Apache
/
hadoop
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge trunk into HDFS-3077 branch. 

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/branches/HDFS-3077@1377092 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Aaron Myers 2012-08-24 20:38:08 +00:00
parent aa6f18e81b 92cb6b093c
commit a8ff292669
253 changed files with 5903 additions and 3211 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
hadoop-common-project/hadoop-annotations/pom.xml
View File

@ -34,7 +34,7 @@
<dependency>
<groupId>jdiff</groupId>
<artifactId>jdiff</artifactId>
<scope>compile</scope>
<scope>provided</scope>
</dependency>
</dependencies>

9
hadoop-common-project/hadoop-auth-examples/pom.xml
View File

@ -42,15 +42,20 @@
<artifactId>hadoop-auth</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.slf4j</groupId>
<artifactId>slf4j-api</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>log4j</groupId>
<artifactId>log4j</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
<dependency>
<groupId>org.slf4j</groupId>
<artifactId>slf4j-log4j12</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
</dependencies>

5
hadoop-common-project/hadoop-auth/pom.xml
View File

@ -38,6 +38,7 @@
<dependencies>
<dependency>
<!-- Used, even though 'mvn dependency:analyze' doesn't find it -->
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-annotations</artifactId>
<scope>provided</scope>
@ -75,12 +76,12 @@
<dependency>
<groupId>log4j</groupId>
<artifactId>log4j</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
<dependency>
<groupId>org.slf4j</groupId>
<artifactId>slf4j-log4j12</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
</dependencies>

88
hadoop-common-project/hadoop-common/CHANGES.txt
View File

@ -91,6 +91,10 @@ Trunk (unreleased changes)
HADOOP-8624. ProtobufRpcEngine should log all RPCs if TRACE logging is
enabled (todd)
HADOOP-8711. IPC Server supports adding exceptions for which
the message is printed and the stack trace is not printed to avoid chatter.
(Brandon Li via Suresh)
BUG FIXES
HADOOP-8177. MBeans shouldn't try to register when it fails to create MBeanName.
@ -198,6 +202,10 @@ Branch-2 ( Unreleased changes )
HADOOP-8388. Remove unused BlockLocation serialization.
(Colin Patrick McCabe via eli)
HADOOP-8689. Make trash a server side configuration option. (eli)
HADOOP-8710. Remove ability for users to easily run the trash emptire. (eli)
NEW FEATURES
HDFS-3042. Automatic failover support for NameNode HA (todd)
@ -218,6 +226,9 @@ Branch-2 ( Unreleased changes )
HADOOP-7754. Expose file descriptors from Hadoop-wrapped local
FileSystems (todd and ahmed via tucu)
HADOOP-8240. Add a new API to allow users to specify a checksum type
on FileSystem.create(..). (Kihwal Lee via szetszwo)
IMPROVEMENTS
HADOOP-8340. SNAPSHOT build versions should compare as less than their eventual
@ -289,6 +300,20 @@ Branch-2 ( Unreleased changes )
HADOOP-8687. Upgrade log4j to 1.2.17. (eli)
HADOOP-8278. Make sure components declare correct set of dependencies.
(tomwhite)
HADOOP-8700. Use enum to define the checksum constants in DataChecksum.
(szetszwo)
HADOOP-8686. Fix warnings in native code. (Colin Patrick McCabe via eli)
HADOOP-8239. Add subclasses of MD5MD5CRC32FileChecksum to support file
checksum with CRC32C. (Kihwal Lee via szetszwo)
HADOOP-8075. Lower native-hadoop library log from info to debug.
(Hızır Sefa İrken via eli)
BUG FIXES
HADOOP-8372. NetUtils.normalizeHostName() incorrectly handles hostname
@ -389,6 +414,31 @@ Branch-2 ( Unreleased changes )
HADOOP-8659. Native libraries must build with soft-float ABI for Oracle JVM
on ARM. (Trevor Robinson via todd)
HADOOP-8654. TextInputFormat delimiter bug (Gelesh and Jason Lowe via
bobby)
HADOOP-8614. IOUtils#skipFully hangs forever on EOF.
(Colin Patrick McCabe via eli)
HADOOP-8720. TestLocalFileSystem should use test root subdirectory.
(Vlad Rozov via eli)
HADOOP-8721. ZKFC should not retry 45 times when attempting a graceful
fence during a failover. (Vinayakumar B via atm)
HADOOP-8632. Configuration leaking class-loaders (Costin Leau via bobby)
HADOOP-4572. Can not access user logs - Jetty is not configured by default
to serve aliases/symlinks (ahmed via tucu)
HADOOP-8660. TestPseudoAuthenticator failing with NPE. (tucu)
HADOOP-8699. some common testcases create core-site.xml in test-classes
making other testcases to fail. (tucu)
HADOOP-8031. Configuration class fails to find embedded .jar resources;
should use URL.openStream() (genman via tucu)
BREAKDOWN OF HDFS-3042 SUBTASKS
HADOOP-8220. ZKFailoverController doesn't handle failure to become active
@ -421,11 +471,6 @@ Branch-2 ( Unreleased changes )
HADOOP-8405. ZKFC tests leak ZK instances. (todd)
HADOOP-8660. TestPseudoAuthenticator failing with NPE. (tucu)
HADOOP-8699. some common testcases create core-site.xml in test-classes
making other testcases to fail. (tucu)
Release 2.0.0-alpha - 05-23-2012
INCOMPATIBLE CHANGES
@ -793,10 +838,15 @@ Release 2.0.0-alpha - 05-23-2012
HADOOP-7868. Hadoop native fails to compile when default linker
option is -Wl,--as-needed. (Trevor Robinson via eli)
HADOOP-8655. Fix TextInputFormat for large deliminators. (Gelesh via
bobby)
Release 0.23.3 - UNRELEASED
INCOMPATIBLE CHANGES
HADOOP-7967. Need generalized multi-token filesystem support (daryn)
NEW FEATURES
IMPROVEMENTS
@ -899,6 +949,34 @@ Release 0.23.3 - UNRELEASED
HADOOP-8633. Interrupted FsShell copies may leave tmp files (Daryn Sharp
via tgraves)
HADOOP-8703. distcpV2: turn CRC checking off for 0 byte size (Dave
Thompson via bobby)
HADOOP-8390. TestFileSystemCanonicalization fails with JDK7 (Trevor
Robinson via tgraves)
HADOOP-8692. TestLocalDirAllocator fails intermittently with JDK7
(Trevor Robinson via tgraves)
HADOOP-8693. TestSecurityUtil fails intermittently with JDK7 (Trevor
Robinson via tgraves)
HADOOP-8697. TestWritableName fails intermittently with JDK7 (Trevor
Robinson via tgraves)
HADOOP-8695. TestPathData fails intermittently with JDK7 (Trevor
Robinson via tgraves)
HADOOP-8611. Allow fall-back to the shell-based implementation when
JNI-based users-group mapping fails (Robert Parker via bobby)
HADOOP-8225. DistCp fails when invoked by Oozie (daryn via bobby)
HADOOP-8709. globStatus changed behavior from 0.20/1.x (Jason Lowe via
bobby)
HADOOP-8725. MR is broken when security is off (daryn via bobby)
Release 0.23.2 - UNRELEASED
INCOMPATIBLE CHANGES

