSameBlockIndexAreDecommn() {
+    LOG.info("Starting test testTwoDatanodesWithSameBlockIndexAreDecommn");
+    int lbsCount = 2; // two located block groups
+    List<Integer> decommnNodeIndices = new ArrayList<>();
+    decommnNodeIndices.add(0);
+    decommnNodeIndices.add(1);
+    decommnNodeIndices.add(4);
+    decommnNodeIndices.add(5);
+    // representing blockIndex 1, later this also decommissioned
+    decommnNodeIndices.add(1);
+
+    List<Integer> targetNodeIndices = new ArrayList<>();
+    targetNodeIndices.addAll(decommnNodeIndices);
+    // map contains decommissioned node details in each located strip block
+    // which will be used for assertions
+    HashMap<Integer, List<String>> decommissionedNodes = new HashMap<>(
+        lbsCount * decommnNodeIndices.size());
+    List<LocatedBlock> lbs = createLocatedStripedBlocks(lbsCount,
+        NUM_DATA_BLOCKS, NUM_PARITY_BLOCKS, decommnNodeIndices,
+        targetNodeIndices, decommissionedNodes);
+
+    // prepare expected block index and token list.
+    List<HashMap<DatanodeInfo, Byte>> locToIndexList = new ArrayList<>();
+    List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList =
+        new ArrayList<>();
+    prepareBlockIndexAndTokenList(lbs, locToIndexList, locToTokenList);
+
+    dm.sortLocatedBlocks(null, lbs);
+    assertDecommnNodePosition(BLK_GROUP_WIDTH, decommissionedNodes, lbs);
+    assertBlockIndexAndTokenPosition(lbs, locToIndexList, locToTokenList);
+  }
+
+  /**
+   * Test to verify sorting with decommissioned datanodes exists in storage
+   * list which is smaller than stripe size.
+   *
+   * We have storage list, marked decommissioned internal blocks with a '
+   * d0, d1, d2, d3, d6, d7, d8, d9, d10, d11
+   * mapping to indices
+   * 0', 1, 2', 3, 6, 7, 8, 0, 2', 2
+   *
+   * Decommissioned node indices: 0', 2', 2'
+   *
+   * Here decommissioned has done twice to the datanode block index 2.
+   * So in the original list nodes d0, d2, d10 are decommissioned state.
+   *
+   * After sorting the expected block indices list should be,
+   * 0, 1, 2, 3, 6, 7, 8, 0', 2', 2'
+   *
+   * After sorting the expected storage list will be,
+   * d9, d1, d11, d3, d6, d7, d8, d0, d2, d10.
+   *
+   * Note: after sorting block indices will not be in ascending order.
+   */
+  @Test(timeout = 10000)
+  public void testSmallerThanOneStripeWithMultpleDecommnNodes()
+      throws Exception {
+    LOG.info("Starting test testSmallerThanOneStripeWithDecommn");
+    int lbsCount = 2; // two located block groups
+    List<Integer> decommnNodeIndices = new ArrayList<>();
+    decommnNodeIndices.add(0);
+    decommnNodeIndices.add(2);
+    // representing blockIndex 1, later this also decommissioned
+    decommnNodeIndices.add(2);
+
+    List<Integer> targetNodeIndices = new ArrayList<>();
+    targetNodeIndices.addAll(decommnNodeIndices);
+    // map contains decommissioned node details in each located strip block
+    // which will be used for assertions
+    HashMap<Integer, List<String>> decommissionedNodes = new HashMap<>(
+        lbsCount * decommnNodeIndices.size());
+    int dataBlksNum = NUM_DATA_BLOCKS - 2;
+    List<LocatedBlock> lbs = createLocatedStripedBlocks(lbsCount, dataBlksNum,
+        NUM_PARITY_BLOCKS, decommnNodeIndices, targetNodeIndices,
+        decommissionedNodes);
+
+    // prepare expected block index and token list.
+    List<HashMap<DatanodeInfo, Byte>> locToIndexList = new ArrayList<>();
+    List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList =
+        new ArrayList<>();
+    prepareBlockIndexAndTokenList(lbs, locToIndexList, locToTokenList);
+
+    dm.sortLocatedBlocks(null, lbs);
+
+    // After this index all are decommissioned nodes.
+    int blkGrpWidth = dataBlksNum + NUM_PARITY_BLOCKS;
+    assertDecommnNodePosition(blkGrpWidth, decommissionedNodes, lbs);
+    assertBlockIndexAndTokenPosition(lbs, locToIndexList, locToTokenList);
+  }
+
+  /**
+   * Test to verify sorting with decommissioned datanodes exists in storage
+   * list but the corresponding new target datanode doesn't exists.
+   *
+   * We have storage list, marked decommissioned internal blocks with a '
+   * d0, d1, d2, d3, d4, d5, d6, d7, d8, d9, d10, d11
+   * mapping to indices
+   * 0', 1', 2', 3, 4', 5', 6, 7, 8, 0, 2, 4
+   *
+   * Decommissioned node indices: 0', 1', 2', 4', 5'
+   *
+   * 1 and 5 nodes doesn't exists in the target list. This can happen, the
+   * target node block corrupted or lost after the successful decommissioning.
+   * So in the original list nodes corresponding to the decommissioned block
+   * index 1 and 5 doesn't have any target entries.
+   *
+   * After sorting the expected block indices list should be,
+   * 0, 2, 3, 4, 6, 7, 8, 0', 1', 2', 4', 5'
+   *
+   * After sorting the expected storage list will be,
+   * d9, d10, d3, d11, d6, d7, d8, d0, d1, d2, d4, d5.
+   *
+   * Note: after sorting block indices will not be in ascending order.
+   */
+  @Test(timeout = 10000)
+  public void testTargetDecommnDatanodeDoesntExists() {
+    LOG.info("Starting test testTargetDecommnDatanodeDoesntExists");
+    int lbsCount = 2; // two located block groups
+    List<Integer> decommnNodeIndices = new ArrayList<>();
+    decommnNodeIndices.add(0);
+    decommnNodeIndices.add(1);
+    decommnNodeIndices.add(2);
+    decommnNodeIndices.add(4);
+    decommnNodeIndices.add(5);
+
+    List<Integer> targetNodeIndices = new ArrayList<>();
+    targetNodeIndices.add(0);
+    targetNodeIndices.add(2);
+    targetNodeIndices.add(4);
+    // 1 and 5 nodes doesn't exists in the target list. One such case is, the
+    // target node block corrupted or lost after the successful decommissioning
+
+    // map contains decommissioned node details in each located strip block
+    // which will be used for assertions
+    HashMap<Integer, List<String>> decommissionedNodes = new HashMap<>(
+        lbsCount * decommnNodeIndices.size());
+    List<LocatedBlock> lbs = createLocatedStripedBlocks(lbsCount,
+        NUM_DATA_BLOCKS, NUM_PARITY_BLOCKS, decommnNodeIndices,
+        targetNodeIndices, decommissionedNodes);
+
+    // prepare expected block index and token list.
+    List<HashMap<DatanodeInfo, Byte>> locToIndexList = new ArrayList<>();
+    List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList =
+        new ArrayList<>();
+    prepareBlockIndexAndTokenList(lbs, locToIndexList, locToTokenList);
+
+    dm.sortLocatedBlocks(null, lbs);
+
+    // After this index all are decommissioned nodes. Needs to reconstruct two
+    // more block indices.
+    int blkGrpWidth = NUM_DATA_BLOCKS + NUM_PARITY_BLOCKS - 2;
+    assertDecommnNodePosition(blkGrpWidth, decommissionedNodes, lbs);
+    assertBlockIndexAndTokenPosition(lbs, locToIndexList, locToTokenList);
+  }
+
+  /**
+   * Test to verify sorting with multiple in-service and decommissioned
+   * datanodes exists in storage lists.
+   *
+   * We have storage list, marked decommissioned internal blocks with a '
+   * d0, d1, d2, d3, d4, d5, d6, d7, d8, d9, d10, d11, d12, d13
+   * mapping to indices
+   * 0', 1', 2, 3, 4, 5, 6, 7', 8', 0, 1, 7, 8, 1
+   *
+   * Decommissioned node indices: 0', 1', 7', 8'
+   *
+   * Additional In-Service node d13 at the end, block index: 1
+   *
+   * So in the original list nodes d0, d1, d7, d8 are decommissioned state.
+   *
+   * After sorting the expected block indices list will be,
+   * 0, 1, 2, 3, 4, 5, 6, 7, 8, 1, 0', 1', 7', 8'
+   *
+   * After sorting the expected storage list will be,
+   * d9, d10, d2, d3, d4, d5, d6, d11, d12, d13, d0, d1, d7, d8.
+   *
+   * Note: after sorting block indices will not be in ascending order.
+   */
+  @Test(timeout = 10000)
+  public void testWithMultipleInServiceAndDecommnDatanodes() {
+    LOG.info("Starting test testWithMultipleInServiceAndDecommnDatanodes");
+    int lbsCount = 2; // two located block groups
+    List<Integer> decommnNodeIndices = new ArrayList<>();
+    decommnNodeIndices.add(0);
+    decommnNodeIndices.add(1);
+    decommnNodeIndices.add(7);
+    decommnNodeIndices.add(8);
+    List<Integer> targetNodeIndices = new ArrayList<>();
+    targetNodeIndices.addAll(decommnNodeIndices);
+
+    // at the end add an additional In-Service node to blockIndex=1
+    targetNodeIndices.add(1);
+
+    // map contains decommissioned node details in each located strip block
+    // which will be used for assertions
+    HashMap<Integer, List<String>> decommissionedNodes = new HashMap<>(
+        lbsCount * decommnNodeIndices.size());
+    List<LocatedBlock> lbs = createLocatedStripedBlocks(lbsCount,
+        NUM_DATA_BLOCKS, NUM_PARITY_BLOCKS, decommnNodeIndices,
+        targetNodeIndices, decommissionedNodes);
+    List <DatanodeInfo> staleDns = new ArrayList<>();
+    for (LocatedBlock lb : lbs) {
+      DatanodeInfo[] locations = lb.getLocations();
+      DatanodeInfo staleDn = locations[locations.length - 1];
+      staleDn
+          .setLastUpdateMonotonic(Time.monotonicNow() - (STALE_INTERVAL * 2));
+      staleDns.add(staleDn);
+    }
+
+    // prepare expected block index and token list.
+    List<HashMap<DatanodeInfo, Byte>> locToIndexList = new ArrayList<>();
+    List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList =
+        new ArrayList<>();
+    prepareBlockIndexAndTokenList(lbs, locToIndexList, locToTokenList);
+
+    dm.sortLocatedBlocks(null, lbs);
+
+    assertDecommnNodePosition(BLK_GROUP_WIDTH + 1, decommissionedNodes, lbs);
+    assertBlockIndexAndTokenPosition(lbs, locToIndexList, locToTokenList);
+
+    for (LocatedBlock lb : lbs) {
+      byte[] blockIndices = ((LocatedStripedBlock) lb).getBlockIndices();
+      // after sorting stale block index will be placed after normal nodes.
+      Assert.assertEquals("Failed to move stale node to bottom!", 1,
+          blockIndices[9]);
+      DatanodeInfo[] locations = lb.getLocations();
+      // After sorting stale node d13 will be placed after normal nodes
+      Assert.assertEquals("Failed to move stale dn after normal one!",
+          staleDns.remove(0), locations[9]);
+    }
+  }
+
+  /**
+   * Verify that decommissioned/stale nodes must be positioned after normal
+   * nodes.
+   */
+  private void assertDecommnNodePosition(int blkGrpWidth,
+      HashMap<Integer, List<String>> decommissionedNodes,
+      List<LocatedBlock> lbs) {
+    for (int i = 0; i < lbs.size(); i++) { // for each block
+      LocatedBlock blk = lbs.get(i);
+      DatanodeInfo[] nodes = blk.getLocations();
+      List<String> decommissionedNodeList = decommissionedNodes.get(i);
+
+      for (int j = 0; j < nodes.length; j++) { // for each replica
+        DatanodeInfo dnInfo = nodes[j];
+        LOG.info("Block Locations size={}, locs={}, j=", nodes.length,
+            dnInfo.toString(), j);
+        if (j < blkGrpWidth) {
+          Assert.assertEquals("Node shouldn't be decommissioned",
+              AdminStates.NORMAL, dnInfo.getAdminState());
+        } else {
+          // check against decommissioned list
+          Assert.assertTrue(
+              "For block " + blk.getBlock() + " decommissioned node " + dnInfo
+                  + " is not last node in list: " + j + "th index of "
+                  + nodes.length,
+              decommissionedNodeList.contains(dnInfo.getXferAddr()));
+          Assert.assertEquals("Node should be decommissioned",
+              AdminStates.DECOMMISSIONED, dnInfo.getAdminState());
+        }
+      }
+    }
+  }
+
+  private List<LocatedBlock> createLocatedStripedBlocks(int blkGrpCount,
+      int dataNumBlk, int numParityBlk, List<Integer> decommnNodeIndices,
+      List<Integer> targetNodeIndices,
+      HashMap<Integer, List<String>> decommissionedNodes) {
+
+    final List<LocatedBlock> lbs = new ArrayList<>(blkGrpCount);
+    for (int i = 0; i < blkGrpCount; i++) {
+      ArrayList<String> decommNodeInfo = new ArrayList<String>();
+      decommissionedNodes.put(new Integer(i), decommNodeInfo);
+      List<Integer> dummyDecommnNodeIndices = new ArrayList<>();
+      dummyDecommnNodeIndices.addAll(decommnNodeIndices);
+
+      LocatedStripedBlock lsb = createEachLocatedBlock(dataNumBlk, numParityBlk,
+          dummyDecommnNodeIndices, targetNodeIndices, decommNodeInfo);
+      lbs.add(lsb);
+    }
+    return lbs;
+  }
+
+  private LocatedStripedBlock createEachLocatedBlock(int numDataBlk,
+      int numParityBlk, List<Integer> decommnNodeIndices,
+      List<Integer> targetNodeIndices, ArrayList<String> decommNodeInfo) {
+    final long blockGroupID = Long.MIN_VALUE;
+    int totalDns = numDataBlk + numParityBlk + targetNodeIndices.size();
+    DatanodeInfo[] locs = new DatanodeInfo[totalDns];
+    String[] storageIDs = new String[totalDns];
+    StorageType[] storageTypes = new StorageType[totalDns];
+    byte[] blkIndices = new byte[totalDns];
+
+    // Adding data blocks
+    int index = 0;
+    for (; index < numDataBlk; index++) {
+      blkIndices[index] = (byte) index;
+      // Location port always equal to logical index of a block,
+      // for easier verification
+      locs[index] = DFSTestUtil.getLocalDatanodeInfo(blkIndices[index]);
+      locs[index].setLastUpdateMonotonic(Time.monotonicNow());
+      storageIDs[index] = locs[index].getDatanodeUuid();
+      storageTypes[index] = StorageType.DISK;
+      // set decommissioned state
+      if (decommnNodeIndices.contains(index)) {
+        locs[index].setDecommissioned();
+        decommNodeInfo.add(locs[index].toString());
+        // Removing it from the list to ensure that all the given nodes are
+        // successfully marked as decomissioned.
+        decommnNodeIndices.remove(new Integer(index));
+      }
+    }
+    // Adding parity blocks after data blocks
+    index = NUM_DATA_BLOCKS;
+    for (int j = numDataBlk; j < numDataBlk + numParityBlk; j++, index++) {
+      blkIndices[j] = (byte) index;
+      // Location port always equal to logical index of a block,
+      // for easier verification
+      locs[j] = DFSTestUtil.getLocalDatanodeInfo(blkIndices[j]);
+      locs[j].setLastUpdateMonotonic(Time.monotonicNow());
+      storageIDs[j] = locs[j].getDatanodeUuid();
+      storageTypes[j] = StorageType.DISK;
+      // set decommissioned state
+      if (decommnNodeIndices.contains(index)) {
+        locs[j].setDecommissioned();
+        decommNodeInfo.add(locs[j].toString());
+        // Removing it from the list to ensure that all the given nodes are
+        // successfully marked as decomissioned.
+        decommnNodeIndices.remove(new Integer(index));
+      }
+    }
+    // Add extra target nodes to storage list after the parity blocks
+    int basePortValue = NUM_DATA_BLOCKS + NUM_PARITY_BLOCKS;
+    index = numDataBlk + numParityBlk;
+    for (int i = 0; i < targetNodeIndices.size(); i++, index++) {
+      int blkIndexPos = targetNodeIndices.get(i);
+      blkIndices[index] = (byte) blkIndexPos;
+      // Location port always equal to logical index of a block,
+      // for easier verification
+      locs[index] = DFSTestUtil.getLocalDatanodeInfo(basePortValue++);
+      locs[index].setLastUpdateMonotonic(Time.monotonicNow());
+      storageIDs[index] = locs[index].getDatanodeUuid();
+      storageTypes[index] = StorageType.DISK;
+      // set decommissioned state. This can happen, the target node is again
+      // decommissioned by administrator
+      if (decommnNodeIndices.contains(blkIndexPos)) {
+        locs[index].setDecommissioned();
+        decommNodeInfo.add(locs[index].toString());
+        // Removing it from the list to ensure that all the given nodes are
+        // successfully marked as decomissioned.
+        decommnNodeIndices.remove(new Integer(blkIndexPos));
+      }
+    }
+    return new LocatedStripedBlock(
+        new ExtendedBlock("pool", blockGroupID,
+            StripedFileTestUtil.BLOCK_STRIPED_CELL_SIZE, 1001),
+        locs, storageIDs, storageTypes, blkIndices, 0, false, null);
+  }
+
+  private static DatanodeManager mockDatanodeManager() throws IOException {
+    Configuration conf = new Configuration();
+    conf.setBoolean(
+        DFSConfigKeys.DFS_NAMENODE_AVOID_STALE_DATANODE_FOR_READ_KEY, true);
+    conf.setLong(DFSConfigKeys.DFS_NAMENODE_STALE_DATANODE_INTERVAL_KEY,
+        STALE_INTERVAL);
+    FSNamesystem fsn = Mockito.mock(FSNamesystem.class);
+    BlockManager bm = Mockito.mock(BlockManager.class);
+    BlockReportLeaseManager blm = new BlockReportLeaseManager(conf);
+    Mockito.when(bm.getBlockReportLeaseManager()).thenReturn(blm);
+    DatanodeManager dm = new DatanodeManager(bm, fsn, conf);
+    return dm;
+  }
+
+  private void prepareBlockIndexAndTokenList(List<LocatedBlock> lbs,
+      List<HashMap<DatanodeInfo, Byte>> locToIndexList,
+      List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList) {
+    for (LocatedBlock lb : lbs) {
+      HashMap<DatanodeInfo, Byte> locToIndex = new HashMap<DatanodeInfo, Byte>();
+      locToIndexList.add(locToIndex);
+
+      HashMap<DatanodeInfo, Token<BlockTokenIdentifier>> locToToken =
+          new HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>();
+      locToTokenList.add(locToToken);
+
+      DatanodeInfo[] di = lb.getLocations();
+      LocatedStripedBlock stripedBlk = (LocatedStripedBlock) lb;
+      for (int i = 0; i < di.length; i++) {
+        locToIndex.put(di[i], stripedBlk.getBlockIndices()[i]);
+        locToToken.put(di[i], stripedBlk.getBlockTokens()[i]);
+      }
+    }
+  }
+
+  /**
+   * Verify block index and token values. Must update block indices and block
+   * tokens after sorting.
+   */
+  private void assertBlockIndexAndTokenPosition(List<LocatedBlock> lbs,
+      List<HashMap<DatanodeInfo, Byte>> locToIndexList,
+      List<HashMap<DatanodeInfo, Token<BlockTokenIdentifier>>> locToTokenList) {
+    for (int i = 0; i < lbs.size(); i++) {
+      LocatedBlock lb = lbs.get(i);
+      LocatedStripedBlock stripedBlk = (LocatedStripedBlock) lb;
+      HashMap<DatanodeInfo, Byte> locToIndex = locToIndexList.get(i);
+      HashMap<DatanodeInfo, Token<BlockTokenIdentifier>> locToToken =
+          locToTokenList.get(i);
+      DatanodeInfo[] di = lb.getLocations();
+      for (int j = 0; j < di.length; j++) {
+        Assert.assertEquals("Block index value mismatches after sorting",
+            (byte) locToIndex.get(di[j]), stripedBlk.getBlockIndices()[j]);
+        Assert.assertEquals("Block token value mismatches after sorting",
+            locToToken.get(di[j]), stripedBlk.getBlockTokens()[j]);
+      }
+    }
+  }
+}

