HDFS-11644. Support for querying outputstream capabilities. Contributed by Manoj Govindassamy.

This commit is contained in:
Andrew Wang 2017-05-13 21:41:11 -07:00
parent aeca471ec1
commit 18b89e902b
5 changed files with 126 additions and 3 deletions

View File

@ -31,7 +31,7 @@ import org.apache.hadoop.classification.InterfaceStability;
@InterfaceAudience.Public
@InterfaceStability.Stable
public class FSDataOutputStream extends DataOutputStream
implements Syncable, CanSetDropBehind {
implements Syncable, CanSetDropBehind, StreamCapabilities {
private final OutputStream wrappedStream;
private static class PositionCache extends FilterOutputStream {
@ -116,6 +116,14 @@ public class FSDataOutputStream extends DataOutputStream
return wrappedStream;
}
@Override
public boolean hasCapability(String capability) {
if (wrappedStream instanceof StreamCapabilities) {
return ((StreamCapabilities) wrappedStream).hasCapability(capability);
}
return false;
}
@Override // Syncable
@Deprecated
public void sync() throws IOException {

View File

@ -0,0 +1,67 @@
/**
* 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.fs;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
/**
* Interface to query streams for supported capabilities.
*/
@InterfaceAudience.Public
@InterfaceStability.Evolving
public interface StreamCapabilities {
/**
* Capabilities that a stream can support and be queried for.
*/
enum StreamCapability {
/**
* Stream hflush capability to flush out the data in client's buffer.
* Streams with this capability implement {@link Syncable} and support
* {@link Syncable#hflush()}.
*/
HFLUSH("hflush"),
/**
* Stream hsync capability to flush out the data in client's buffer and
* the disk device. Streams with this capability implement {@link Syncable}
* and support {@link Syncable#hsync()}.
*/
HSYNC("hsync");
private final String capability;
StreamCapability(String value) {
this.capability = value;
}
public final String getValue() {
return capability;
}
}
/**
* Query the stream for a specific capability.
*
* @param capability string to query the stream support for.
* @return True if the stream supports capability.
*/
boolean hasCapability(String capability);
}

View File

@ -17,6 +17,9 @@
*/
package org.apache.hadoop.hdfs;
import static org.apache.hadoop.fs.StreamCapabilities.StreamCapability.HFLUSH;
import static org.apache.hadoop.fs.StreamCapabilities.StreamCapability.HSYNC;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InterruptedIOException;
@ -33,6 +36,7 @@ import org.apache.hadoop.fs.FSOutputSummer;
import org.apache.hadoop.fs.FileAlreadyExistsException;
import org.apache.hadoop.fs.FileEncryptionInfo;
import org.apache.hadoop.fs.ParentNotDirectoryException;
import org.apache.hadoop.fs.StreamCapabilities;
import org.apache.hadoop.fs.Syncable;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.hdfs.client.HdfsClientConfigKeys;
@ -88,7 +92,7 @@ import com.google.common.base.Preconditions;
****************************************************************/
@InterfaceAudience.Private
public class DFSOutputStream extends FSOutputSummer
implements Syncable, CanSetDropBehind {
implements Syncable, CanSetDropBehind, StreamCapabilities {
static final Logger LOG = LoggerFactory.getLogger(DFSOutputStream.class);
/**
* Number of times to retry creating a file when there are transient
@ -554,6 +558,15 @@ public class DFSOutputStream extends FSOutputSummer
}
}
@Override
public boolean hasCapability(String capability) {
if (capability.equalsIgnoreCase(HSYNC.getValue()) ||
capability.equalsIgnoreCase((HFLUSH.getValue()))) {
return true;
}
return false;
}
/**
* Flush/Sync buffered data to DataNodes.
*

View File

@ -17,9 +17,11 @@
*/
package org.apache.hadoop.hdfs;
import java.io.ByteArrayInputStream;
import java.io.DataOutputStream;
import java.io.File;
import java.io.IOException;
import java.io.InputStream;
import java.lang.reflect.Field;
import java.lang.reflect.InvocationTargetException;
import java.lang.reflect.Method;
@ -32,8 +34,10 @@ import java.util.Random;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.CreateFlag;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.FsTracer;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.StreamCapabilities.StreamCapability;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.hdfs.DataStreamer.LastExceptionInStreamer;
import org.apache.hadoop.hdfs.client.impl.DfsClientConf;
@ -48,6 +52,7 @@ import org.apache.hadoop.hdfs.server.blockmanagement.BlockManager;
import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeDescriptor;
import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeManager;
import org.apache.hadoop.hdfs.server.protocol.DatanodeStorage;
import org.apache.hadoop.io.IOUtils;
import org.apache.hadoop.test.GenericTestUtils;
import org.apache.hadoop.test.PathUtils;
import org.apache.htrace.core.SpanId;
@ -55,6 +60,8 @@ import org.junit.AfterClass;
import org.junit.Assert;
import org.junit.BeforeClass;
import org.junit.Test;
import static org.junit.Assert.assertTrue;
import static org.mockito.Matchers.anyBoolean;
import static org.mockito.Matchers.anyLong;
import org.mockito.Mockito;
@ -346,6 +353,24 @@ public class TestDFSOutputStream {
verify(spyClient, times(1)).endFileLease(anyLong());
}
@Test
public void testStreamFlush() throws Exception {
FileSystem fs = cluster.getFileSystem();
FSDataOutputStream os = fs.create(new Path("/normal-file"));
// Verify output stream supports hsync() and hflush().
assertTrue("DFSOutputStream should support hflush()!",
os.hasCapability(StreamCapability.HFLUSH.getValue()));
assertTrue("DFSOutputStream should support hsync()!",
os.hasCapability(StreamCapability.HSYNC.getValue()));
byte[] bytes = new byte[1024];
InputStream is = new ByteArrayInputStream(bytes);
IOUtils.copyBytes(is, os, bytes.length);
os.hflush();
IOUtils.copyBytes(is, os, bytes.length);
os.hsync();
os.close();
}
@AfterClass
public static void tearDown() {
if (cluster != null) {

View File

@ -22,6 +22,7 @@ import java.io.DataOutputStream;
import java.io.IOException;
import java.io.OutputStream;
import org.apache.hadoop.fs.StreamCapabilities;
import org.apache.hadoop.fs.Syncable;
/**
@ -30,12 +31,21 @@ import org.apache.hadoop.fs.Syncable;
* wrapped stream passed in to the constructor. This is required
* for HBase when wrapping a PageBlobOutputStream used as a write-ahead log.
*/
public class SyncableDataOutputStream extends DataOutputStream implements Syncable {
public class SyncableDataOutputStream extends DataOutputStream
implements Syncable, StreamCapabilities {
public SyncableDataOutputStream(OutputStream out) {
super(out);
}
@Override
public boolean hasCapability(String capability) {
if (out instanceof StreamCapabilities) {
return ((StreamCapabilities) out).hasCapability(capability);
}
return false;
}
@Override
@Deprecated
public void sync() throws IOException {