73
hadoop-common-project/hadoop-common/pom.xml
View File

@ -74,13 +74,13 @@
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-net</groupId>
<artifactId>commons-net</artifactId>
<groupId>commons-io</groupId>
<artifactId>commons-io</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-io</groupId>
<artifactId>commons-io</artifactId>
<groupId>commons-net</groupId>
<artifactId>commons-net</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
@ -98,17 +98,13 @@
<artifactId>jetty-util</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>asm</groupId>
<artifactId>asm</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>com.sun.jersey</groupId>
<artifactId>jersey-core</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<!-- Used, even though 'mvn dependency:analyze' doesn't find it -->
<groupId>com.sun.jersey</groupId>
<artifactId>jersey-json</artifactId>
<scope>compile</scope>
@ -121,33 +117,28 @@
<dependency>
<groupId>tomcat</groupId>
<artifactId>jasper-compiler</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
<dependency>
<groupId>tomcat</groupId>
<artifactId>jasper-runtime</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
<dependency>
<groupId>javax.servlet.jsp</groupId>
<artifactId>jsp-api</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
<dependency>
<groupId>commons-el</groupId>
<artifactId>commons-el</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
<dependency>
<groupId>commons-logging</groupId>
<artifactId>commons-logging</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-logging</groupId>
<artifactId>commons-logging-api</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>log4j</groupId>
<artifactId>log4j</artifactId>
@ -158,26 +149,6 @@
<artifactId>jets3t</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.apache.mina</groupId>
<artifactId>mina-core</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.ftpserver</groupId>
<artifactId>ftplet-api</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.ftpserver</groupId>
<artifactId>ftpserver-core</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.ftpserver</groupId>
<artifactId>ftpserver-deprecated</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
@ -188,11 +159,6 @@
<artifactId>commons-lang</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-collections</groupId>
<artifactId>commons-collections</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-configuration</groupId>
<artifactId>commons-configuration</artifactId>
@ -206,16 +172,11 @@
<dependency>
<groupId>org.slf4j</groupId>
<artifactId>slf4j-log4j12</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
<dependency>
<groupId>org.eclipse.jdt</groupId>
<artifactId>core</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>oro</groupId>
<artifactId>oro</artifactId>
<groupId>org.codehaus.jackson</groupId>
<artifactId>jackson-core-asl</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
@ -223,11 +184,6 @@
<artifactId>jackson-mapper-asl</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.aspectj</groupId>
<artifactId>aspectjrt</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.mockito</groupId>
<artifactId>mockito-all</artifactId>
@ -258,11 +214,6 @@
<artifactId>hadoop-auth</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>com.googlecode.json-simple</groupId>
<artifactId>json-simple</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>com.jcraft</groupId>
<artifactId>jsch</artifactId>

78
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/conf/Configuration.java
View File

@ -30,6 +30,7 @@
import java.io.OutputStreamWriter;
import java.io.Reader;
import java.io.Writer;
import java.lang.ref.WeakReference;
import java.net.InetSocketAddress;
import java.net.URL;
import java.util.ArrayList;
@ -219,8 +220,8 @@ public String toString() {
private static final CopyOnWriteArrayList<String> defaultResources =
new CopyOnWriteArrayList<String>();
private static final Map<ClassLoader, Map<String, Class<?>>>
CACHE_CLASSES = new WeakHashMap<ClassLoader, Map<String, Class<?>>>();
private static final Map<ClassLoader, Map<String, WeakReference<Class<?>>>>
CACHE_CLASSES = new WeakHashMap<ClassLoader, Map<String, WeakReference<Class<?>>>>();
/**
* Sentinel value to store negative cache results in {@link #CACHE_CLASSES}.
@ -1531,28 +1532,33 @@ public Class<?> getClassByName(String name) throws ClassNotFoundException {
* @return the class object, or null if it could not be found.
*/
public Class<?> getClassByNameOrNull(String name) {
Map<String, Class<?>> map;
Map<String, WeakReference<Class<?>>> map;
synchronized (CACHE_CLASSES) {
map = CACHE_CLASSES.get(classLoader);
if (map == null) {
map = Collections.synchronizedMap(
new WeakHashMap<String, Class<?>>());
new WeakHashMap<String, WeakReference<Class<?>>>());
CACHE_CLASSES.put(classLoader, map);
}
}
Class<?> clazz = map.get(name);
Class<?> clazz = null;
WeakReference<Class<?>> ref = map.get(name);
if (ref != null) {
clazz = ref.get();
}
if (clazz == null) {
try {
clazz = Class.forName(name, true, classLoader);
} catch (ClassNotFoundException e) {
// Leave a marker that the class isn't found
map.put(name, NEGATIVE_CACHE_SENTINEL);
map.put(name, new WeakReference<Class<?>>(NEGATIVE_CACHE_SENTINEL));
return null;
}
// two putters can race here, but they'll put the same class
map.put(name, clazz);
map.put(name, new WeakReference<Class<?>>(clazz));
return clazz;
} else if (clazz == NEGATIVE_CACHE_SENTINEL) {
return null; // not found
@ -1856,6 +1862,32 @@ public Iterator<Map.Entry<String, String>> iterator() {
return result.entrySet().iterator();
}
private Document parse(DocumentBuilder builder, URL url)
throws IOException, SAXException {
if (!quietmode) {
LOG.info("parsing URL " + url);
}
if (url == null) {
return null;
}
return parse(builder, url.openStream());
}
private Document parse(DocumentBuilder builder, InputStream is)
throws IOException, SAXException {
if (!quietmode) {
LOG.info("parsing input stream " + is);
}
if (is == null) {
return null;
}
try {
return builder.parse(is);
} finally {
is.close();
}
}
private void loadResources(Properties properties,
ArrayList<Resource> resources,
boolean quiet) {
@ -1905,21 +1937,10 @@ private Resource loadResource(Properties properties, Resource wrapper, boolean q
boolean returnCachedProperties = false;
if (resource instanceof URL) { // an URL resource
URL url = (URL)resource;
if (url != null) {
if (!quiet) {
LOG.info("parsing " + url);
}
doc = builder.parse(url.toString());
}
doc = parse(builder, (URL)resource);
} else if (resource instanceof String) { // a CLASSPATH resource
URL url = getResource((String)resource);
if (url != null) {
if (!quiet) {
LOG.info("parsing " + url);
}
doc = builder.parse(url.toString());
}
doc = parse(builder, url);
} else if (resource instanceof Path) { // a file resource
// Can't use FileSystem API or we get an infinite loop
// since FileSystem uses Configuration API. Use java.io.File instead.
@ -1927,22 +1948,13 @@ private Resource loadResource(Properties properties, Resource wrapper, boolean q
.getAbsoluteFile();
if (file.exists()) {
if (!quiet) {
LOG.info("parsing " + file);
}
InputStream in = new BufferedInputStream(new FileInputStream(file));
try {
doc = builder.parse(in);
} finally {
in.close();
LOG.info("parsing File " + file);
}
doc = parse(builder, new BufferedInputStream(new FileInputStream(file)));
}
} else if (resource instanceof InputStream) {
try {
doc = builder.parse((InputStream)resource);
returnCachedProperties = true;
} finally {
((InputStream)resource).close();
}
doc = parse(builder, (InputStream) resource);
returnCachedProperties = true;
} else if (resource instanceof Properties) {
overlay(properties, (Properties)resource);
} else if (resource instanceof Element) {

26
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/AbstractFileSystem.java
View File