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/datanode/TestBlockReplacement.java

@@ -81,7 +81,6 @@ public class TestBlockReplacement {
     long bytesToSend = TOTAL_BYTES; 
     long start = Time.monotonicNow();
     DataTransferThrottler throttler = new DataTransferThrottler(bandwidthPerSec);
-    long totalBytes = 0L;
     long bytesSent = 1024*512L; // 0.5MB
     throttler.throttle(bytesSent);
     bytesToSend -= bytesSent;
@@ -93,7 +92,7 @@ public class TestBlockReplacement {
     } catch (InterruptedException ignored) {}
     throttler.throttle(bytesToSend);
     long end = Time.monotonicNow();
-    assertTrue(totalBytes*1000/(end-start)<=bandwidthPerSec);
+    assertTrue(TOTAL_BYTES * 1000 / (end - start) <= bandwidthPerSec);
   }
   
   @Test

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/datanode/TestDataNodeVolumeFailureToleration.java

@@ -34,6 +34,8 @@ import org.apache.hadoop.hdfs.DFSTestUtil;
 import org.apache.hadoop.hdfs.HdfsConfiguration;
 import org.apache.hadoop.hdfs.MiniDFSCluster;
 import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeManager;
+import org.apache.hadoop.test.GenericTestUtils;
+import org.apache.hadoop.util.DiskChecker.DiskErrorException;
 import org.junit.After;
 import org.junit.Before;
 import org.junit.Test;
@@ -229,9 +231,22 @@ public class TestDataNodeVolumeFailureToleration {
         prepareDirToFail(dirs[i]);
       }
       restartDatanodes(volumesTolerated, manageDfsDirs);
-      assertEquals(expectedBPServiceState, cluster.getDataNodes().get(0)
-          .isBPServiceAlive(cluster.getNamesystem().getBlockPoolId()));
+    } catch (DiskErrorException e) {
+      GenericTestUtils.assertExceptionContains("Invalid value configured for "
+          + "dfs.datanode.failed.volumes.tolerated", e);
     } finally {
+      boolean bpServiceState;
+      // If the datanode not registered successfully,
+      // because the invalid value configured for tolerated volumes
+      if (cluster.getDataNodes().size() == 0) {
+        bpServiceState = false;
+      } else {
+        bpServiceState =
+            cluster.getDataNodes().get(0)
+                    .isBPServiceAlive(cluster.getNamesystem().getBlockPoolId());
+      }
+      assertEquals(expectedBPServiceState, bpServiceState);
+
       for (File dir : dirs) {
         FileUtil.chmod(dir.toString(), "755");
       }

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/TestSpaceReservation.java

@@ -572,4 +572,64 @@ public class TestSpaceReservation {
       return numFailures;
     }
   }
+
+  @Test(timeout = 30000)
+  public void testReservedSpaceForAppend() throws Exception {
+    final short replication = 3;
+    startCluster(BLOCK_SIZE, replication, -1);
+    final String methodName = GenericTestUtils.getMethodName();
+    final Path file = new Path("/" + methodName + ".01.dat");
+
+    // Write 1 byte to the file and kill the writer.
+    FSDataOutputStream os = fs.create(file, replication);
+    os.write(new byte[1024]);
+    os.close();
+
+    final Path file2 = new Path("/" + methodName + ".02.dat");
+
+    // Write 1 byte to the file and keep it open.
+    FSDataOutputStream os2 = fs.create(file2, replication);
+    os2.write(new byte[1]);
+    os2.hflush();
+    int expectedFile2Reserved = BLOCK_SIZE - 1;
+    checkReservedSpace(expectedFile2Reserved);
+
+    // append one byte and verify reservedspace before and after closing
+    os = fs.append(file);
+    os.write(new byte[1]);
+    os.hflush();
+    int expectedFile1Reserved = BLOCK_SIZE - 1025;
+    checkReservedSpace(expectedFile2Reserved + expectedFile1Reserved);
+    os.close();
+    checkReservedSpace(expectedFile2Reserved);
+
+    // append one byte and verify reservedspace before and after abort
+    os = fs.append(file);
+    os.write(new byte[1]);
+    os.hflush();
+    expectedFile1Reserved--;
+    checkReservedSpace(expectedFile2Reserved + expectedFile1Reserved);
+    DFSTestUtil.abortStream(((DFSOutputStream) os.getWrappedStream()));
+    checkReservedSpace(expectedFile2Reserved);
+  }
+
+  private void checkReservedSpace(final long expectedReserved) throws TimeoutException,
+      InterruptedException, IOException {
+    for (final DataNode dn : cluster.getDataNodes()) {
+      try (FsDatasetSpi.FsVolumeReferences volumes = dn.getFSDataset()
+          .getFsVolumeReferences()) {
+        final FsVolumeImpl volume = (FsVolumeImpl) volumes.get(0);
+        GenericTestUtils.waitFor(new Supplier<Boolean>() {
+          @Override
+          public Boolean get() {
+            LOG.info(
+                "dn " + dn.getDisplayName() + " space : " + volume
+                    .getReservedForReplicas() + ", Expected ReservedSpace :"
+                    + expectedReserved);
+            return (volume.getReservedForReplicas() == expectedReserved);
+          }
+        }, 100, 3000);
+      }
+    }
+  }
 }

