CDPD-8485. HADOOP-16910 : ABFS Streams to update FileSystem.Statistics counters on IO. (apache#1918).

Contributed by Mehakmeet Singh.

Change-Id: If7d82e40427a3ccc1418b935cd4574dee262b37b

Author: Mehakmeet Singh

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

@@ -188,7 +188,7 @@ public FSDataOutputStream create(final Path f, final FsPermission permission, fi
     Path qualifiedPath = makeQualified(f);
 
     try {
-      OutputStream outputStream = abfsStore.createFile(qualifiedPath, overwrite,
+      OutputStream outputStream = abfsStore.createFile(qualifiedPath, statistics, overwrite,
           permission == null ? FsPermission.getFileDefault() : permission, FsPermission.getUMask(getConf()));
       return new FSDataOutputStream(outputStream, statistics);
     } catch(AzureBlobFileSystemException ex) {
@@ -250,7 +250,7 @@ public FSDataOutputStream append(final Path f, final int bufferSize, final Progr
     Path qualifiedPath = makeQualified(f);
 
     try {
-      OutputStream outputStream = abfsStore.openFileForWrite(qualifiedPath, false);
+      OutputStream outputStream = abfsStore.openFileForWrite(qualifiedPath, statistics, false);
       return new FSDataOutputStream(outputStream, statistics);
     } catch(AzureBlobFileSystemException ex) {
       checkException(f, ex);

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

@@ -52,6 +52,8 @@
 import com.google.common.annotations.VisibleForTesting;
 import com.google.common.base.Preconditions;
 import com.google.common.base.Strings;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
 
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
@@ -99,8 +101,6 @@
 import org.apache.hadoop.io.IOUtils;
 import org.apache.hadoop.security.UserGroupInformation;
 import org.apache.http.client.utils.URIBuilder;
-import org.slf4j.Logger;
-import org.slf4j.LoggerFactory;
 
 import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.CHAR_EQUALS;
 import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.CHAR_FORWARD_SLASH;
@@ -403,8 +403,10 @@ public void deleteFilesystem() throws AzureBlobFileSystemException {
     }
   }
 
-  public OutputStream createFile(final Path path, final boolean overwrite, final FsPermission permission,
-                                 final FsPermission umask) throws AzureBlobFileSystemException {
+  public OutputStream createFile(final Path path,
+      final FileSystem.Statistics statistics,
+      final boolean overwrite, final FsPermission permission,
+      final FsPermission umask) throws AzureBlobFileSystemException {
     try (AbfsPerfInfo perfInfo = startTracking("createFile", "createPath")) {
       boolean isNamespaceEnabled = getIsNamespaceEnabled();
       LOG.debug("createFile filesystem: {} path: {} overwrite: {} permission: {} umask: {} isNamespaceEnabled: {}",
@@ -421,12 +423,13 @@ public OutputStream createFile(final Path path, final boolean overwrite, final F
       perfInfo.registerResult(op.getResult()).registerSuccess(true);
 
       return new AbfsOutputStream(
-              client,
-              AbfsHttpConstants.FORWARD_SLASH + getRelativePath(path),
-              0,
-              abfsConfiguration.getWriteBufferSize(),
-              abfsConfiguration.isFlushEnabled(),
-              abfsConfiguration.isOutputStreamFlushDisabled());
+          client,
+          statistics,
+          AbfsHttpConstants.FORWARD_SLASH + getRelativePath(path),
+          0,
+          abfsConfiguration.getWriteBufferSize(),
+          abfsConfiguration.isFlushEnabled(),
+          abfsConfiguration.isOutputStreamFlushDisabled());
     }
   }
 
@@ -482,7 +485,7 @@ public AbfsInputStream openFileForRead(final Path path, final FileSystem.Statist
     }
   }
 
-  public OutputStream openFileForWrite(final Path path, final boolean overwrite) throws
+  public OutputStream openFileForWrite(final Path path, final FileSystem.Statistics statistics, final boolean overwrite) throws
           AzureBlobFileSystemException {
     try (AbfsPerfInfo perfInfo = startTracking("openFileForWrite", "getPathStatus")) {
       LOG.debug("openFileForWrite filesystem: {} path: {} overwrite: {}",
@@ -509,12 +512,13 @@ public OutputStream openFileForWrite(final Path path, final boolean overwrite) t
       perfInfo.registerSuccess(true);
 
       return new AbfsOutputStream(
-              client,
-              AbfsHttpConstants.FORWARD_SLASH + getRelativePath(path),
-              offset,
-              abfsConfiguration.getWriteBufferSize(),
-              abfsConfiguration.isFlushEnabled(),
-              abfsConfiguration.isOutputStreamFlushDisabled());
+          client,
+          statistics,
+          AbfsHttpConstants.FORWARD_SLASH + getRelativePath(path),
+          offset,
+          abfsConfiguration.getWriteBufferSize(),
+          abfsConfiguration.isFlushEnabled(),
+          abfsConfiguration.isOutputStreamFlushDisabled());
     }
   }
 

hadoop-tools/hadoop-azure/src/main/java/org/apache/hadoop/fs/azurebfs/services/AbfsInputStream.java

@@ -131,6 +131,7 @@ public synchronized int read(final byte[] b, final int off, final int len) throw
     int lastReadBytes;
     int totalReadBytes = 0;
     streamStatistics.readOperationStarted(off, len);
+    incrementReadOps();
     do {
       if (nextReadPos >= fCursor - limit && nextReadPos <= fCursor) {
         // data can be read from buffer.
@@ -261,6 +262,7 @@ private int readInternal(final long position, final byte[] b, final int offset,
       if (receivedBytes > 0) {
         LOG.debug("Received data from read ahead, not doing remote read");
         streamStatistics.bytesReadFromBuffer(receivedBytes);
+        incrementReadOps();
         return receivedBytes;
       }
 