@ -39,6 +39,7 @@
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem.Statistics;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.Options.CreateOpts;
import org.apache.hadoop.fs.Options.Rename;
import org.apache.hadoop.fs.permission.FsPermission;
@ -46,6 +47,7 @@
import org.apache.hadoop.security.AccessControlException;
import org.apache.hadoop.security.SecurityUtil;
import org.apache.hadoop.security.token.Token;
import org.apache.hadoop.util.DataChecksum;
import org.apache.hadoop.util.Progressable;
/**
@ -467,6 +469,7 @@ public final FSDataOutputStream create(final Path f,
short replication = -1;
long blockSize = -1;
int bytesPerChecksum = -1;
ChecksumOpt checksumOpt = null;
FsPermission permission = null;
Progressable progress = null;
Boolean createParent = null;
@ -496,6 +499,12 @@ public final FSDataOutputStream create(final Path f,
"BytesPerChecksum option is set multiple times");
}
bytesPerChecksum = ((CreateOpts.BytesPerChecksum) iOpt).getValue();
} else if (CreateOpts.ChecksumParam.class.isInstance(iOpt)) {
if (checksumOpt != null) {
throw new HadoopIllegalArgumentException(
"CreateChecksumType option is set multiple times");
}
checksumOpt = ((CreateOpts.ChecksumParam) iOpt).getValue();
} else if (CreateOpts.Perms.class.isInstance(iOpt)) {
if (permission != null) {
throw new HadoopIllegalArgumentException(
@ -533,9 +542,16 @@ public final FSDataOutputStream create(final Path f,
if (blockSize == -1) {
blockSize = ssDef.getBlockSize();
}
if (bytesPerChecksum == -1) {
bytesPerChecksum = ssDef.getBytesPerChecksum();
}
// Create a checksum option honoring user input as much as possible.
// If bytesPerChecksum is specified, it will override the one set in
// checksumOpt. Any missing value will be filled in using the default.
ChecksumOpt defaultOpt = new ChecksumOpt(
ssDef.getChecksumType(),
ssDef.getBytesPerChecksum());
checksumOpt = ChecksumOpt.processChecksumOpt(defaultOpt,
checksumOpt, bytesPerChecksum);
if (bufferSize == -1) {
bufferSize = ssDef.getFileBufferSize();
}
@ -552,7 +568,7 @@ public final FSDataOutputStream create(final Path f,
}
return this.createInternal(f, createFlag, permission, bufferSize,
replication, blockSize, progress, bytesPerChecksum, createParent);
replication, blockSize, progress, checksumOpt, createParent);
}
/**
@ -563,7 +579,7 @@ public final FSDataOutputStream create(final Path f,
public abstract FSDataOutputStream createInternal(Path f,
EnumSet<CreateFlag> flag, FsPermission absolutePermission,
int bufferSize, short replication, long blockSize, Progressable progress,
int bytesPerChecksum, boolean createParent)
ChecksumOpt checksumOpt, boolean createParent)
throws AccessControlException, FileAlreadyExistsException,
FileNotFoundException, ParentNotDirectoryException,
UnsupportedFileSystemException, UnresolvedLinkException, IOException;

16
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/ChecksumFs.java
View File

@ -28,6 +28,7 @@
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.util.Progressable;
import org.apache.hadoop.util.PureJavaCrc32;
@ -324,13 +325,17 @@ public ChecksumFSOutputSummer(final ChecksumFs fs, final Path file,
final EnumSet<CreateFlag> createFlag,
final FsPermission absolutePermission, final int bufferSize,
final short replication, final long blockSize,
final Progressable progress, final int bytesPerChecksum,
final Progressable progress, final ChecksumOpt checksumOpt,
final boolean createParent) throws IOException {
super(new PureJavaCrc32(), fs.getBytesPerSum(), 4);
// checksumOpt is passed down to the raw fs. Unless it implements
// checksum impelemts internally, checksumOpt will be ignored.
// If the raw fs does checksum internally, we will end up with
// two layers of checksumming. i.e. checksumming checksum file.
this.datas = fs.getRawFs().createInternal(file, createFlag,
absolutePermission, bufferSize, replication, blockSize, progress,
bytesPerChecksum, createParent);
checksumOpt, createParent);
// Now create the chekcsumfile; adjust the buffsize
int bytesPerSum = fs.getBytesPerSum();
@ -338,7 +343,7 @@ public ChecksumFSOutputSummer(final ChecksumFs fs, final Path file,
this.sums = fs.getRawFs().createInternal(fs.getChecksumFile(file),
EnumSet.of(CreateFlag.CREATE, CreateFlag.OVERWRITE),
absolutePermission, sumBufferSize, replication, blockSize, progress,
bytesPerChecksum, createParent);
checksumOpt, createParent);
sums.write(CHECKSUM_VERSION, 0, CHECKSUM_VERSION.length);
sums.writeInt(bytesPerSum);
}
@ -361,12 +366,11 @@ protected void writeChunk(byte[] b, int offset, int len, byte[] checksum)
public FSDataOutputStream createInternal(Path f,
EnumSet<CreateFlag> createFlag, FsPermission absolutePermission,
int bufferSize, short replication, long blockSize, Progressable progress,
int bytesPerChecksum, boolean createParent) throws IOException {
ChecksumOpt checksumOpt, boolean createParent) throws IOException {
final FSDataOutputStream out = new FSDataOutputStream(
new ChecksumFSOutputSummer(this, f, createFlag, absolutePermission,
bufferSize, replication, blockSize, progress,
bytesPerChecksum, createParent), null);
checksumOpt, createParent), null);
return out;
}

11
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/CommonConfigurationKeys.java
View File

@ -154,6 +154,11 @@ public class CommonConfigurationKeys extends CommonConfigurationKeysPublic {
"ha.failover-controller.graceful-fence.rpc-timeout.ms";
public static final int HA_FC_GRACEFUL_FENCE_TIMEOUT_DEFAULT = 5000;
/* FC connection retries for graceful fencing */
public static final String HA_FC_GRACEFUL_FENCE_CONNECTION_RETRIES =
"ha.failover-controller.graceful-fence.connection.retries";
public static final int HA_FC_GRACEFUL_FENCE_CONNECTION_RETRIES_DEFAULT = 1;
/* Timeout that the CLI (manual) FC waits for monitorHealth, getServiceState */
public static final String HA_FC_CLI_CHECK_TIMEOUT_KEY =
"ha.failover-controller.cli-check.rpc-timeout.ms";
@ -167,6 +172,12 @@ public class CommonConfigurationKeys extends CommonConfigurationKeysPublic {
public static final String DEFAULT_HADOOP_HTTP_STATIC_USER =
"dr.who";
/** Enable/Disable aliases serving from jetty */
public static final String HADOOP_JETTY_LOGS_SERVE_ALIASES =
"hadoop.jetty.logs.serve.aliases";
public static final boolean DEFAULT_HADOOP_JETTY_LOGS_SERVE_ALIASES =
true;
/* Path to the Kerberos ticket cache. Setting this will force
* UserGroupInformation to use only this ticket cache file when creating a
* FileSystem instance.