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNameNodeReconfigure.java

@@ -34,6 +34,8 @@ import org.apache.hadoop.hdfs.MiniDFSCluster;
 import org.apache.hadoop.hdfs.HdfsConfiguration;
 import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeManager;
 
+import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.HADOOP_CALLER_CONTEXT_ENABLED_KEY;
+import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.HADOOP_CALLER_CONTEXT_ENABLED_DEFAULT;
 import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_HEARTBEAT_INTERVAL_KEY;
 import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_HEARTBEAT_INTERVAL_DEFAULT;
 import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_HEARTBEAT_RECHECK_INTERVAL_KEY;
@@ -50,13 +52,60 @@ public class TestNameNodeReconfigure {
   public void setUp() throws IOException {
     Configuration conf = new HdfsConfiguration();
     cluster = new MiniDFSCluster.Builder(conf).build();
+    cluster.waitActive();
+  }
+
+  @Test
+  public void testReconfigureCallerContextEnabled()
+      throws ReconfigurationException {
+    final NameNode nameNode = cluster.getNameNode();
+    final FSNamesystem nameSystem = nameNode.getNamesystem();
+
+    // try invalid values
+    nameNode.reconfigureProperty(HADOOP_CALLER_CONTEXT_ENABLED_KEY, "text");
+    assertEquals(HADOOP_CALLER_CONTEXT_ENABLED_KEY + " has wrong value", false,
+        nameSystem.getCallerContextEnabled());
+    assertEquals(
+        HADOOP_CALLER_CONTEXT_ENABLED_KEY + " has wrong value",
+        false,
+        nameNode.getConf().getBoolean(HADOOP_CALLER_CONTEXT_ENABLED_KEY,
+            HADOOP_CALLER_CONTEXT_ENABLED_DEFAULT));
+
+    // enable CallerContext
+    nameNode.reconfigureProperty(HADOOP_CALLER_CONTEXT_ENABLED_KEY, "true");
+    assertEquals(HADOOP_CALLER_CONTEXT_ENABLED_KEY + " has wrong value", true,
+        nameSystem.getCallerContextEnabled());
+    assertEquals(
+        HADOOP_CALLER_CONTEXT_ENABLED_KEY + " has wrong value",
+        true,
+        nameNode.getConf().getBoolean(HADOOP_CALLER_CONTEXT_ENABLED_KEY,
+            HADOOP_CALLER_CONTEXT_ENABLED_DEFAULT));
+
+    // disable CallerContext
+    nameNode.reconfigureProperty(HADOOP_CALLER_CONTEXT_ENABLED_KEY, "false");
+    assertEquals(HADOOP_CALLER_CONTEXT_ENABLED_KEY + " has wrong value", false,
+        nameSystem.getCallerContextEnabled());
+    assertEquals(
+        HADOOP_CALLER_CONTEXT_ENABLED_KEY + " has wrong value",
+        false,
+        nameNode.getConf().getBoolean(HADOOP_CALLER_CONTEXT_ENABLED_KEY,
+            HADOOP_CALLER_CONTEXT_ENABLED_DEFAULT));
+
+    // revert to default
+    nameNode.reconfigureProperty(HADOOP_CALLER_CONTEXT_ENABLED_KEY, null);
+
+    // verify default
+    assertEquals(HADOOP_CALLER_CONTEXT_ENABLED_KEY + " has wrong value", false,
+        nameSystem.getCallerContextEnabled());
+    assertEquals(HADOOP_CALLER_CONTEXT_ENABLED_KEY + " has wrong value", null,
+        nameNode.getConf().get(HADOOP_CALLER_CONTEXT_ENABLED_KEY));
   }
 
   /**
    * Test that we can modify configuration properties.
    */
   @Test
-  public void testReconfigure() throws ReconfigurationException {
+  public void testReconfigureHearbeatCheck1() throws ReconfigurationException {
     final NameNode nameNode = cluster.getNameNode();
     final DatanodeManager datanodeManager = nameNode.namesystem
         .getBlockManager().getDatanodeManager();

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

@@ -45,7 +45,67 @@ public class TestHdfsConfigFields extends TestConfigurationFieldsBase {
 
     // Set error modes
     errorIfMissingConfigProps = true;
-    errorIfMissingXmlProps = false;
+    errorIfMissingXmlProps = true;
+
+    // Initialize used variables
+    configurationPropsToSkipCompare = new HashSet<String>();
+
+    // Ignore testing based parameter
+    configurationPropsToSkipCompare.add("ignore.secure.ports.for.testing");
+
+    // Remove deprecated properties listed in Configuration#DeprecationDelta
+    configurationPropsToSkipCompare.add(DFSConfigKeys.DFS_DF_INTERVAL_KEY);
+
+    // Remove default properties
+    configurationPropsToSkipCompare
+        .add(DFSConfigKeys.DFS_IMAGE_COMPRESSION_CODEC_DEFAULT);
+    configurationPropsToSkipCompare
+        .add(DFSConfigKeys.DFS_WEBHDFS_AUTHENTICATION_FILTER_DEFAULT);
+
+    // Remove support property
+    configurationPropsToSkipCompare
+        .add(DFSConfigKeys.DFS_NAMENODE_MIN_SUPPORTED_DATANODE_VERSION_KEY);
+    configurationPropsToSkipCompare
+        .add(DFSConfigKeys.DFS_DATANODE_MIN_SUPPORTED_NAMENODE_VERSION_KEY);
+
+    // Purposely hidden, based on comments in DFSConfigKeys
+    configurationPropsToSkipCompare
+        .add(DFSConfigKeys.DFS_DATANODE_XCEIVER_STOP_TIMEOUT_MILLIS_KEY);
+
+    // Fully deprecated properties?
+    configurationPropsToSkipCompare
+        .add("dfs.corruptfilesreturned.max");
+    configurationPropsToSkipCompare
+        .add("dfs.datanode.hdfs-blocks-metadata.enabled");
+    configurationPropsToSkipCompare
+        .add("dfs.metrics.session-id");
+    configurationPropsToSkipCompare
+        .add("dfs.datanode.synconclose");
+    configurationPropsToSkipCompare
+        .add("dfs.datanode.non.local.lazy.persist");
+    configurationPropsToSkipCompare
+        .add("dfs.namenode.tolerate.heartbeat.multiplier");
+    configurationPropsToSkipCompare
+        .add("dfs.namenode.stripe.min");
+    configurationPropsToSkipCompare
+        .add("dfs.namenode.replqueue.threshold-pct");
+
+    // Removed by HDFS-6440
+    configurationPropsToSkipCompare
+        .add("dfs.ha.log-roll.rpc.timeout");
+
+    // Example (not real) property in hdfs-default.xml
+    configurationPropsToSkipCompare.add("dfs.ha.namenodes");
+
+    // Property used for internal testing only
+    configurationPropsToSkipCompare
+        .add(DFSConfigKeys.DFS_DATANODE_DUPLICATE_REPLICA_DELETION);
+
+    // Property not intended for users
+    configurationPropsToSkipCompare
+        .add(DFSConfigKeys.DFS_DATANODE_STARTUP_KEY);
+    configurationPropsToSkipCompare
+        .add(DFSConfigKeys.DFS_NAMENODE_STARTUP_KEY);
 
     // Allocate
     xmlPropsToSkipCompare = new HashSet<String>();
@@ -58,21 +118,12 @@ public class TestHdfsConfigFields extends TestConfigurationFieldsBase {
     // Used dynamically as part of DFSConfigKeys.DFS_NAMENODE_EDITS_PLUGIN_PREFIX
     xmlPropsToSkipCompare.add("dfs.namenode.edits.journal-plugin.qjournal");
 
-    // Example (not real) property in hdfs-default.xml
-    xmlPropsToSkipCompare.add("dfs.ha.namenodes.EXAMPLENAMESERVICE");
-
     // Defined in org.apache.hadoop.fs.CommonConfigurationKeys
     xmlPropsToSkipCompare.add("hadoop.user.group.metrics.percentiles.intervals");
 
     // Used oddly by DataNode to create new config String
     xmlPropsToSkipCompare.add("hadoop.hdfs.configuration.version");
 
-    // Kept in the NfsConfiguration class in the hadoop-hdfs-nfs module
-    xmlPrefixToSkipCompare.add("nfs");
-
-    // Not a hardcoded property.  Used by SaslRpcClient
-    xmlPrefixToSkipCompare.add("dfs.namenode.kerberos.principal.pattern");
-
     // Skip comparing in branch-2.  Removed in trunk with HDFS-7985.
     xmlPropsToSkipCompare.add("dfs.webhdfs.enabled");
 
@@ -82,5 +133,21 @@ public class TestHdfsConfigFields extends TestConfigurationFieldsBase {
     // Ignore HTrace properties
     xmlPropsToSkipCompare.add("fs.client.htrace");
     xmlPropsToSkipCompare.add("hadoop.htrace");
+
+    // Ignore SpanReceiveHost properties
+    xmlPropsToSkipCompare.add("dfs.htrace.spanreceiver.classes");
+    xmlPropsToSkipCompare.add("dfs.client.htrace.spanreceiver.classes");
+
+    // Remove deprecated properties listed in Configuration#DeprecationDelta
+    xmlPropsToSkipCompare.add(DFSConfigKeys.DFS_DF_INTERVAL_KEY);
+
+    // Kept in the NfsConfiguration class in the hadoop-hdfs-nfs module
+    xmlPrefixToSkipCompare.add("nfs");
+
+    // Not a hardcoded property.  Used by SaslRpcClient
+    xmlPrefixToSkipCompare.add("dfs.namenode.kerberos.principal.pattern");
+
+    // Skip over example property
+    xmlPrefixToSkipCompare.add("dfs.ha.namenodes");
   }
 }

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

@@ -111,9 +111,6 @@ public class TestJMXGet {
         jmx.getValue("NumLiveDataNodes")));
     assertGauge("CorruptBlocks", Long.parseLong(jmx.getValue("CorruptBlocks")),
                 getMetrics("FSNamesystem"));
-    DFSTestUtil.waitForMetric(jmx, "NumOpenConnections", numDatanodes);
-    assertEquals(numDatanodes, Integer.parseInt(
-        jmx.getValue("NumOpenConnections")));
 
     cluster.shutdown();
     MBeanServerConnection mbsc = ManagementFactory.getPlatformMBeanServer();

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

@@ -537,7 +537,7 @@ class Fetcher<K,V> extends Thread {
             + " len: " + compressedLength + " to " + mapOutput.getDescription());
         mapOutput.shuffle(host, is, compressedLength, decompressedLength,
             metrics, reporter);
-      } catch (java.lang.InternalError e) {
+      } catch (java.lang.InternalError | Exception e) {
         LOG.warn("Failed to shuffle for fetcher#"+id, e);
         throw new IOException(e);
       }

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

@@ -346,6 +346,43 @@ public class TestFetcher {
   
   @SuppressWarnings("unchecked")
   @Test(timeout=10000) 
+  public void testCopyFromHostOnAnyException() throws Exception {
+    InMemoryMapOutput<Text, Text> immo = mock(InMemoryMapOutput.class);
+
+    Fetcher<Text,Text> underTest = new FakeFetcher<Text,Text>(job, id, ss, mm,
+        r, metrics, except, key, connection);
+
+    String replyHash = SecureShuffleUtils.generateHash(encHash.getBytes(), key);
+
+    when(connection.getResponseCode()).thenReturn(200);
+    when(connection.getHeaderField(
+        SecureShuffleUtils.HTTP_HEADER_REPLY_URL_HASH)).thenReturn(replyHash);
+    ShuffleHeader header = new ShuffleHeader(map1ID.toString(), 10, 10, 1);
+    ByteArrayOutputStream bout = new ByteArrayOutputStream();
+    header.write(new DataOutputStream(bout));
+    ByteArrayInputStream in = new ByteArrayInputStream(bout.toByteArray());
+    when(connection.getInputStream()).thenReturn(in);
+    when(connection.getHeaderField(ShuffleHeader.HTTP_HEADER_NAME))
+        .thenReturn(ShuffleHeader.DEFAULT_HTTP_HEADER_NAME);
+    when(connection.getHeaderField(ShuffleHeader.HTTP_HEADER_VERSION))
+        .thenReturn(ShuffleHeader.DEFAULT_HTTP_HEADER_VERSION);
+    when(mm.reserve(any(TaskAttemptID.class), anyLong(), anyInt()))
+        .thenReturn(immo);
+
+    doThrow(new ArrayIndexOutOfBoundsException()).when(immo)
+        .shuffle(any(MapHost.class), any(InputStream.class), anyLong(),
+            anyLong(), any(ShuffleClientMetrics.class), any(Reporter.class));
+
+    underTest.copyFromHost(host);
+
+    verify(connection)
+        .addRequestProperty(SecureShuffleUtils.HTTP_HEADER_URL_HASH,
+          encHash);
+    verify(ss, times(1)).copyFailed(map1ID, host, true, false);
+  }
+
+  @SuppressWarnings("unchecked")
+  @Test(timeout=10000)
   public void testCopyFromHostWithRetry() throws Exception {
     InMemoryMapOutput<Text, Text> immo = mock(InMemoryMapOutput.class);
     ss = mock(ShuffleSchedulerImpl.class);