@@ -301,6 +303,7 @@ int readRemote(long position, byte[] b, int offset, int length) throws IOExcepti
           position, b.length, offset, length);
       op = client.read(path, position, b, offset, length, tolerateOobAppends ? "*" : eTag);
       perfInfo.registerResult(op.getResult()).registerSuccess(true);
+      incrementReadOps();
     } catch (AzureBlobFileSystemException ex) {
       if (ex instanceof AbfsRestOperationException) {
         AbfsRestOperationException ere = (AbfsRestOperationException) ex;
@@ -318,6 +321,15 @@ int readRemote(long position, byte[] b, int offset, int length) throws IOExcepti
     return (int) bytesRead;
   }
 
+  /**
+   * Increment Read Operations.
+   */
+  private void incrementReadOps() {
+    if (statistics != null) {
+      statistics.incrementReadOps(1);
+    }
+  }
+
   /**
    * Seek to given position in stream.
    * @param n position to seek to

hadoop-tools/hadoop-azure/src/main/java/org/apache/hadoop/fs/azurebfs/services/AbfsOutputStream.java

@@ -39,6 +39,7 @@
 import org.apache.hadoop.fs.azurebfs.contracts.exceptions.AbfsRestOperationException;
 import org.apache.hadoop.fs.azurebfs.contracts.exceptions.AzureBlobFileSystemException;
 import org.apache.hadoop.io.ElasticByteBufferPool;
+import org.apache.hadoop.fs.FileSystem.Statistics;
 import org.apache.hadoop.fs.FSExceptionMessages;
 import org.apache.hadoop.fs.StreamCapabilities;
 import org.apache.hadoop.fs.Syncable;
@@ -78,14 +79,18 @@ public class AbfsOutputStream extends OutputStream implements Syncable, StreamCa
   private final ElasticByteBufferPool byteBufferPool
           = new ElasticByteBufferPool();
 
+  private final Statistics statistics;
+
   public AbfsOutputStream(
       final AbfsClient client,
+      final Statistics statistics,
       final String path,
       final long position,
       final int bufferSize,
       final boolean supportFlush,
       final boolean disableOutputStreamFlush) {
     this.client = client;
+    this.statistics = statistics;
     this.path = path;
     this.position = position;
     this.closed = false;
@@ -181,6 +186,16 @@ public synchronized void write(final byte[] data, final int off, final int lengt
 
       writableBytes = bufferSize - bufferIndex;
     }
+    incrementWriteOps();
+  }
+
+  /**
+   * Increment Write Operations.
+   */
+  private void incrementWriteOps() {
+    if (statistics != null) {
+      statistics.incrementWriteOps(1);
+    }
   }
 
   /**

hadoop-tools/hadoop-azure/src/test/java/org/apache/hadoop/fs/azurebfs/AbstractAbfsTestWithTimeout.java

@@ -17,12 +17,19 @@
  */
 package org.apache.hadoop.fs.azurebfs;
 
+import java.io.IOException;
+
 import org.junit.Assert;
 import org.junit.Before;
 import org.junit.BeforeClass;
 import org.junit.Rule;
 import org.junit.rules.TestName;
 import org.junit.rules.Timeout;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import org.apache.hadoop.fs.FSDataInputStream;
+import org.apache.hadoop.fs.Path;
 
 import static org.apache.hadoop.fs.azurebfs.constants.TestConfigurationKeys.TEST_TIMEOUT;
 
@@ -31,6 +38,9 @@
  * This class does not attempt to bind to Azure.
  */
 public class AbstractAbfsTestWithTimeout extends Assert {
+  private static final Logger LOG =
+      LoggerFactory.getLogger(AbstractAbfsTestWithTimeout.class);
+
   /**
    * The name of the current method.
    */
@@ -67,4 +77,53 @@ public void nameThread() {
   protected int getTestTimeoutMillis() {
     return TEST_TIMEOUT;
   }
+
+  /**
+   * Describe a test in the logs.
+   *
+   * @param text text to print
+   * @param args arguments to format in the printing
+   */
+  protected void describe(String text, Object... args) {
+    LOG.info("\n\n{}: {}\n",
+        methodName.getMethodName(),
+        String.format(text, args));
+  }
+
+  /**
+   * Validate Contents written on a file in Abfs.
+   *
+   * @param fs                AzureBlobFileSystem
+   * @param path              Path of the file
+   * @param originalByteArray original byte array
+   * @return if content is validated true else, false
+   * @throws IOException
+   */
+  protected boolean validateContent(AzureBlobFileSystem fs, Path path,
+      byte[] originalByteArray)
+      throws IOException {
+    int pos = 0;
+    int lenOfOriginalByteArray = originalByteArray.length;
+
+    try (FSDataInputStream in = fs.open(path)) {
+      byte valueOfContentAtPos = (byte) in.read();
+
+      while (valueOfContentAtPos != -1 && pos < lenOfOriginalByteArray) {
+        if (originalByteArray[pos] != valueOfContentAtPos) {
+          assertEquals("Mismatch in content validation at position {}", pos,
+              originalByteArray[pos], valueOfContentAtPos);
+          return false;
+        }
+        valueOfContentAtPos = (byte) in.read();
+        pos++;
+      }
+      if (valueOfContentAtPos != -1) {
+        assertEquals("Expected end of file", -1, valueOfContentAtPos);
+        return false;
+      }
+      return true;
+    }
+
+  }
+
 }