10
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/DelegateToFileSystem.java
View File

@ -21,12 +21,14 @@
import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;
import java.util.Arrays;
import java.util.EnumSet;
import java.util.List;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.security.token.Token;
@ -61,7 +63,7 @@ public Path getInitialWorkingDirectory() {
public FSDataOutputStream createInternal (Path f,
EnumSet<CreateFlag> flag, FsPermission absolutePermission, int bufferSize,
short replication, long blockSize, Progressable progress,
int bytesPerChecksum, boolean createParent) throws IOException {
ChecksumOpt checksumOpt, boolean createParent) throws IOException {
checkPath(f);
// Default impl assumes that permissions do not matter
@ -80,8 +82,8 @@ public FSDataOutputStream createInternal (Path f,
}
// parent does exist - go ahead with create of file.
}
return fsImpl.primitiveCreate(f, absolutePermission, flag,
bufferSize, replication, blockSize, progress, bytesPerChecksum);
return fsImpl.primitiveCreate(f, absolutePermission, flag,
bufferSize, replication, blockSize, progress, checksumOpt);
}
@Override
@ -217,6 +219,6 @@ public String getCanonicalServiceName() {
@Override //AbstractFileSystem
public List<Token<?>> getDelegationTokens(String renewer) throws IOException {
return fsImpl.getDelegationTokens(renewer);
return Arrays.asList(fsImpl.addDelegationTokens(renewer, null));
}
}

6
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/DelegationTokenRenewer.java
View File

@ -110,7 +110,11 @@ private boolean renew() throws IOException, InterruptedException {
fs.getRenewToken().renew(fs.getConf());
} catch (IOException ie) {
try {
fs.setDelegationToken(fs.getDelegationTokens(null).get(0));
Token<?>[] tokens = fs.addDelegationTokens(null, null);
if (tokens.length == 0) {
throw new IOException("addDelegationTokens returned no tokens");
}
fs.setDelegationToken(tokens[0]);
} catch (IOException ie2) {
throw new IOException("Can't renew or get new delegation token ", ie);
}

18
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/FileContext.java
View File

@ -127,7 +127,8 @@
* <li> replication factor
* <li> block size
* <li> buffer size
* <li> bytesPerChecksum (if used).
* <li> encryptDataTransfer
* <li> checksum option. (checksumType and bytesPerChecksum)
* </ul>
*
* <p>
@ -613,7 +614,8 @@ public Path makeQualified(final Path path) {
* <li>BufferSize - buffersize used in FSDataOutputStream
* <li>Blocksize - block size for file blocks
* <li>ReplicationFactor - replication for blocks
* <li>BytesPerChecksum - bytes per checksum
* <li>ChecksumParam - Checksum parameters. server default is used
* if not specified.
* </ul>
* </ul>
*
@ -2012,7 +2014,11 @@ private FileStatus[] globStatusInternal(final URI uri,
new GlobFilter(components[components.length - 1], filter);
if (fp.hasPattern()) { // last component has a pattern
// list parent directories and then glob the results
results = listStatus(parentPaths, fp);
try {
results = listStatus(parentPaths, fp);
} catch (FileNotFoundException e) {
results = null;
}
hasGlob[0] = true;
} else { // last component does not have a pattern
// get all the path names
@ -2063,7 +2069,11 @@ private Path[] globPathsLevel(Path[] parents, String[] filePattern,
}
GlobFilter fp = new GlobFilter(filePattern[level]);
if (fp.hasPattern()) {
parents = FileUtil.stat2Paths(listStatus(parents, fp));
try {
parents = FileUtil.stat2Paths(listStatus(parents, fp));
} catch (FileNotFoundException e) {
parents = null;
}
hasGlob[0] = true;
} else {
for (int i = 0; i < parents.length; i++) {

190
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/FileSystem.java
View File