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AInputStream.java

@@ -20,7 +20,6 @@ package org.apache.hadoop.fs.s3a;
 
 import com.amazonaws.services.s3.AmazonS3Client;
 import com.amazonaws.services.s3.model.GetObjectRequest;
-import com.amazonaws.services.s3.model.S3Object;
 import com.amazonaws.services.s3.model.S3ObjectInputStream;
 import org.apache.hadoop.fs.FSExceptionMessages;
 import org.apache.hadoop.fs.FSInputStream;
@@ -37,82 +36,128 @@ public class S3AInputStream extends FSInputStream {
   private long pos;
   private boolean closed;
   private S3ObjectInputStream wrappedStream;
-  private FileSystem.Statistics stats;
-  private AmazonS3Client client;
-  private String bucket;
-  private String key;
-  private long contentLength;
+  private final FileSystem.Statistics stats;
+  private final AmazonS3Client client;
+  private final String bucket;
+  private final String key;
+  private final long contentLength;
+  private final String uri;
   public static final Logger LOG = S3AFileSystem.LOG;
   public static final long CLOSE_THRESHOLD = 4096;
 
-  public S3AInputStream(String bucket, String key, long contentLength, AmazonS3Client client,
-                        FileSystem.Statistics stats) {
+  // Used by lazy seek
+  private long nextReadPos;
+
+  //Amount of data requested from the request
+  private long requestedStreamLen;
+
+  public S3AInputStream(String bucket, String key, long contentLength,
+      AmazonS3Client client, FileSystem.Statistics stats) {
     this.bucket = bucket;
     this.key = key;
     this.contentLength = contentLength;
     this.client = client;
     this.stats = stats;
     this.pos = 0;
+    this.nextReadPos = 0;
     this.closed = false;
     this.wrappedStream = null;
+    this.uri = "s3a://" + this.bucket + "/" + this.key;
   }
 
-  private void openIfNeeded() throws IOException {
-    if (wrappedStream == null) {
-      reopen(0);
-    }
-  }
-
-  private synchronized void reopen(long pos) throws IOException {
+  /**
+   * Opens up the stream at specified target position and for given length.
+   *
+   * @param targetPos target position
+   * @param length length requested
+   * @throws IOException
+   */
+  private synchronized void reopen(long targetPos, long length)
+      throws IOException {
+    requestedStreamLen = (length < 0) ? this.contentLength :
+        Math.max(this.contentLength, (CLOSE_THRESHOLD + (targetPos + length)));
 
     if (wrappedStream != null) {
       if (LOG.isDebugEnabled()) {
-        LOG.debug("Aborting old stream to open at pos " + pos);
+        LOG.debug("Closing the previous stream");
       }
-      wrappedStream.abort();
-    }
-
-    if (pos < 0) {
-      throw new EOFException(FSExceptionMessages.NEGATIVE_SEEK
-          +" " + pos);
+      closeStream(requestedStreamLen);
     }
 
-    if (contentLength > 0 && pos > contentLength-1) {
-      throw new EOFException(
-          FSExceptionMessages.CANNOT_SEEK_PAST_EOF
-          + " " + pos);
+    if (LOG.isDebugEnabled()) {
+      LOG.debug("Requesting for "
+          + "targetPos=" + targetPos
+          + ", length=" + length
+          + ", requestedStreamLen=" + requestedStreamLen
+          + ", streamPosition=" + pos
+          + ", nextReadPosition=" + nextReadPos
+      );
     }
 
-    LOG.debug("Actually opening file " + key + " at pos " + pos);
-
-    GetObjectRequest request = new GetObjectRequest(bucket, key);
-    request.setRange(pos, contentLength-1);
-
+    GetObjectRequest request = new GetObjectRequest(bucket, key)
+        .withRange(targetPos, requestedStreamLen);
     wrappedStream = client.getObject(request).getObjectContent();
 
     if (wrappedStream == null) {
       throw new IOException("Null IO stream");
     }
 
-    this.pos = pos;
+    this.pos = targetPos;
   }
 
   @Override
   public synchronized long getPos() throws IOException {
-    return pos;
+    return (nextReadPos < 0) ? 0 : nextReadPos;
   }
 
   @Override
-  public synchronized void seek(long pos) throws IOException {
+  public synchronized void seek(long targetPos) throws IOException {
     checkNotClosed();
 
-    if (this.pos == pos) {
+    // Do not allow negative seek
+    if (targetPos < 0) {
+      throw new EOFException(FSExceptionMessages.NEGATIVE_SEEK
+          + " " + targetPos);
+    }
+
+    if (this.contentLength <= 0) {
       return;
     }
 
-    LOG.debug(
-        "Reopening " + this.key + " to seek to new offset " + (pos - this.pos));
-    reopen(pos);
+    // Lazy seek
+    nextReadPos = targetPos;
+  }
+
+  /**
+   * Adjust the stream to a specific position.
+   *
+   * @param targetPos target seek position
+   * @param length length of content that needs to be read from targetPos
+   * @throws IOException
+   */
+  private void seekInStream(long targetPos, long length) throws IOException {
+    checkNotClosed();
+    if (wrappedStream == null) {
+      return;
+    }
+
+    // compute how much more to skip
+    long diff = targetPos - pos;
+    if (targetPos > pos) {
+      if ((diff + length) <= wrappedStream.available()) {
+        // already available in buffer
+        pos += wrappedStream.skip(diff);
+        if (pos != targetPos) {
+          throw new IOException("Failed to seek to " + targetPos
+              + ". Current position " + pos);
+        }
+        return;
+      }
+    }
+
+    // close the stream; if read the object will be opened at the new pos
+    closeStream(this.requestedStreamLen);
+    pos = targetPos;
   }
 
   @Override
@@ -120,27 +165,48 @@ public class S3AInputStream extends FSInputStream {
     return false;
   }
 
+  /**
+   * Perform lazy seek and adjust stream to correct position for reading.
+   *
+   * @param targetPos position from where data should be read
+   * @param len length of the content that needs to be read
+   */
+  private void lazySeek(long targetPos, long len) throws IOException {
+    //For lazy seek
+    if (targetPos != this.pos) {
+      seekInStream(targetPos, len);
+    }
+
+    //re-open at specific location if needed
+    if (wrappedStream == null) {
+      reopen(targetPos, len);
+    }
+  }
+
   @Override
   public synchronized int read() throws IOException {
     checkNotClosed();
+    if (this.contentLength == 0 || (nextReadPos >= contentLength)) {
+      return -1;
+    }
 
-    openIfNeeded();
+    lazySeek(nextReadPos, 1);
 
     int byteRead;
     try {
       byteRead = wrappedStream.read();
-    } catch (SocketTimeoutException e) {
-      LOG.info("Got timeout while trying to read from stream, trying to recover " + e);
-      reopen(pos);
-      byteRead = wrappedStream.read();
-    } catch (SocketException e) {
-      LOG.info("Got socket exception while trying to read from stream, trying to recover " + e);
-      reopen(pos);
+    } catch (SocketTimeoutException | SocketException e) {
+      LOG.info("Got exception while trying to read from stream,"
+          + " trying to recover " + e);
+      reopen(pos, 1);
       byteRead = wrappedStream.read();
+    } catch (EOFException e) {
+      return -1;
     }
 
     if (byteRead >= 0) {
       pos++;
+      nextReadPos++;
     }
 
     if (stats != null && byteRead >= 0) {
@@ -150,26 +216,34 @@ public class S3AInputStream extends FSInputStream {
   }
 
   @Override
-  public synchronized int read(byte[] buf, int off, int len) throws IOException {
+  public synchronized int read(byte[] buf, int off, int len)
+      throws IOException {
     checkNotClosed();
 
-    openIfNeeded();
+    validatePositionedReadArgs(nextReadPos, buf, off, len);
+    if (len == 0) {
+      return 0;
+    }
+
+    if (this.contentLength == 0 || (nextReadPos >= contentLength)) {
+      return -1;
+    }
+
+    lazySeek(nextReadPos, len);
 
     int byteRead;
     try {
       byteRead = wrappedStream.read(buf, off, len);
-    } catch (SocketTimeoutException e) {
-      LOG.info("Got timeout while trying to read from stream, trying to recover " + e);
-      reopen(pos);
-      byteRead = wrappedStream.read(buf, off, len);
-    } catch (SocketException e) {
-      LOG.info("Got socket exception while trying to read from stream, trying to recover " + e);
-      reopen(pos);
+    } catch (SocketTimeoutException | SocketException e) {
+      LOG.info("Got exception while trying to read from stream,"
+          + " trying to recover " + e);
+      reopen(pos, len);
       byteRead = wrappedStream.read(buf, off, len);
     }
 
     if (byteRead > 0) {
       pos += byteRead;
+      nextReadPos += byteRead;
     }
 
     if (stats != null && byteRead > 0) {
@@ -189,15 +263,43 @@ public class S3AInputStream extends FSInputStream {
   public synchronized void close() throws IOException {
     super.close();
     closed = true;
+    closeStream(this.contentLength);
+  }
+
+  /**
+   * Close a stream: decide whether to abort or close, based on
+   * the length of the stream and the current position.
+   *
+   * This does not set the {@link #closed} flag.
+   * @param length length of the stream.
+   * @throws IOException
+   */
+  private void closeStream(long length) throws IOException {
     if (wrappedStream != null) {
-      if (contentLength - pos <= CLOSE_THRESHOLD) {
-        // Close, rather than abort, so that the http connection can be reused.
-        wrappedStream.close();
-      } else {
+      String reason = null;
+      boolean shouldAbort = length - pos > CLOSE_THRESHOLD;
+      if (!shouldAbort) {
+        try {
+          reason = "Closed stream";
+          wrappedStream.close();
+        } catch (IOException e) {
+          // exception escalates to an abort
+          LOG.debug("When closing stream", e);
+          shouldAbort = true;
+        }
+      }
+      if (shouldAbort) {
         // Abort, rather than just close, the underlying stream.  Otherwise, the
         // remaining object payload is read from S3 while closing the stream.
         wrappedStream.abort();
+        reason = "Closed stream with abort";
       }
+      if (LOG.isDebugEnabled()) {
+        LOG.debug(reason + "; streamPos=" + pos
+            + ", nextReadPos=" + nextReadPos
+            + ", contentLength=" + length);
+      }
+      wrappedStream = null;
     }
   }
 
@@ -216,4 +318,55 @@ public class S3AInputStream extends FSInputStream {
   public boolean markSupported() {
     return false;
   }
+
+  @Override
+  public String toString() {
+    final StringBuilder sb = new StringBuilder(
+        "S3AInputStream{");
+    sb.append(uri);
+    sb.append(" pos=").append(pos);
+    sb.append(" nextReadPos=").append(nextReadPos);
+    sb.append(" contentLength=").append(contentLength);
+    sb.append('}');
+    return sb.toString();
+  }
+
+  /**
+   * Subclass {@code readFully()} operation which only seeks at the start
+   * of the series of operations; seeking back at the end.
+   *
+   * This is significantly higher performance if multiple read attempts are
+   * needed to fetch the data, as it does not break the HTTP connection.
+   *
+   * To maintain thread safety requirements, this operation is synchronized
+   * for the duration of the sequence.
+   * {@inheritDoc}
+   *
+   */
+  @Override
+  public void readFully(long position, byte[] buffer, int offset, int length)
+      throws IOException {
+    checkNotClosed();
+    validatePositionedReadArgs(position, buffer, offset, length);
+    if (length == 0) {
+      return;
+    }
+    int nread = 0;
+    synchronized (this) {
+      long oldPos = getPos();
+      try {
+        seek(position);
+        while (nread < length) {
+          int nbytes = read(buffer, offset + nread, length - nread);
+          if (nbytes < 0) {
+            throw new EOFException(FSExceptionMessages.EOF_IN_READ_FULLY);
+          }
+          nread += nbytes;
+        }
+
+      } finally {
+        seek(oldPos);
+      }
+    }
+  }
 }

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/TestS3AConfiguration.java

@@ -123,6 +123,7 @@ public class TestS3AConfiguration {
   @Test
   public void testProxyPortWithoutHost() throws Exception {
     conf = new Configuration();
+    conf.unset(Constants.PROXY_HOST);
     conf.setInt(Constants.MAX_ERROR_RETRIES, 2);
     conf.setInt(Constants.PROXY_PORT, 1);
     try {
@@ -140,6 +141,7 @@ public class TestS3AConfiguration {
   @Test
   public void testAutomaticProxyPortSelection() throws Exception {
     conf = new Configuration();
+    conf.unset(Constants.PROXY_PORT);
     conf.setInt(Constants.MAX_ERROR_RETRIES, 2);
     conf.set(Constants.PROXY_HOST, "127.0.0.1");
     conf.set(Constants.SECURE_CONNECTIONS, "true");