@ -45,18 +45,23 @@
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.conf.Configured;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.Options.Rename;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.io.MultipleIOException;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.net.NetUtils;
import org.apache.hadoop.security.Credentials;
import org.apache.hadoop.security.SecurityUtil;
import org.apache.hadoop.security.UserGroupInformation;
import org.apache.hadoop.security.token.Token;
import org.apache.hadoop.util.DataChecksum;
import org.apache.hadoop.util.Progressable;
import org.apache.hadoop.util.ReflectionUtils;
import org.apache.hadoop.util.ShutdownHookManager;
import com.google.common.annotations.VisibleForTesting;
/****************************************************************
* An abstract base class for a fairly generic filesystem. It
* may be implemented as a distributed filesystem, or as a "local"
@ -222,15 +227,25 @@ protected int getDefaultPort() {
/**
* Get a canonical service name for this file system. The token cache is
* the only user of this value, and uses it to lookup this filesystem's
* service tokens. The token cache will not attempt to acquire tokens if the
* service is null.
* the only user of the canonical service name, and uses it to lookup this
* filesystem's service tokens.
* If file system provides a token of its own then it must have a canonical
* name, otherwise canonical name can be null.
*
* Default Impl: If the file system has child file systems
* (such as an embedded file system) then it is assumed that the fs has no
* tokens of its own and hence returns a null name; otherwise a service
* name is built using Uri and port.
*
* @return a service string that uniquely identifies this file system, null
* if the filesystem does not implement tokens
* @see SecurityUtil#buildDTServiceName(URI, int)
*/
@InterfaceAudience.LimitedPrivate({ "HDFS", "MapReduce" })
public String getCanonicalServiceName() {
return SecurityUtil.buildDTServiceName(getUri(), getDefaultPort());
return (getChildFileSystems() == null)
? SecurityUtil.buildDTServiceName(getUri(), getDefaultPort())
: null;
}
/** @deprecated call #getUri() instead.*/
@ -396,68 +411,95 @@ public Path makeQualified(Path path) {
}
/**
* Deprecated - use @link {@link #getDelegationTokens(String)}
* Get a new delegation token for this file system.
* This is an internal method that should have been declared protected
* but wasn't historically.
* Callers should use {@link #addDelegationTokens(String, Credentials)}
*
* @param renewer the account name that is allowed to renew the token.
* @return a new delegation token
* @throws IOException
*/
@InterfaceAudience.LimitedPrivate({"HDFS", "MapReduce"})
@Deprecated
@InterfaceAudience.Private()
public Token<?> getDelegationToken(String renewer) throws IOException {
return null;
}
/**
* Get one or more delegation tokens associated with the filesystem. Normally
* a file system returns a single delegation token. A file system that manages
* multiple file systems underneath, could return set of delegation tokens for
* all the file systems it manages.
* Obtain all delegation tokens used by this FileSystem that are not
* already present in the given Credentials. Existing tokens will neither
* be verified as valid nor having the given renewer. Missing tokens will
* be acquired and added to the given Credentials.
*
* @param renewer the account name that is allowed to renew the token.
* Default Impl: works for simple fs with its own token
* and also for an embedded fs whose tokens are those of its
* children file system (i.e. the embedded fs has not tokens of its
* own).
*
* @param renewer the user allowed to renew the delegation tokens
* @param credentials cache in which to add new delegation tokens
* @return list of new delegation tokens
* If delegation tokens not supported then return a list of size zero.
* @throws IOException
*/
@InterfaceAudience.LimitedPrivate( { "HDFS", "MapReduce" })
public List<Token<?>> getDelegationTokens(String renewer) throws IOException {
return new ArrayList<Token<?>>(0);
}
/**
* @see #getDelegationTokens(String)
* This is similar to getDelegationTokens, with the added restriction that if
* a token is already present in the passed Credentials object - that token
* is returned instead of a new delegation token.
*
* If the token is found to be cached in the Credentials object, this API does
* not verify the token validity or the passed in renewer.
*
*
* @param renewer the account name that is allowed to renew the token.
* @param credentials a Credentials object containing already knowing
* delegationTokens.
* @return a list of delegation tokens.
* @throws IOException
*/
@InterfaceAudience.LimitedPrivate({ "HDFS", "MapReduce" })
public List<Token<?>> getDelegationTokens(String renewer,
Credentials credentials) throws IOException {
List<Token<?>> allTokens = getDelegationTokens(renewer);
List<Token<?>> newTokens = new ArrayList<Token<?>>();
if (allTokens != null) {
for (Token<?> token : allTokens) {
Token<?> knownToken = credentials.getToken(token.getService());
if (knownToken == null) {
newTokens.add(token);
} else {
newTokens.add(knownToken);
public Token<?>[] addDelegationTokens(
final String renewer, Credentials credentials) throws IOException {
if (credentials == null) {
credentials = new Credentials();
}
final List<Token<?>> tokens = new ArrayList<Token<?>>();
collectDelegationTokens(renewer, credentials, tokens);
return tokens.toArray(new Token<?>[tokens.size()]);
}
/**
* Recursively obtain the tokens for this FileSystem and all descended
* FileSystems as determined by getChildFileSystems().
* @param renewer the user allowed to renew the delegation tokens
* @param credentials cache in which to add the new delegation tokens
* @param tokens list in which to add acquired tokens
* @throws IOException
*/
private void collectDelegationTokens(final String renewer,
final Credentials credentials,
final List<Token<?>> tokens)
throws IOException {
final String serviceName = getCanonicalServiceName();
// Collect token of the this filesystem and then of its embedded children
if (serviceName != null) { // fs has token, grab it
final Text service = new Text(serviceName);
Token<?> token = credentials.getToken(service);
if (token == null) {
token = getDelegationToken(renewer);
if (token != null) {
tokens.add(token);
credentials.addToken(service, token);
}
}
}
return newTokens;
// Now collect the tokens from the children
final FileSystem[] children = getChildFileSystems();
if (children != null) {
for (final FileSystem fs : children) {
fs.collectDelegationTokens(renewer, credentials, tokens);
}
}
}
/**
* Get all the immediate child FileSystems embedded in this FileSystem.
* It does not recurse and get grand children. If a FileSystem
* has multiple child FileSystems, then it should return a unique list
* of those FileSystems. Default is to return null to signify no children.
*
* @return FileSystems used by this FileSystem
*/
@InterfaceAudience.LimitedPrivate({ "HDFS" })
@VisibleForTesting
public FileSystem[] getChildFileSystems() {
return null;
}
/** create a file with the provided permission
* The permission of the file is set to be the provided permission as in
* setPermission, not permission&~umask
@ -616,12 +658,17 @@ public BlockLocation[] getFileBlockLocations(Path p,
@Deprecated
public FsServerDefaults getServerDefaults() throws IOException {
Configuration conf = getConf();
// CRC32 is chosen as default as it is available in all
// releases that support checksum.
// The client trash configuration is ignored.
return new FsServerDefaults(getDefaultBlockSize(),
conf.getInt("io.bytes.per.checksum", 512),
64 * 1024,
getDefaultReplication(),
conf.getInt("io.file.buffer.size", 4096),
false);
false,
CommonConfigurationKeysPublic.FS_TRASH_INTERVAL_DEFAULT,
DataChecksum.Type.CRC32);
}
/**
@ -847,11 +894,40 @@ public FSDataOutputStream create(Path f,
short replication,
long blockSize,
Progressable progress) throws IOException {
// only DFS support this
return create(f, permission, flags.contains(CreateFlag.OVERWRITE), bufferSize, replication, blockSize, progress);
return create(f, permission, flags, bufferSize, replication,
blockSize, progress, null);
}
/**
* Create an FSDataOutputStream at the indicated Path with a custom
* checksum option
* @param f the file name to open
* @param permission
* @param flags {@link CreateFlag}s to use for this stream.
* @param bufferSize the size of the buffer to be used.
* @param replication required block replication for the file.
* @param blockSize
* @param progress
* @param checksumOpt checksum parameter. If null, the values
* found in conf will be used.
* @throws IOException
* @see #setPermission(Path, FsPermission)
*/
public FSDataOutputStream create(Path f,
FsPermission permission,
EnumSet<CreateFlag> flags,
int bufferSize,
short replication,
long blockSize,
Progressable progress,
ChecksumOpt checksumOpt) throws IOException {
// Checksum options are ignored by default. The file systems that
// implement checksum need to override this method. The full
// support is currently only available in DFS.
return create(f, permission, flags.contains(CreateFlag.OVERWRITE),
bufferSize, replication, blockSize, progress);
}
/*.
* This create has been added to support the FileContext that processes
* the permission
@ -863,7 +939,7 @@ public FSDataOutputStream create(Path f,
protected FSDataOutputStream primitiveCreate(Path f,
FsPermission absolutePermission, EnumSet<CreateFlag> flag, int bufferSize,
short replication, long blockSize, Progressable progress,
int bytesPerChecksum) throws IOException {
ChecksumOpt checksumOpt) throws IOException {
boolean pathExists = exists(f);
CreateFlag.validate(f, pathExists, flag);
@ -1543,7 +1619,11 @@ private FileStatus[] globStatusInternal(Path pathPattern, PathFilter filter)
GlobFilter fp = new GlobFilter(components[components.length - 1], filter);
if (fp.hasPattern()) { // last component has a pattern
// list parent directories and then glob the results
results = listStatus(parentPaths, fp);
try {
results = listStatus(parentPaths, fp);
} catch (FileNotFoundException e) {
results = null;
}
hasGlob[0] = true;
} else { // last component does not have a pattern
// remove the quoting of metachars in a non-regexp expansion
@ -1592,7 +1672,11 @@ private Path[] globPathsLevel(Path[] parents, String[] filePattern,
}
GlobFilter fp = new GlobFilter(filePattern[level]);
if (fp.hasPattern()) {
parents = FileUtil.stat2Paths(listStatus(parents, fp));
try {
parents = FileUtil.stat2Paths(listStatus(parents, fp));
} catch (FileNotFoundException e) {
parents = null;
}
hasGlob[0] = true;
} else { // the component does not have a pattern
// remove the quoting of metachars in a non-regexp expansion

32
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/FilterFileSystem.java
View File

@ -22,15 +22,12 @@
import java.net.URI;
import java.net.URISyntaxException;
import java.util.EnumSet;
import java.util.List;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.fs.ContentSummary;
import org.apache.hadoop.security.Credentials;
import org.apache.hadoop.security.token.Token;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.util.Progressable;
/****************************************************************
@ -414,10 +411,11 @@ public void setPermission(Path p, FsPermission permission
@Override
protected FSDataOutputStream primitiveCreate(Path f,
FsPermission absolutePermission, EnumSet<CreateFlag> flag,
int bufferSize, short replication, long blockSize, Progressable progress, int bytesPerChecksum)
int bufferSize, short replication, long blockSize,
Progressable progress, ChecksumOpt checksumOpt)
throws IOException {
return fs.primitiveCreate(f, absolutePermission, flag,
bufferSize, replication, blockSize, progress, bytesPerChecksum);
bufferSize, replication, blockSize, progress, checksumOpt);
}
@Override
@ -428,25 +426,7 @@ protected boolean primitiveMkdir(Path f, FsPermission abdolutePermission)
}
@Override // FileSystem
public String getCanonicalServiceName() {
return fs.getCanonicalServiceName();
}
@Override // FileSystem
@SuppressWarnings("deprecation")
public Token<?> getDelegationToken(String renewer) throws IOException {
return fs.getDelegationToken(renewer);
}
@Override // FileSystem
public List<Token<?>> getDelegationTokens(String renewer) throws IOException {
return fs.getDelegationTokens(renewer);
}
@Override
// FileSystem
public List<Token<?>> getDelegationTokens(String renewer,
Credentials credentials) throws IOException {
return fs.getDelegationTokens(renewer, credentials);
public FileSystem[] getChildFileSystems() {
return new FileSystem[]{fs};
}
}