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/scale/TestS3ADeleteManyFiles.java

@@ -44,6 +44,15 @@ public class TestS3ADeleteManyFiles extends S3AScaleTestBase {
   @Rule
   public Timeout testTimeout = new Timeout(30 * 60 * 1000);
 
+  /**
+   * CAUTION: If this test starts failing, please make sure that the
+   * {@link org.apache.hadoop.fs.s3a.Constants#MAX_THREADS} configuration is not
+   * set too low. Alternatively, consider reducing the
+   * <code>scale.test.operation.count</code> parameter in
+   * <code>getOperationCount()</code>.
+   *
+   * @see #getOperationCount()
+   */
   @Test
   public void testBulkRenameAndDelete() throws Throwable {
     final Path scaleTestDir = getTestPath();

hadoop-tools/hadoop-azure/src/main/java/org/apache/hadoop/fs/azure/NativeAzureFileSystem.java

@@ -725,6 +725,8 @@ public class NativeAzureFileSystem extends FileSystem {
       // Return to the caller with the result.
       //
         return result;
+      } catch(EOFException e) {
+        return -1;
       } catch(IOException e) {
 
         Throwable innerException = NativeAzureFileSystemHelper.checkForAzureStorageException(e);
@@ -773,7 +775,7 @@ public class NativeAzureFileSystem extends FileSystem {
           pos += result;
         }
 
-        if (null != statistics) {
+        if (null != statistics && result > 0) {
           statistics.incrementBytesRead(result);
         }
 

hadoop-tools/hadoop-sls/src/main/data/2jobs2min-rumen-jh.json

@@ -4701,7 +4701,6 @@
     "dfs.namenode.avoid.read.stale.datanode" : "false",
     "mapreduce.job.reduces" : "0",
     "mapreduce.map.sort.spill.percent" : "0.8",
-    "dfs.client.file-block-storage-locations.timeout" : "60",
     "dfs.datanode.drop.cache.behind.writes" : "false",
     "mapreduce.job.end-notification.retry.interval" : "1",
     "mapreduce.job.maps" : "96",
@@ -4800,7 +4799,6 @@
     "dfs.datanode.directoryscan.interval" : "21600",
     "yarn.resourcemanager.address" : "a2115.smile.com:8032",
     "yarn.nodemanager.health-checker.interval-ms" : "600000",
-    "dfs.client.file-block-storage-locations.num-threads" : "10",
     "yarn.resourcemanager.container-tokens.master-key-rolling-interval-secs" : "86400",
     "mapreduce.reduce.markreset.buffer.percent" : "0.0",
     "hadoop.security.group.mapping.ldap.directory.search.timeout" : "10000",
@@ -9806,7 +9804,6 @@
     "dfs.namenode.avoid.read.stale.datanode" : "false",
     "mapreduce.job.reduces" : "0",
     "mapreduce.map.sort.spill.percent" : "0.8",
-    "dfs.client.file-block-storage-locations.timeout" : "60",
     "dfs.datanode.drop.cache.behind.writes" : "false",
     "mapreduce.job.end-notification.retry.interval" : "1",
     "mapreduce.job.maps" : "96",
@@ -9905,7 +9902,6 @@
     "dfs.datanode.directoryscan.interval" : "21600",
     "yarn.resourcemanager.address" : "a2115.smile.com:8032",
     "yarn.nodemanager.health-checker.interval-ms" : "600000",
-    "dfs.client.file-block-storage-locations.num-threads" : "10",
     "yarn.resourcemanager.container-tokens.master-key-rolling-interval-secs" : "86400",
     "mapreduce.reduce.markreset.buffer.percent" : "0.0",
     "hadoop.security.group.mapping.ldap.directory.search.timeout" : "10000",
@@ -10412,7 +10408,6 @@
 "dfs.namenode.avoid.read.stale.datanode" : "false",
 "mapreduce.job.reduces" : "0",
 "mapreduce.map.sort.spill.percent" : "0.8",
-"dfs.client.file-block-storage-locations.timeout" : "60",
 "dfs.datanode.drop.cache.behind.writes" : "false",
 "mapreduce.job.end-notification.retry.interval" : "1",
 "mapreduce.job.maps" : "96",
@@ -10511,7 +10506,6 @@
 "dfs.datanode.directoryscan.interval" : "21600",
 "yarn.resourcemanager.address" : "a2115.smile.com:8032",
 "yarn.nodemanager.health-checker.interval-ms" : "600000",
-"dfs.client.file-block-storage-locations.num-threads" : "10",
 "yarn.resourcemanager.container-tokens.master-key-rolling-interval-secs" : "86400",
 "mapreduce.reduce.markreset.buffer.percent" : "0.0",
 "hadoop.security.group.mapping.ldap.directory.search.timeout" : "10000",

hadoop-tools/hadoop-sls/src/main/java/org/apache/hadoop/yarn/sls/nodemanager/NodeInfo.java

@@ -199,15 +199,6 @@ public class NodeInfo {
     public ResourceUtilization getNodeUtilization() {
       return null;
     }
-
-    @Override
-    public long getUntrackedTimeStamp() {
-      return 0;
-    }
-
-    @Override
-    public void setUntrackedTimeStamp(long timeStamp) {
-    }
   }
 
   public static RMNode newNodeInfo(String rackName, String hostName,

hadoop-tools/hadoop-sls/src/main/java/org/apache/hadoop/yarn/sls/scheduler/RMNodeWrapper.java

@@ -188,13 +188,4 @@ public class RMNodeWrapper implements RMNode {
   public ResourceUtilization getNodeUtilization() {
     return node.getNodeUtilization();
   }
-
-  @Override
-  public long getUntrackedTimeStamp() {
-    return 0;
-  }
-
-  @Override
-  public void setUntrackedTimeStamp(long timeStamp) {
-  }
 }

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/api/records/ContainerId.java

@@ -165,13 +165,12 @@ public abstract class ContainerId implements Comparable<ContainerId>{
 
   @Override
   public int compareTo(ContainerId other) {
-    if (this.getApplicationAttemptId().compareTo(
-        other.getApplicationAttemptId()) == 0) {
-      return Long.valueOf(getContainerId())
-          .compareTo(Long.valueOf(other.getContainerId()));
+    int result = this.getApplicationAttemptId().compareTo(
+        other.getApplicationAttemptId());
+    if (result == 0) {
+      return Long.compare(getContainerId(), other.getContainerId());
     } else {
-      return this.getApplicationAttemptId().compareTo(
-          other.getApplicationAttemptId());
+      return result;
     }
   }
 

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/conf/YarnConfiguration.java

@@ -647,15 +647,6 @@ public class YarnConfiguration extends Configuration {
   public static final String DEFAULT_RM_NODEMANAGER_MINIMUM_VERSION =
       "NONE";
 
-  /**
-   * Timeout(msec) for an untracked node to remain in shutdown or decommissioned
-   * state.
-   */
-  public static final String RM_NODEMANAGER_UNTRACKED_REMOVAL_TIMEOUT_MSEC =
-      RM_PREFIX + "node-removal-untracked.timeout-ms";
-  public static final int
-      DEFAULT_RM_NODEMANAGER_UNTRACKED_REMOVAL_TIMEOUT_MSEC = 60000;
-
   /**
    * RM proxy users' prefix
    */

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-client/src/main/java/org/apache/hadoop/yarn/client/api/impl/NMClientImpl.java

@@ -171,8 +171,6 @@ public class NMClientImpl extends NMClient {
       throw RPCUtil.getRemoteException("Container "
           + startedContainer.containerId.toString() + " is already started");
     }
-    startedContainers
-        .put(startedContainer.getContainerId(), startedContainer);
   }
 
   @Override
@@ -182,7 +180,8 @@ public class NMClientImpl extends NMClient {
     // Do synchronization on StartedContainer to prevent race condition
     // between startContainer and stopContainer only when startContainer is
     // in progress for a given container.
-    StartedContainer startingContainer = createStartedContainer(container);
+    StartedContainer startingContainer =
+        new StartedContainer(container.getId(), container.getNodeId());
     synchronized (startingContainer) {
       addStartingContainer(startingContainer);
       
@@ -210,18 +209,14 @@ public class NMClientImpl extends NMClient {
         }
         allServiceResponse = response.getAllServicesMetaData();
         startingContainer.state = ContainerState.RUNNING;
-      } catch (YarnException e) {
+      } catch (YarnException | IOException e) {
         startingContainer.state = ContainerState.COMPLETE;
         // Remove the started container if it failed to start
-        removeStartedContainer(startingContainer);
-        throw e;
-      } catch (IOException e) {
-        startingContainer.state = ContainerState.COMPLETE;
-        removeStartedContainer(startingContainer);
+        startedContainers.remove(startingContainer.containerId);
         throw e;
       } catch (Throwable t) {
         startingContainer.state = ContainerState.COMPLETE;
-        removeStartedContainer(startingContainer);
+        startedContainers.remove(startingContainer.containerId);
         throw RPCUtil.getRemoteException(t);
       } finally {
         if (proxy != null) {
@@ -263,7 +258,7 @@ public class NMClientImpl extends NMClient {
   @Override
   public void stopContainer(ContainerId containerId, NodeId nodeId)
       throws YarnException, IOException {
-    StartedContainer startedContainer = getStartedContainer(containerId);
+    StartedContainer startedContainer = startedContainers.get(containerId);
 
     // Only allow one request of stopping the container to move forward
     // When entering the block, check whether the precursor has already stopped
@@ -276,7 +271,7 @@ public class NMClientImpl extends NMClient {
         stopContainerInternal(containerId, nodeId);
         // Only after successful
         startedContainer.state = ContainerState.COMPLETE;
-        removeStartedContainer(startedContainer);
+        startedContainers.remove(startedContainer.containerId);
       }
     } else {
       stopContainerInternal(containerId, nodeId);
@@ -333,23 +328,6 @@ public class NMClientImpl extends NMClient {
       }
     }
   }
-  
-  protected synchronized StartedContainer createStartedContainer(
-      Container container) throws YarnException, IOException {
-    StartedContainer startedContainer = new StartedContainer(container.getId(),
-        container.getNodeId());
-    return startedContainer;
-  }
-
-  protected synchronized void
-      removeStartedContainer(StartedContainer container) {
-    startedContainers.remove(container.containerId);
-  }
-
-  protected synchronized StartedContainer getStartedContainer(
-      ContainerId containerId) {
-    return startedContainers.get(containerId);
-  }
 
   public AtomicBoolean getCleanupRunningContainers() {
     return cleanupRunningContainers;

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-client/src/main/java/org/apache/hadoop/yarn/client/cli/TopCLI.java

@@ -241,7 +241,7 @@ public class TopCLI extends YarnCLI {
         @Override
         public int
             compare(ApplicationInformation a1, ApplicationInformation a2) {
-          return Long.valueOf(a1.usedMemory).compareTo(a2.usedMemory);
+          return Long.compare(a1.usedMemory, a2.usedMemory);
         }
       };
   public static final Comparator<ApplicationInformation> ReservedMemoryComparator =
@@ -249,7 +249,7 @@ public class TopCLI extends YarnCLI {
         @Override
         public int
             compare(ApplicationInformation a1, ApplicationInformation a2) {
-          return Long.valueOf(a1.reservedMemory).compareTo(a2.reservedMemory);
+          return Long.compare(a1.reservedMemory, a2.reservedMemory);
         }
       };
   public static final Comparator<ApplicationInformation> UsedVCoresComparator =
@@ -273,7 +273,7 @@ public class TopCLI extends YarnCLI {
         @Override
         public int
             compare(ApplicationInformation a1, ApplicationInformation a2) {
-          return Long.valueOf(a1.vcoreSeconds).compareTo(a2.vcoreSeconds);
+          return Long.compare(a1.vcoreSeconds, a2.vcoreSeconds);
         }
       };
   public static final Comparator<ApplicationInformation> MemorySecondsComparator =
@@ -281,7 +281,7 @@ public class TopCLI extends YarnCLI {
         @Override
         public int
             compare(ApplicationInformation a1, ApplicationInformation a2) {
-          return Long.valueOf(a1.memorySeconds).compareTo(a2.memorySeconds);
+          return Long.compare(a1.memorySeconds, a2.memorySeconds);
         }
       };
   public static final Comparator<ApplicationInformation> ProgressComparator =
@@ -297,7 +297,7 @@ public class TopCLI extends YarnCLI {
         @Override
         public int
             compare(ApplicationInformation a1, ApplicationInformation a2) {
-          return Long.valueOf(a1.runningTime).compareTo(a2.runningTime);
+          return Long.compare(a1.runningTime, a2.runningTime);
         }
       };
   public static final Comparator<ApplicationInformation> AppNameComparator =