5
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/FilterFs.java
View File

@ -27,6 +27,7 @@
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.fs.FileSystem.Statistics;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.security.AccessControlException;
import org.apache.hadoop.security.token.Token;
import org.apache.hadoop.util.Progressable;
@ -81,11 +82,11 @@ public Path getHomeDirectory() {
public FSDataOutputStream createInternal(Path f,
EnumSet<CreateFlag> flag, FsPermission absolutePermission, int bufferSize,
short replication, long blockSize, Progressable progress,
int bytesPerChecksum, boolean createParent)
ChecksumOpt checksumOpt, boolean createParent)
throws IOException, UnresolvedLinkException {
checkPath(f);
return myFs.createInternal(f, flag, absolutePermission, bufferSize,
replication, blockSize, progress, bytesPerChecksum, createParent);
replication, blockSize, progress, checksumOpt, createParent);
}
@Override

19
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/FsServerDefaults.java
View File

@ -26,6 +26,8 @@
import org.apache.hadoop.io.Writable;
import org.apache.hadoop.io.WritableFactories;
import org.apache.hadoop.io.WritableFactory;
import org.apache.hadoop.io.WritableUtils;
import org.apache.hadoop.util.DataChecksum;
/****************************************************
* Provides server default configuration values to clients.
@ -49,19 +51,24 @@ public Writable newInstance() {
private short replication;
private int fileBufferSize;
private boolean encryptDataTransfer;
private long trashInterval;
private DataChecksum.Type checksumType;
public FsServerDefaults() {
}
public FsServerDefaults(long blockSize, int bytesPerChecksum,
int writePacketSize, short replication, int fileBufferSize,
boolean encryptDataTransfer) {
boolean encryptDataTransfer, long trashInterval,
DataChecksum.Type checksumType) {
this.blockSize = blockSize;
this.bytesPerChecksum = bytesPerChecksum;
this.writePacketSize = writePacketSize;
this.replication = replication;
this.fileBufferSize = fileBufferSize;
this.encryptDataTransfer = encryptDataTransfer;
this.trashInterval = trashInterval;
this.checksumType = checksumType;
}
public long getBlockSize() {
@ -88,6 +95,14 @@ public boolean getEncryptDataTransfer() {
return encryptDataTransfer;
}
public long getTrashInterval() {
return trashInterval;
}
public DataChecksum.Type getChecksumType() {
return checksumType;
}
// /////////////////////////////////////////
// Writable
// /////////////////////////////////////////
@ -98,6 +113,7 @@ public void write(DataOutput out) throws IOException {
out.writeInt(writePacketSize);
out.writeShort(replication);
out.writeInt(fileBufferSize);
WritableUtils.writeEnum(out, checksumType);
}
@InterfaceAudience.Private
@ -107,5 +123,6 @@ public void readFields(DataInput in) throws IOException {
writePacketSize = in.readInt();
replication = in.readShort();
fileBufferSize = in.readInt();
checksumType = WritableUtils.readEnum(in, DataChecksum.Type.class);
}
}

41
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/MD5MD5CRC32CastagnoliFileChecksum.java Normal file
View File

@ -0,0 +1,41 @@
/**
* 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.io.MD5Hash;
import org.apache.hadoop.util.DataChecksum;
/** For CRC32 with the Castagnoli polynomial */
public class MD5MD5CRC32CastagnoliFileChecksum extends MD5MD5CRC32FileChecksum {
/** Same as this(0, 0, null) */
public MD5MD5CRC32CastagnoliFileChecksum() {
this(0, 0, null);
}
/** Create a MD5FileChecksum */
public MD5MD5CRC32CastagnoliFileChecksum(int bytesPerCRC, long crcPerBlock, MD5Hash md5) {
super(bytesPerCRC, crcPerBlock, md5);
}
@Override
public DataChecksum.Type getCrcType() {
// default to the one that is understood by all releases.
return DataChecksum.Type.CRC32C;
}
}

64
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/MD5MD5CRC32FileChecksum.java
View File

@ -23,12 +23,17 @@
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.io.MD5Hash;
import org.apache.hadoop.io.WritableUtils;
import org.apache.hadoop.util.DataChecksum;
import org.xml.sax.Attributes;
import org.xml.sax.SAXException;
import org.znerd.xmlenc.XMLOutputter;
import org.apache.hadoop.fs.MD5MD5CRC32CastagnoliFileChecksum;
import org.apache.hadoop.fs.MD5MD5CRC32GzipFileChecksum;
/** MD5 of MD5 of CRC32. */
@InterfaceAudience.LimitedPrivate({"HDFS"})
@InterfaceStability.Unstable
@ -54,7 +59,19 @@ public MD5MD5CRC32FileChecksum(int bytesPerCRC, long crcPerBlock, MD5Hash md5) {
/** {@inheritDoc} */
public String getAlgorithmName() {
return "MD5-of-" + crcPerBlock + "MD5-of-" + bytesPerCRC + "CRC32";
return "MD5-of-" + crcPerBlock + "MD5-of-" + bytesPerCRC +
getCrcType().name();
}
public static DataChecksum.Type getCrcTypeFromAlgorithmName(String algorithm)
throws IOException {
if (algorithm.endsWith(DataChecksum.Type.CRC32.name())) {
return DataChecksum.Type.CRC32;
} else if (algorithm.endsWith(DataChecksum.Type.CRC32C.name())) {
return DataChecksum.Type.CRC32C;
}
throw new IOException("Unknown checksum type in " + algorithm);
}
/** {@inheritDoc} */
@ -65,6 +82,16 @@ public byte[] getBytes() {
return WritableUtils.toByteArray(this);
}
/** returns the CRC type */
public DataChecksum.Type getCrcType() {
// default to the one that is understood by all releases.
return DataChecksum.Type.CRC32;
}
public ChecksumOpt getChecksumOpt() {
return new ChecksumOpt(getCrcType(), bytesPerCRC);
}
/** {@inheritDoc} */
public void readFields(DataInput in) throws IOException {
bytesPerCRC = in.readInt();
@ -86,6 +113,7 @@ public static void write(XMLOutputter xml, MD5MD5CRC32FileChecksum that
if (that != null) {
xml.attribute("bytesPerCRC", "" + that.bytesPerCRC);
xml.attribute("crcPerBlock", "" + that.crcPerBlock);
xml.attribute("crcType", ""+ that.getCrcType().name());
xml.attribute("md5", "" + that.md5);
}
xml.endTag();
@ -97,16 +125,40 @@ public static MD5MD5CRC32FileChecksum valueOf(Attributes attrs
final String bytesPerCRC = attrs.getValue("bytesPerCRC");
final String crcPerBlock = attrs.getValue("crcPerBlock");
final String md5 = attrs.getValue("md5");
String crcType = attrs.getValue("crcType");
DataChecksum.Type finalCrcType;
if (bytesPerCRC == null || crcPerBlock == null || md5 == null) {
return null;
}
try {
return new MD5MD5CRC32FileChecksum(Integer.valueOf(bytesPerCRC),
Integer.valueOf(crcPerBlock), new MD5Hash(md5));
} catch(Exception e) {
// old versions don't support crcType.
if (crcType == null || crcType == "") {
finalCrcType = DataChecksum.Type.CRC32;
} else {
finalCrcType = DataChecksum.Type.valueOf(crcType);
}
switch (finalCrcType) {
case CRC32:
return new MD5MD5CRC32GzipFileChecksum(
Integer.valueOf(bytesPerCRC),
Integer.valueOf(crcPerBlock),
new MD5Hash(md5));
case CRC32C:
return new MD5MD5CRC32CastagnoliFileChecksum(
Integer.valueOf(bytesPerCRC),
Integer.valueOf(crcPerBlock),
new MD5Hash(md5));
default:
// we should never get here since finalCrcType will
// hold a valid type or we should have got an exception.
return null;
}
} catch (Exception e) {
throw new SAXException("Invalid attributes: bytesPerCRC=" + bytesPerCRC
+ ", crcPerBlock=" + crcPerBlock + ", md5=" + md5, e);
+ ", crcPerBlock=" + crcPerBlock + ", crcType=" + crcType
+ ", md5=" + md5, e);
}
}
@ -114,4 +166,4 @@ public static MD5MD5CRC32FileChecksum valueOf(Attributes attrs
public String toString() {
return getAlgorithmName() + ":" + md5;
}
}
}

40
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/MD5MD5CRC32GzipFileChecksum.java Normal file
View File

@ -0,0 +1,40 @@
/**
* 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.io.MD5Hash;
import org.apache.hadoop.util.DataChecksum;
/** For CRC32 with the Gzip polynomial */
public class MD5MD5CRC32GzipFileChecksum extends MD5MD5CRC32FileChecksum {
/** Same as this(0, 0, null) */
public MD5MD5CRC32GzipFileChecksum() {
this(0, 0, null);
}
/** Create a MD5FileChecksum */
public MD5MD5CRC32GzipFileChecksum(int bytesPerCRC, long crcPerBlock, MD5Hash md5) {
super(bytesPerCRC, crcPerBlock, md5);
}
@Override
public DataChecksum.Type getCrcType() {
// default to the one that is understood by all releases.
return DataChecksum.Type.CRC32;
}
}

129
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/Options.java
View File

@ -20,7 +20,9 @@
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.util.DataChecksum;
import org.apache.hadoop.util.Progressable;
import org.apache.hadoop.HadoopIllegalArgumentException;
/**
* This class contains options related to file system operations.
@ -46,6 +48,10 @@ public static ReplicationFactor repFac(short rf) {
public static BytesPerChecksum bytesPerChecksum(short crc) {
return new BytesPerChecksum(crc);
}
public static ChecksumParam checksumParam(
ChecksumOpt csumOpt) {
return new ChecksumParam(csumOpt);
}
public static Perms perms(FsPermission perm) {
return new Perms(perm);
}
@ -91,7 +97,8 @@ protected BufferSize(int bs) {
}
public int getValue() { return bufferSize; }
}
/** This is not needed if ChecksumParam is specified. **/
public static class BytesPerChecksum extends CreateOpts {
private final int bytesPerChecksum;
protected BytesPerChecksum(short bpc) {
@ -103,6 +110,14 @@ protected BytesPerChecksum(short bpc) {
}
public int getValue() { return bytesPerChecksum; }
}
public static class ChecksumParam extends CreateOpts {
private final ChecksumOpt checksumOpt;
protected ChecksumParam(ChecksumOpt csumOpt) {
checksumOpt = csumOpt;
}
public ChecksumOpt getValue() { return checksumOpt; }
}
public static class Perms extends CreateOpts {
private final FsPermission permissions;
@ -206,4 +221,116 @@ public byte value() {
return code;
}
}
/**
* This is used in FileSystem and FileContext to specify checksum options.
*/
public static class ChecksumOpt {
private final int crcBlockSize;
private final DataChecksum.Type crcType;
/**
* Create a uninitialized one
*/
public ChecksumOpt() {
crcBlockSize = -1;
crcType = DataChecksum.Type.DEFAULT;
}
/**
* Normal ctor
* @param type checksum type
* @param size bytes per checksum
*/
public ChecksumOpt(DataChecksum.Type type, int size) {
crcBlockSize = size;
crcType = type;
}
public int getBytesPerChecksum() {
return crcBlockSize;
}
public DataChecksum.Type getChecksumType() {
return crcType;
}
/**
* Create a ChecksumOpts that disables checksum
*/
public static ChecksumOpt createDisabled() {
return new ChecksumOpt(DataChecksum.Type.NULL, -1);
}
/**
* A helper method for processing user input and default value to
* create a combined checksum option. This is a bit complicated because
* bytesPerChecksum is kept for backward compatibility.
*
* @param defaultOpt Default checksum option
* @param userOpt User-specified checksum option. Ignored if null.
* @param userBytesPerChecksum User-specified bytesPerChecksum
* Ignored if < 0.
*/
public static ChecksumOpt processChecksumOpt(ChecksumOpt defaultOpt,
ChecksumOpt userOpt, int userBytesPerChecksum) {
// The following is done to avoid unnecessary creation of new objects.
// tri-state variable: 0 default, 1 userBytesPerChecksum, 2 userOpt
short whichSize;
// true default, false userOpt
boolean useDefaultType;
// bytesPerChecksum - order of preference
// user specified value in bytesPerChecksum
// user specified value in checksumOpt
// default.
if (userBytesPerChecksum > 0) {
whichSize = 1; // userBytesPerChecksum
} else if (userOpt != null && userOpt.getBytesPerChecksum() > 0) {
whichSize = 2; // userOpt
} else {
whichSize = 0; // default
}
// checksum type - order of preference
// user specified value in checksumOpt
// default.
if (userOpt != null &&
userOpt.getChecksumType() != DataChecksum.Type.DEFAULT) {
useDefaultType = false;
} else {
useDefaultType = true;
}
// Short out the common and easy cases
if (whichSize == 0 && useDefaultType) {
return defaultOpt;
} else if (whichSize == 2 && !useDefaultType) {
return userOpt;
}
// Take care of the rest of combinations
DataChecksum.Type type = useDefaultType ? defaultOpt.getChecksumType() :
userOpt.getChecksumType();
if (whichSize == 0) {
return new ChecksumOpt(type, defaultOpt.getBytesPerChecksum());
} else if (whichSize == 1) {
return new ChecksumOpt(type, userBytesPerChecksum);
} else {
return new ChecksumOpt(type, userOpt.getBytesPerChecksum());
}
}
/**
* A helper method for processing user input and default value to
* create a combined checksum option.
*
* @param defaultOpt Default checksum option
* @param userOpt User-specified checksum option
*/
public static ChecksumOpt processChecksumOpt(ChecksumOpt defaultOpt,
ChecksumOpt userOpt) {
return processChecksumOpt(defaultOpt, userOpt, -1);
}
}
}

5
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/Trash.java
View File

@ -117,9 +117,4 @@ TrashPolicy getTrashPolicy() {
public Runnable getEmptier() throws IOException {
return trashPolicy.getEmptier();
}
/** Run an emptier.*/
public static void main(String[] args) throws Exception {
new Trash(new Configuration()).getEmptier().run();
}
}

45
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/TrashPolicyDefault.java
View File