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/java/org/apache/hadoop/yarn/webapp/WebApps.java

@@ -270,7 +270,7 @@ public class WebApps {
         }
 
         if (httpScheme.equals(WebAppUtils.HTTPS_PREFIX)) {
-          WebAppUtils.loadSslConfiguration(builder);
+          WebAppUtils.loadSslConfiguration(builder, conf);
         }
 
         HttpServer2 server = builder.build();

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/resources/webapps/static/dt-1.9.4/css/jui-dt.css

@@ -109,8 +109,8 @@ table.display thead th div.DataTables_sort_wrapper span {
 
 .dataTables_wrapper {
 	position: relative;
-	min-height: 302px;
-	_height: 302px;
+	min-height: 35px;
+	_height: 35px;
 	clear: both;
 }
 

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/resources/yarn-default.xml

@@ -2722,17 +2722,4 @@
     <name>yarn.timeline-service.webapp.rest-csrf.methods-to-ignore</name>
     <value>GET,OPTIONS,HEAD</value>
   </property>
-
-  <property>
-    <description>
-    The least amount of time(msec.) an inactive (decommissioned or shutdown) node can
-    stay in the nodes list of the resourcemanager after being declared untracked.
-    A node is marked untracked if and only if it is absent from both include and
-    exclude nodemanager lists on the RM. All inactive nodes are checked twice per
-    timeout interval or every 10 minutes, whichever is lesser, and marked appropriately.
-    The same is done when refreshNodes command (graceful or otherwise) is invoked.
-    </description>
-    <name>yarn.resourcemanager.node-removal-untracked.timeout-ms</name>
-    <value>60000</value>
-  </property>
 </configuration>

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/test/java/org/apache/hadoop/yarn/conf/TestYarnConfiguration.java

@@ -54,7 +54,7 @@ public class TestYarnConfiguration {
     String rmWebUrl = WebAppUtils.getRMWebAppURLWithScheme(conf);
     String[] parts = rmWebUrl.split(":");
     Assert.assertEquals("RM Web URL Port is incrrect", 24543,
-        Integer.valueOf(parts[parts.length - 1]).intValue());
+        Integer.parseInt(parts[parts.length - 1]));
     Assert.assertNotSame(
         "RM Web Url not resolved correctly. Should not be rmtesting",
         "http://rmtesting:24543", rmWebUrl);
@@ -178,7 +178,7 @@ public class TestYarnConfiguration {
     conf.set(YarnConfiguration.RM_RESOURCE_TRACKER_ADDRESS, "yo.yo.yo");
     serverAddress = new InetSocketAddress(
         YarnConfiguration.DEFAULT_RM_RESOURCE_TRACKER_ADDRESS.split(":")[0],
-        Integer.valueOf(YarnConfiguration.DEFAULT_RM_RESOURCE_TRACKER_ADDRESS.split(":")[1]));
+        Integer.parseInt(YarnConfiguration.DEFAULT_RM_RESOURCE_TRACKER_ADDRESS.split(":")[1]));
 
     resourceTrackerConnectAddress = conf.updateConnectAddr(
         YarnConfiguration.RM_BIND_HOST,
@@ -194,7 +194,7 @@ public class TestYarnConfiguration {
     conf.set(YarnConfiguration.RM_BIND_HOST, "0.0.0.0");
     serverAddress = new InetSocketAddress(
         YarnConfiguration.DEFAULT_RM_RESOURCE_TRACKER_ADDRESS.split(":")[0],
-        Integer.valueOf(YarnConfiguration.DEFAULT_RM_RESOURCE_TRACKER_ADDRESS.split(":")[1]));
+        Integer.parseInt(YarnConfiguration.DEFAULT_RM_RESOURCE_TRACKER_ADDRESS.split(":")[1]));
 
     resourceTrackerConnectAddress = conf.updateConnectAddr(
         YarnConfiguration.RM_BIND_HOST,
@@ -213,7 +213,7 @@ public class TestYarnConfiguration {
 
     serverAddress = new InetSocketAddress(
         YarnConfiguration.DEFAULT_NM_LOCALIZER_ADDRESS.split(":")[0],
-        Integer.valueOf(YarnConfiguration.DEFAULT_NM_LOCALIZER_ADDRESS.split(":")[1]));
+        Integer.parseInt(YarnConfiguration.DEFAULT_NM_LOCALIZER_ADDRESS.split(":")[1]));
 
     InetSocketAddress localizerAddress = conf.updateConnectAddr(
         YarnConfiguration.NM_BIND_HOST,

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/test/java/org/apache/hadoop/yarn/nodelabels/NodeLabelTestBase.java

@@ -140,7 +140,7 @@ public class NodeLabelTestBase {
       int idx = str.indexOf(':');
       NodeId id =
           NodeId.newInstance(str.substring(0, idx),
-              Integer.valueOf(str.substring(idx + 1)));
+              Integer.parseInt(str.substring(idx + 1)));
       return id;
     } else {
       return NodeId.newInstance(str, CommonNodeLabelsManager.WILDCARD_PORT);

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/test/java/org/apache/hadoop/yarn/util/TestFSDownload.java

@@ -431,7 +431,7 @@ public class TestFSDownload {
     try {
       for (Map.Entry<LocalResource,Future<Path>> p : pending.entrySet()) {
         Path localized = p.getValue().get();
-        assertEquals(sizes[Integer.valueOf(localized.getName())], p.getKey()
+        assertEquals(sizes[Integer.parseInt(localized.getName())], p.getKey()
             .getSize());
 
         FileStatus status = files.getFileStatus(localized.getParent());

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-registry/src/main/java/org/apache/hadoop/registry/client/impl/zk/RegistrySecurity.java

@@ -772,7 +772,7 @@ public class RegistrySecurity extends AbstractService {
    * @return true if the SASL client system property is set.
    */
   public static boolean isClientSASLEnabled() {
-    return Boolean.valueOf(System.getProperty(
+    return Boolean.parseBoolean(System.getProperty(
         ZookeeperConfigOptions.PROP_ZK_ENABLE_SASL_CLIENT, "true"));
   }
 
@@ -862,7 +862,7 @@ public class RegistrySecurity extends AbstractService {
     String sasl =
         System.getProperty(PROP_ZK_ENABLE_SASL_CLIENT,
             DEFAULT_ZK_ENABLE_SASL_CLIENT);
-    boolean saslEnabled = Boolean.valueOf(sasl);
+    boolean saslEnabled = Boolean.parseBoolean(sasl);
     builder.append(describeProperty(PROP_ZK_ENABLE_SASL_CLIENT,
         DEFAULT_ZK_ENABLE_SASL_CLIENT));
     if (saslEnabled) {

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/NodeStatusUpdaterImpl.java

@@ -284,6 +284,7 @@ public class NodeStatusUpdaterImpl extends AbstractService implements
         return;
       }
       this.isStopped = true;
+      sendOutofBandHeartBeat();
       try {
         statusUpdater.join();
         registerWithRM();

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/containermanager/launcher/ContainerLaunch.java

@@ -1017,7 +1017,7 @@ public class ContainerLaunch implements Callable<Integer> {
         //variable can be set to indicate that distcache entries should come
         //first
 
-        boolean preferLocalizedJars = Boolean.valueOf(
+        boolean preferLocalizedJars = Boolean.parseBoolean(
           environment.get(Environment.CLASSPATH_PREPEND_DISTCACHE.name())
           );
 

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/containermanager/localizer/LocalizedResource.java

@@ -200,9 +200,11 @@ public class LocalizedResource implements EventHandler<ResourceEvent> {
         LOG.warn("Can't handle this event at current state", e);
       }
       if (oldState != newState) {
-        LOG.info("Resource " + resourcePath + (localPath != null ? 
-          "(->" + localPath + ")": "") + " transitioned from " + oldState
-            + " to " + newState);
+        if (LOG.isDebugEnabled()) {
+          LOG.debug("Resource " + resourcePath + (localPath != null ?
+              "(->" + localPath + ")": "") + " transitioned from " + oldState
+              + " to " + newState);
+        }
       }
     } finally {
       this.writeLock.unlock();

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/util/ProcessIdFileReader.java

@@ -79,7 +79,7 @@ public class ProcessIdFileReader {
             else {
               // Otherwise, find first line containing a numeric pid.
               try {
-                Long pid = Long.valueOf(temp);
+                long pid = Long.parseLong(temp);
                 if (pid > 0) {
                   processId = temp;
                   break;

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/webapp/ApplicationPage.java

@@ -34,6 +34,7 @@ import org.apache.hadoop.yarn.util.ConverterUtils;
 import org.apache.hadoop.yarn.webapp.SubView;
 import org.apache.hadoop.yarn.webapp.YarnWebParams;
 import org.apache.hadoop.yarn.webapp.hamlet.Hamlet;
+import org.apache.hadoop.yarn.webapp.hamlet.Hamlet.DIV;
 import org.apache.hadoop.yarn.webapp.hamlet.Hamlet.TABLE;
 import org.apache.hadoop.yarn.webapp.view.HtmlBlock;
 import org.apache.hadoop.yarn.webapp.view.InfoBlock;
@@ -75,10 +76,21 @@ public class ApplicationPage extends NMView implements YarnWebParams {
 
     @Override
     protected void render(Block html) {
-      ApplicationId applicationID =
-          ConverterUtils.toApplicationId(this.recordFactory,
-              $(APPLICATION_ID));
+      ApplicationId applicationID = null;
+      try {
+        applicationID = ConverterUtils.toApplicationId(this.recordFactory,
+            $(APPLICATION_ID));
+      } catch (IllegalArgumentException e) {
+        html.p()._("Invalid Application Id " + $(APPLICATION_ID))._();
+        return;
+      }
+      DIV<Hamlet> div = html.div("#content");
       Application app = this.nmContext.getApplications().get(applicationID);
+      if (app == null) {
+        div.h1("Unknown application with id " + applicationID
+            + ". Application might have been completed")._();
+        return;
+      }
       AppInfo info = new AppInfo(app);
       info("Application's information")
             ._("ApplicationId", info.getId())

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/test/java/org/apache/hadoop/yarn/server/nodemanager/TestNodeManagerResync.java

@@ -108,6 +108,7 @@ public class TestNodeManagerResync {
   static final String user = "nobody";
   private FileContext localFS;
   private CyclicBarrier syncBarrier;
+  private CyclicBarrier updateBarrier;
   private AtomicBoolean assertionFailedInThread = new AtomicBoolean(false);
   private AtomicBoolean isNMShutdownCalled = new AtomicBoolean(false);
   private final NodeManagerEvent resyncEvent =
@@ -125,6 +126,7 @@ public class TestNodeManagerResync {
     remoteLogsDir.mkdirs();
     nmLocalDir.mkdirs();
     syncBarrier = new CyclicBarrier(2);
+    updateBarrier = new CyclicBarrier(2);
   }
 
   @After
@@ -803,9 +805,11 @@ public class TestNodeManagerResync {
                 .getContainerStatuses(gcsRequest).getContainerStatuses().get(0);
             assertEquals(Resource.newInstance(1024, 1),
                 containerStatus.getCapability());
+            updateBarrier.await();
             // Call the actual rebootNodeStatusUpdaterAndRegisterWithRM().
             // This function should be synchronized with
             // increaseContainersResource().
+            updateBarrier.await();
             super.rebootNodeStatusUpdaterAndRegisterWithRM();
             // Check status after registerWithRM
             containerStatus = getContainerManager()
@@ -831,17 +835,24 @@ public class TestNodeManagerResync {
         List<Token> increaseTokens = new ArrayList<Token>();
         // Add increase request.
         Resource targetResource = Resource.newInstance(4096, 2);
-        try {
-          increaseTokens.add(getContainerToken(targetResource));
-          IncreaseContainersResourceRequest increaseRequest =
-              IncreaseContainersResourceRequest.newInstance(increaseTokens);
-          IncreaseContainersResourceResponse increaseResponse =
-              getContainerManager()
-                  .increaseContainersResource(increaseRequest);
-          Assert.assertEquals(
-              1, increaseResponse.getSuccessfullyIncreasedContainers()
-                  .size());
-          Assert.assertTrue(increaseResponse.getFailedRequests().isEmpty());
+        try{
+          try {
+            updateBarrier.await();
+            increaseTokens.add(getContainerToken(targetResource));
+            IncreaseContainersResourceRequest increaseRequest =
+                IncreaseContainersResourceRequest.newInstance(increaseTokens);
+            IncreaseContainersResourceResponse increaseResponse =
+                getContainerManager()
+                    .increaseContainersResource(increaseRequest);
+            Assert.assertEquals(
+                1, increaseResponse.getSuccessfullyIncreasedContainers()
+                    .size());
+            Assert.assertTrue(increaseResponse.getFailedRequests().isEmpty());
+          } catch (Exception e) {
+            e.printStackTrace();
+          } finally {
+            updateBarrier.await();
+          }
         } catch (Exception e) {
           e.printStackTrace();
         }

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/test/java/org/apache/hadoop/yarn/server/nodemanager/containermanager/logaggregation/TestLogAggregationService.java

@@ -223,16 +223,27 @@ public class TestLogAggregationService extends BaseContainerManagerTest {
         any(UserGroupInformation.class));
     verify(delSrvc).delete(eq(user), eq((Path) null),
       eq(new Path(app1LogDir.getAbsolutePath())));
-    delSrvc.stop();
     
     String containerIdStr = ConverterUtils.toString(container11);
     File containerLogDir = new File(app1LogDir, containerIdStr);
+    int count = 0;
+    int maxAttempts = 50;
     for (String fileType : new String[] { "stdout", "stderr", "syslog" }) {
       File f = new File(containerLogDir, fileType);
-      Assert.assertFalse("check "+f, f.exists());
+      count = 0;
+      while ((f.exists()) && (count < maxAttempts)) {
+        count++;
+        Thread.sleep(100);
+      }
+      Assert.assertFalse("File [" + f + "] was not deleted", f.exists());
     }
-
-    Assert.assertFalse(app1LogDir.exists());
+    count = 0;
+    while ((app1LogDir.exists()) && (count < maxAttempts)) {
+      count++;
+      Thread.sleep(100);
+    }
+    Assert.assertFalse("Directory [" + app1LogDir + "] was not deleted",
+      app1LogDir.exists());
 
     Path logFilePath =
         logAggregationService.getRemoteNodeLogFileForApp(application1,

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/test/java/org/apache/hadoop/yarn/server/nodemanager/webapp/TestNMAppsPage.java

@@ -0,0 +1,86 @@
+/**
+ * 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.hadoop.yarn.server.nodemanager.webapp;
+
+import static org.mockito.Mockito.mock;
+import static org.mockito.Mockito.when;
+
+import java.util.Arrays;
+import java.util.Collection;
+
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.yarn.server.nodemanager.Context;
+import org.apache.hadoop.yarn.server.nodemanager.NodeManager;
+import org.apache.hadoop.yarn.server.nodemanager.NodeManager.NMContext;
+import org.apache.hadoop.yarn.server.nodemanager.recovery.NMNullStateStoreService;
+import org.apache.hadoop.yarn.server.nodemanager.security.NMContainerTokenSecretManager;
+import org.apache.hadoop.yarn.server.nodemanager.security.NMTokenSecretManagerInNM;
+import org.apache.hadoop.yarn.server.nodemanager.webapp.ApplicationPage.ApplicationBlock;
+import org.apache.hadoop.yarn.server.security.ApplicationACLsManager;
+import org.apache.hadoop.yarn.webapp.YarnWebParams;
+import org.apache.hadoop.yarn.webapp.test.WebAppTests;
+import org.junit.Test;
+import org.junit.runner.RunWith;
+import org.junit.runners.Parameterized;
+
+import com.google.inject.Binder;
+import com.google.inject.Injector;
+import com.google.inject.Module;
+
+@RunWith(Parameterized.class)
+public class TestNMAppsPage {
+
+  String applicationid;
+
+  public TestNMAppsPage(String appid) {
+    this.applicationid = appid;
+  }
+
+  @Parameterized.Parameters
+  public static Collection<Object[]> getAppIds() {
+    return Arrays.asList(new Object[][] { { "appid" },
+        { "application_123123213_0001" }, { "" } });
+  }
+
+  @Test
+  public void testNMAppsPage() {
+    Configuration conf = new Configuration();
+    final NMContext nmcontext = new NMContext(
+        new NMContainerTokenSecretManager(conf), new NMTokenSecretManagerInNM(),
+        null, new ApplicationACLsManager(conf), new NMNullStateStoreService());
+    Injector injector = WebAppTests.createMockInjector(NMContext.class,
+        nmcontext, new Module() {
+          @Override
+          public void configure(Binder binder) {
+            NodeManager nm = TestNMAppsPage.mocknm(nmcontext);
+            binder.bind(NodeManager.class).toInstance(nm);
+            binder.bind(Context.class).toInstance(nmcontext);
+          }
+        });
+    ApplicationBlock instance = injector.getInstance(ApplicationBlock.class);
+    instance.set(YarnWebParams.APPLICATION_ID, applicationid);
+    instance.render();
+  }
+
+  protected static NodeManager mocknm(NMContext nmcontext) {
+    NodeManager rm = mock(NodeManager.class);
+    when(rm.getNMContext()).thenReturn(nmcontext);
+    return rm;
+  }
+
+}

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/AdminService.java

@@ -682,7 +682,11 @@ public class AdminService extends CompositeService implements
     return conf;
   }
 
-  private void refreshAll() throws ServiceFailedException {
+  /*
+   * Visibility could be private for test its made as default
+   */
+  @VisibleForTesting
+  void refreshAll() throws ServiceFailedException {
     try {
       refreshQueues(RefreshQueuesRequest.newInstance());
       refreshNodes(RefreshNodesRequest.newInstance(DecommissionType.NORMAL));

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/NodesListManager.java

@@ -36,7 +36,6 @@ import org.apache.hadoop.net.Node;
 import org.apache.hadoop.service.AbstractService;
 import org.apache.hadoop.service.CompositeService;
 import org.apache.hadoop.util.HostsFileReader;
-import org.apache.hadoop.util.Time;
 import org.apache.hadoop.yarn.api.records.NodeId;
 import org.apache.hadoop.yarn.api.records.NodeState;
 import org.apache.hadoop.yarn.conf.YarnConfiguration;
@@ -69,8 +68,6 @@ public class NodesListManager extends CompositeService implements
   private String excludesFile;
 
   private Resolver resolver;
-  private Timer removalTimer;
-  private int nodeRemovalCheckInterval;
 
   public NodesListManager(RMContext rmContext) {
     super(NodesListManager.class.getName());
@@ -108,56 +105,9 @@ public class NodesListManager extends CompositeService implements
     } catch (IOException ioe) {
       disableHostsFileReader(ioe);
     }
-
-    final int nodeRemovalTimeout =
-        conf.getInt(
-            YarnConfiguration.RM_NODEMANAGER_UNTRACKED_REMOVAL_TIMEOUT_MSEC,
-            YarnConfiguration.
-                DEFAULT_RM_NODEMANAGER_UNTRACKED_REMOVAL_TIMEOUT_MSEC);
-    nodeRemovalCheckInterval = (Math.min(nodeRemovalTimeout/2,
-        600000));
-    removalTimer = new Timer("Node Removal Timer");
-
-    removalTimer.schedule(new TimerTask() {
-      @Override
-      public void run() {
-        long now = Time.monotonicNow();
-        for (Map.Entry<NodeId, RMNode> entry :
-            rmContext.getInactiveRMNodes().entrySet()) {
-          NodeId nodeId = entry.getKey();
-          RMNode rmNode = entry.getValue();
-          if (isUntrackedNode(rmNode.getHostName())) {
-            if (rmNode.getUntrackedTimeStamp() == 0) {
-              rmNode.setUntrackedTimeStamp(now);
-            } else if (now - rmNode.getUntrackedTimeStamp() >
-                  nodeRemovalTimeout) {
-              RMNode result = rmContext.getInactiveRMNodes().remove(nodeId);
-              if (result != null) {
-                ClusterMetrics clusterMetrics = ClusterMetrics.getMetrics();
-                if (rmNode.getState() == NodeState.SHUTDOWN) {
-                  clusterMetrics.decrNumShutdownNMs();
-                } else {
-                  clusterMetrics.decrDecommisionedNMs();
-                }
-                LOG.info("Removed "+result.getHostName() +
-                    " from inactive nodes list");
-              }
-            }
-          } else {
-            rmNode.setUntrackedTimeStamp(0);
-          }
-        }
-      }
-    }, nodeRemovalCheckInterval, nodeRemovalCheckInterval);
-
     super.serviceInit(conf);
   }
 
-  @Override
-  public void serviceStop() {
-    removalTimer.cancel();
-  }
-
   private void printConfiguredHosts() {
     if (!LOG.isDebugEnabled()) {
       return;
@@ -181,13 +131,10 @@ public class NodesListManager extends CompositeService implements
 
     for (NodeId nodeId: rmContext.getRMNodes().keySet()) {
       if (!isValidNode(nodeId.getHost())) {
-        RMNodeEventType nodeEventType = isUntrackedNode(nodeId.getHost()) ?
-            RMNodeEventType.SHUTDOWN : RMNodeEventType.DECOMMISSION;
         this.rmContext.getDispatcher().getEventHandler().handle(
-            new RMNodeEvent(nodeId, nodeEventType));
+            new RMNodeEvent(nodeId, RMNodeEventType.DECOMMISSION));
       }
     }
-    updateInactiveNodes();
   }
 
   private void refreshHostsReader(Configuration yarnConf) throws IOException,
@@ -224,16 +171,6 @@ public class NodesListManager extends CompositeService implements
     }
   }
 
-  @VisibleForTesting
-  public int getNodeRemovalCheckInterval() {
-    return nodeRemovalCheckInterval;
-  }
-
-  @VisibleForTesting
-  public void setNodeRemovalCheckInterval(int interval) {
-    this.nodeRemovalCheckInterval = interval;
-  }
-
   @VisibleForTesting
   public Resolver getResolver() {
     return resolver;
@@ -437,33 +374,6 @@ public class NodesListManager extends CompositeService implements
     return hostsReader;
   }
 
-  private void updateInactiveNodes() {
-    long now = Time.monotonicNow();
-    for(Entry<NodeId, RMNode> entry :
-        rmContext.getInactiveRMNodes().entrySet()) {
-      NodeId nodeId = entry.getKey();
-      RMNode rmNode = entry.getValue();
-      if (isUntrackedNode(nodeId.getHost()) &&
-          rmNode.getUntrackedTimeStamp() == 0) {
-        rmNode.setUntrackedTimeStamp(now);
-      }
-    }
-  }
-
-  public boolean isUntrackedNode(String hostName) {
-    boolean untracked;
-    String ip = resolver.resolve(hostName);
-
-    synchronized (hostsReader) {
-      Set<String> hostsList = hostsReader.getHosts();
-      Set<String> excludeList = hostsReader.getExcludedHosts();
-      untracked = !hostsList.isEmpty() &&
-          !hostsList.contains(hostName) && !hostsList.contains(ip) &&
-          !excludeList.contains(hostName) && !excludeList.contains(ip);
-    }
-    return untracked;
-  }
-
   /**
    * Refresh the nodes gracefully
    *
@@ -474,13 +384,11 @@ public class NodesListManager extends CompositeService implements
   public void refreshNodesGracefully(Configuration conf) throws IOException,
       YarnException {
     refreshHostsReader(conf);
-    for (Entry<NodeId, RMNode> entry : rmContext.getRMNodes().entrySet()) {
+    for (Entry<NodeId, RMNode> entry:rmContext.getRMNodes().entrySet()) {
       NodeId nodeId = entry.getKey();
       if (!isValidNode(nodeId.getHost())) {
-        RMNodeEventType nodeEventType = isUntrackedNode(nodeId.getHost()) ?
-            RMNodeEventType.SHUTDOWN : RMNodeEventType.GRACEFUL_DECOMMISSION;
         this.rmContext.getDispatcher().getEventHandler().handle(
-            new RMNodeEvent(nodeId, nodeEventType));
+            new RMNodeEvent(nodeId, RMNodeEventType.GRACEFUL_DECOMMISSION));
       } else {
         // Recommissioning the nodes
         if (entry.getValue().getState() == NodeState.DECOMMISSIONING) {
@@ -489,7 +397,6 @@ public class NodesListManager extends CompositeService implements
         }
       }
     }
-    updateInactiveNodes();
   }
 
   /**
@@ -513,11 +420,8 @@ public class NodesListManager extends CompositeService implements
   public void refreshNodesForcefully() {
     for (Entry<NodeId, RMNode> entry : rmContext.getRMNodes().entrySet()) {
       if (entry.getValue().getState() == NodeState.DECOMMISSIONING) {
-        RMNodeEventType nodeEventType =
-            isUntrackedNode(entry.getKey().getHost()) ?
-            RMNodeEventType.SHUTDOWN : RMNodeEventType.DECOMMISSION;
         this.rmContext.getDispatcher().getEventHandler().handle(
-            new RMNodeEvent(entry.getKey(), nodeEventType));
+            new RMNodeEvent(entry.getKey(), RMNodeEventType.DECOMMISSION));
       }
     }
   }

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/RMServerUtils.java

@@ -87,7 +87,7 @@ public class RMServerUtils {
         acceptedStates.contains(NodeState.LOST) ||
         acceptedStates.contains(NodeState.REBOOTED)) {
       for (RMNode rmNode : context.getInactiveRMNodes().values()) {
-        if ((rmNode != null) && acceptedStates.contains(rmNode.getState())) {
+        if (acceptedStates.contains(rmNode.getState())) {
           results.add(rmNode);
         }
       }

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/ResourceTrackerService.java

@@ -320,8 +320,7 @@ public class ResourceTrackerService extends AbstractService implements
     }
 
     // Check if this node is a 'valid' node
-    if (!this.nodesListManager.isValidNode(host) ||
-        this.nodesListManager.isUntrackedNode(host)) {
+    if (!this.nodesListManager.isValidNode(host)) {
       String message =
           "Disallowed NodeManager from  " + host
               + ", Sending SHUTDOWN signal to the NodeManager.";
@@ -452,9 +451,8 @@ public class ResourceTrackerService extends AbstractService implements
 
     // 1. Check if it's a valid (i.e. not excluded) node, if not, see if it is
     // in decommissioning.
-    if ((!this.nodesListManager.isValidNode(nodeId.getHost()) &&
-        !isNodeInDecommissioning(nodeId)) ||
-        this.nodesListManager.isUntrackedNode(nodeId.getHost())) {
+    if (!this.nodesListManager.isValidNode(nodeId.getHost())
+        && !isNodeInDecommissioning(nodeId)) {
       String message =
           "Disallowed NodeManager nodeId: " + nodeId + " hostname: "
               + nodeId.getHost();

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/rmapp/attempt/RMAppAttemptImpl.java

@@ -927,6 +927,20 @@ public class RMAppAttemptImpl implements RMAppAttempt, Recoverable {
     this.justFinishedContainers = attempt.getJustFinishedContainersReference();
     this.finishedContainersSentToAM =
         attempt.getFinishedContainersSentToAMReference();
+    // container complete msg was moved from justFinishedContainers to
+    // finishedContainersSentToAM in ApplicationMasterService#allocate,
+    // if am crashed and not received this response, we should resend
+    // this msg again after am restart
+    if (!this.finishedContainersSentToAM.isEmpty()) {
+      for (NodeId nodeId : this.finishedContainersSentToAM.keySet()) {
+        List<ContainerStatus> containerStatuses =
+            this.finishedContainersSentToAM.get(nodeId);
+        this.justFinishedContainers.putIfAbsent(nodeId,
+            new ArrayList<ContainerStatus>());
+        this.justFinishedContainers.get(nodeId).addAll(containerStatuses);
+      }
+      this.finishedContainersSentToAM.clear();
+    }
   }
 
   private void recoverAppAttemptCredentials(Credentials appAttemptTokens,
@@ -1845,13 +1859,13 @@ public class RMAppAttemptImpl implements RMAppAttempt, Recoverable {
     } else {
       LOG.warn("No ContainerStatus in containerFinishedEvent");
     }
-    finishedContainersSentToAM.putIfAbsent(nodeId,
-      new ArrayList<ContainerStatus>());
-    appAttempt.finishedContainersSentToAM.get(nodeId).add(
-      containerFinishedEvent.getContainerStatus());
 
     if (!appAttempt.getSubmissionContext()
-      .getKeepContainersAcrossApplicationAttempts()) {
+        .getKeepContainersAcrossApplicationAttempts()) {
+      finishedContainersSentToAM.putIfAbsent(nodeId,
+          new ArrayList<ContainerStatus>());
+      appAttempt.finishedContainersSentToAM.get(nodeId).add(
+          containerFinishedEvent.getContainerStatus());
       appAttempt.sendFinishedContainersToNM();
     } else {
       appAttempt.sendFinishedAMContainerToNM(nodeId,

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/rmnode/RMNode.java

@@ -168,8 +168,4 @@ public interface RMNode {
       NodeHeartbeatResponse response);
   
   public List<Container> pullNewlyIncreasedContainers();
-
-  long getUntrackedTimeStamp();
-
-  void setUntrackedTimeStamp(long timer);
 }

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/rmnode/RMNodeImpl.java

@@ -39,7 +39,6 @@ import org.apache.hadoop.classification.InterfaceAudience.Private;
 import org.apache.hadoop.classification.InterfaceStability.Unstable;
 import org.apache.hadoop.net.Node;
 import org.apache.hadoop.security.UserGroupInformation;
-import org.apache.hadoop.util.Time;
 import org.apache.hadoop.yarn.api.protocolrecords.SignalContainerRequest;
 import org.apache.hadoop.yarn.api.records.ApplicationId;
 import org.apache.hadoop.yarn.api.records.Container;
@@ -121,7 +120,6 @@ public class RMNodeImpl implements RMNode, EventHandler<RMNodeEvent> {
   private long lastHealthReportTime;
   private String nodeManagerVersion;
 
-  private long timeStamp;
   /* Aggregated resource utilization for the containers. */
   private ResourceUtilization containersUtilization;
   /* Resource utilization for the node. */
@@ -261,9 +259,6 @@ public class RMNodeImpl implements RMNode, EventHandler<RMNodeEvent> {
 
       .addTransition(NodeState.DECOMMISSIONING, NodeState.DECOMMISSIONING,
           RMNodeEventType.CLEANUP_APP, new CleanUpAppTransition())
-      .addTransition(NodeState.DECOMMISSIONING, NodeState.SHUTDOWN,
-          RMNodeEventType.SHUTDOWN,
-          new DeactivateNodeTransition(NodeState.SHUTDOWN))
 
       // TODO (in YARN-3223) update resource when container finished.
       .addTransition(NodeState.DECOMMISSIONING, NodeState.DECOMMISSIONING,
@@ -351,7 +346,6 @@ public class RMNodeImpl implements RMNode, EventHandler<RMNodeEvent> {
     this.healthReport = "Healthy";
     this.lastHealthReportTime = System.currentTimeMillis();
     this.nodeManagerVersion = nodeManagerVersion;
-    this.timeStamp = 0;
 
     this.latestNodeHeartBeatResponse.setResponseId(0);
 
@@ -1017,7 +1011,7 @@ public class RMNodeImpl implements RMNode, EventHandler<RMNodeEvent> {
   }
 
   /**
-   * Put a node in deactivated (decommissioned or shutdown) status.
+   * Put a node in deactivated (decommissioned) status.
    * @param rmNode
    * @param finalState
    */
@@ -1034,10 +1028,6 @@ public class RMNodeImpl implements RMNode, EventHandler<RMNodeEvent> {
     LOG.info("Deactivating Node " + rmNode.nodeId + " as it is now "
         + finalState);
     rmNode.context.getInactiveRMNodes().put(rmNode.nodeId, rmNode);
-    if (finalState == NodeState.SHUTDOWN &&
-        rmNode.context.getNodesListManager().isUntrackedNode(rmNode.hostName)) {
-      rmNode.setUntrackedTimeStamp(Time.monotonicNow());
-    }
   }
 
   /**
@@ -1393,14 +1383,4 @@ public class RMNodeImpl implements RMNode, EventHandler<RMNodeEvent> {
   public Resource getOriginalTotalCapability() {
     return this.originalTotalCapability;
   }
-
-  @Override
-  public long getUntrackedTimeStamp() {
-    return this.timeStamp;
-  }
-
-  @Override
-  public void setUntrackedTimeStamp(long ts) {
-    this.timeStamp = ts;
-  }
  }

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/AppSchedulingInfo.java

@@ -29,8 +29,8 @@ import java.util.Set;
 import java.util.TreeMap;
 import java.util.TreeSet;
 import java.util.concurrent.ConcurrentHashMap;
-import java.util.concurrent.atomic.AtomicLong;
 import java.util.concurrent.atomic.AtomicBoolean;
+import java.util.concurrent.atomic.AtomicLong;
 
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
@@ -46,6 +46,7 @@ import org.apache.hadoop.yarn.api.records.Resource;
 import org.apache.hadoop.yarn.api.records.ResourceRequest;
 import org.apache.hadoop.yarn.exceptions.YarnException;
 import org.apache.hadoop.yarn.server.resourcemanager.RMServerUtils;
+import org.apache.hadoop.yarn.server.resourcemanager.nodelabels.RMNodeLabelsManager;
 import org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainer;
 import org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainerState;
 import org.apache.hadoop.yarn.util.resource.Resources;
@@ -75,6 +76,7 @@ public class AppSchedulingInfo {
   private AtomicBoolean userBlacklistChanged = new AtomicBoolean(false);
   private final Set<String> amBlacklist = new HashSet<>();
   private Set<String> userBlacklist = new HashSet<>();
+  private Set<String> requestedPartitions = new HashSet<>();
 
   final Set<Priority> priorities = new TreeSet<>(COMPARATOR);
   final Map<Priority, Map<String, ResourceRequest>> resourceRequestMap =
@@ -119,6 +121,10 @@ public class AppSchedulingInfo {
     return pending;
   }
   
+  public Set<String> getRequestedPartitions() {
+    return requestedPartitions;
+  }
+
   /**
    * Clear any pending requests from this application.
    */
@@ -340,6 +346,10 @@ public class AppSchedulingInfo {
       asks.put(resourceName, request);
 
       if (resourceName.equals(ResourceRequest.ANY)) {
+        //update the applications requested labels set
+        requestedPartitions.add(request.getNodeLabelExpression() == null
+            ? RMNodeLabelsManager.NO_LABEL : request.getNodeLabelExpression());
+
         anyResourcesUpdated = true;
 
         // Activate application. Metrics activation is done here.

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/capacity/AbstractCSQueue.java

@@ -439,9 +439,8 @@ public abstract class AbstractCSQueue implements CSQueue {
        * limit-set-by-parent)
        */
       Resource queueMaxResource =
-          Resources.multiplyAndNormalizeDown(resourceCalculator,
-              labelManager.getResourceByLabel(nodePartition, clusterResource),
-              queueCapacities.getAbsoluteMaximumCapacity(nodePartition), minimumAllocation);
+          getQueueMaxResource(nodePartition, clusterResource);
+
       return Resources.min(resourceCalculator, clusterResource,
           queueMaxResource, currentResourceLimits.getLimit());
     } else if (schedulingMode == SchedulingMode.IGNORE_PARTITION_EXCLUSIVITY) {
@@ -452,7 +451,14 @@ public abstract class AbstractCSQueue implements CSQueue {
     
     return Resources.none();
   }
-  
+
+  Resource getQueueMaxResource(String nodePartition, Resource clusterResource) {
+    return Resources.multiplyAndNormalizeDown(resourceCalculator,
+        labelManager.getResourceByLabel(nodePartition, clusterResource),
+        queueCapacities.getAbsoluteMaximumCapacity(nodePartition),
+        minimumAllocation);
+  }
+
   synchronized boolean canAssignToThisQueue(Resource clusterResource,
       String nodePartition, ResourceLimits currentResourceLimits,
       Resource resourceCouldBeUnreserved, SchedulingMode schedulingMode) {

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/capacity/CapacityHeadroomProvider.java

@@ -17,8 +17,12 @@
  */
 package org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity;
 
-import org.apache.hadoop.yarn.server.resourcemanager.scheduler.common.fica.FiCaSchedulerApp;
+import java.util.Set;
+
 import org.apache.hadoop.yarn.api.records.Resource;
+import org.apache.hadoop.yarn.server.resourcemanager.nodelabels.RMNodeLabelsManager;
+import org.apache.hadoop.yarn.server.resourcemanager.scheduler.common.fica.FiCaSchedulerApp;
+import org.apache.hadoop.yarn.util.resource.Resources;
 
 public class CapacityHeadroomProvider {
   
@@ -38,22 +42,32 @@ public class CapacityHeadroomProvider {
   }
   
   public Resource getHeadroom() {
-    
+
     Resource queueCurrentLimit;
     Resource clusterResource;
     synchronized (queueResourceLimitsInfo) {
       queueCurrentLimit = queueResourceLimitsInfo.getQueueCurrentLimit();
       clusterResource = queueResourceLimitsInfo.getClusterResource();
     }
-    Resource headroom = queue.getHeadroom(user, queueCurrentLimit, 
-      clusterResource, application);
-    
+    Set<String> requestedPartitions =
+        application.getAppSchedulingInfo().getRequestedPartitions();
+    Resource headroom;
+    if (requestedPartitions.isEmpty() || (requestedPartitions.size() == 1
+        && requestedPartitions.contains(RMNodeLabelsManager.NO_LABEL))) {
+      headroom = queue.getHeadroom(user, queueCurrentLimit, clusterResource,
+          application);
+    } else {
+      headroom = Resource.newInstance(0, 0);
+      for (String partition : requestedPartitions) {
+        Resource partitionHeadRoom = queue.getHeadroom(user, queueCurrentLimit,
+            clusterResource, application, partition);
+        Resources.addTo(headroom, partitionHeadRoom);
+      }
+    }
     // Corner case to deal with applications being slightly over-limit
     if (headroom.getMemory() < 0) {
       headroom.setMemory(0);
     }
     return headroom;
-  
   }
-
 }