@ -34,7 +34,6 @@
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.conf.Configured;
import org.apache.hadoop.fs.Options.Rename;
import org.apache.hadoop.fs.permission.FsAction;
import org.apache.hadoop.fs.permission.FsPermission;
@ -66,6 +65,7 @@ public class TrashPolicyDefault extends TrashPolicy {
private Path current;
private Path homesParent;
private long emptierInterval;
public TrashPolicyDefault() { }
@ -79,8 +79,27 @@ public void initialize(Configuration conf, FileSystem fs, Path home) {
this.trash = new Path(home, TRASH);
this.homesParent = home.getParent();
this.current = new Path(trash, CURRENT);
this.deletionInterval = (long) (conf.getFloat(FS_TRASH_INTERVAL_KEY,
FS_TRASH_INTERVAL_DEFAULT) * MSECS_PER_MINUTE);
long trashInterval = 0;
try {
trashInterval = fs.getServerDefaults(home).getTrashInterval();
} catch (IOException ioe) {
LOG.warn("Unable to get server defaults", ioe);
}
// If the trash interval is not configured or is disabled on the
// server side then check the config which may be client side.
if (0 == trashInterval) {
this.deletionInterval = (long)(conf.getFloat(
FS_TRASH_INTERVAL_KEY, FS_TRASH_INTERVAL_DEFAULT)
* MSECS_PER_MINUTE);
} else {
this.deletionInterval = trashInterval * MSECS_PER_MINUTE;
}
// For the checkpoint interval use the given config instead of
// checking the server as it's OK if a client starts an emptier
// with a different interval than the server.
this.emptierInterval = (long)(conf.getFloat(
FS_TRASH_CHECKPOINT_INTERVAL_KEY, FS_TRASH_CHECKPOINT_INTERVAL_DEFAULT)
* MSECS_PER_MINUTE);
}
private Path makeTrashRelativePath(Path basePath, Path rmFilePath) {
@ -89,7 +108,7 @@ private Path makeTrashRelativePath(Path basePath, Path rmFilePath) {
@Override
public boolean isEnabled() {
return (deletionInterval != 0);
return deletionInterval != 0;
}
@Override
@ -223,7 +242,7 @@ public Path getCurrentTrashDir() {
@Override
public Runnable getEmptier() throws IOException {
return new Emptier(getConf());
return new Emptier(getConf(), emptierInterval);
}
private class Emptier implements Runnable {
@ -231,16 +250,14 @@ private class Emptier implements Runnable {
private Configuration conf;
private long emptierInterval;
Emptier(Configuration conf) throws IOException {
Emptier(Configuration conf, long emptierInterval) throws IOException {
this.conf = conf;
this.emptierInterval = (long) (conf.getFloat(FS_TRASH_CHECKPOINT_INTERVAL_KEY,
FS_TRASH_CHECKPOINT_INTERVAL_DEFAULT) *
MSECS_PER_MINUTE);
if (this.emptierInterval > deletionInterval ||
this.emptierInterval == 0) {
LOG.warn("The configured interval for checkpoint is " +
this.emptierInterval + " minutes." +
" Using interval of " + deletionInterval +
this.emptierInterval = emptierInterval;
if (emptierInterval > deletionInterval || emptierInterval == 0) {
LOG.info("The configured checkpoint interval is " +
(emptierInterval / MSECS_PER_MINUTE) + " minutes." +
" Using an interval of " +
(deletionInterval / MSECS_PER_MINUTE) +
" minutes that is used for deletion instead");
this.emptierInterval = deletionInterval;
}

13
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/ftp/FtpConfigKeys.java
View File

@ -23,10 +23,16 @@
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.fs.CommonConfigurationKeys;
import org.apache.hadoop.fs.FsServerDefaults;
import org.apache.hadoop.util.DataChecksum;
/**
* This class contains constants for configuration keys used
* in the ftp file system.
*
* Note that the settings for unimplemented features are ignored.
* E.g. checksum related settings are just place holders. Even when
* wrapped with {@link ChecksumFileSystem}, these settings are not
* used.
*/
@InterfaceAudience.Private
@InterfaceStability.Unstable
@ -45,6 +51,9 @@ public class FtpConfigKeys extends CommonConfigurationKeys {
"ftp.client-write-packet-size";
public static final int CLIENT_WRITE_PACKET_SIZE_DEFAULT = 64*1024;
public static final boolean ENCRYPT_DATA_TRANSFER_DEFAULT = false;
public static final long FS_TRASH_INTERVAL_DEFAULT = 0;
public static final DataChecksum.Type CHECKSUM_TYPE_DEFAULT =
DataChecksum.Type.CRC32;
protected static FsServerDefaults getServerDefaults() throws IOException {
return new FsServerDefaults(
@ -53,7 +62,9 @@ protected static FsServerDefaults getServerDefaults() throws IOException {
CLIENT_WRITE_PACKET_SIZE_DEFAULT,
REPLICATION_DEFAULT,
STREAM_BUFFER_SIZE_DEFAULT,
ENCRYPT_DATA_TRANSFER_DEFAULT);
ENCRYPT_DATA_TRANSFER_DEFAULT,
FS_TRASH_INTERVAL_DEFAULT,
CHECKSUM_TYPE_DEFAULT);
}
}

15
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/local/LocalConfigKeys.java
View File

@ -24,11 +24,18 @@
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.fs.CommonConfigurationKeys;
import org.apache.hadoop.fs.FsServerDefaults;
import org.apache.hadoop.util.DataChecksum;
/**
* This class contains constants for configuration keys used
* in the local file system, raw local fs and checksum fs.
*
* Note that the settings for unimplemented features are ignored.
* E.g. checksum related settings are just place holders. Even when
* wrapped with {@link ChecksumFileSystem}, these settings are not
* used.
*/
@InterfaceAudience.Private
@InterfaceStability.Unstable
public class LocalConfigKeys extends CommonConfigurationKeys {
@ -44,7 +51,9 @@ public class LocalConfigKeys extends CommonConfigurationKeys {
"file.client-write-packet-size";
public static final int CLIENT_WRITE_PACKET_SIZE_DEFAULT = 64*1024;
public static final boolean ENCRYPT_DATA_TRANSFER_DEFAULT = false;
public static final long FS_TRASH_INTERVAL_DEFAULT = 0;
public static final DataChecksum.Type CHECKSUM_TYPE_DEFAULT =
DataChecksum.Type.CRC32;
public static FsServerDefaults getServerDefaults() throws IOException {
return new FsServerDefaults(
BLOCK_SIZE_DEFAULT,
@ -52,7 +61,9 @@ public static FsServerDefaults getServerDefaults() throws IOException {
CLIENT_WRITE_PACKET_SIZE_DEFAULT,
REPLICATION_DEFAULT,
STREAM_BUFFER_SIZE_DEFAULT,
ENCRYPT_DATA_TRANSFER_DEFAULT);
ENCRYPT_DATA_TRANSFER_DEFAULT,
FS_TRASH_INTERVAL_DEFAULT,
CHECKSUM_TYPE_DEFAULT);
}
}

18
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/shell/Command.java
View File

@ -44,12 +44,12 @@
@InterfaceStability.Evolving
abstract public class Command extends Configured {
/** default name of the command */
public static String NAME;
/** the command's usage switches and arguments format */
public static String USAGE;
/** the command's long description */
public static String DESCRIPTION;
/** field name indicating the default name of the command */
public static final String COMMAND_NAME_FIELD = "NAME";
/** field name indicating the command's usage switches and arguments format */
public static final String COMMAND_USAGE_FIELD = "USAGE";
/** field name indicating the command's long description */
public static final String COMMAND_DESCRIPTION_FIELD = "DESCRIPTION";
protected String[] args;
protected String name;
@ -397,7 +397,7 @@ public void displayWarning(String message) {
*/
public String getName() {
return (name == null)
? getCommandField("NAME")
? getCommandField(COMMAND_NAME_FIELD)
: name.startsWith("-") ? name.substring(1) : name;
}
@ -415,7 +415,7 @@ public void setName(String name) {
*/
public String getUsage() {
String cmd = "-" + getName();
String usage = isDeprecated() ? "" : getCommandField("USAGE");
String usage = isDeprecated() ? "" : getCommandField(COMMAND_USAGE_FIELD);
return usage.isEmpty() ? cmd : cmd + " " + usage;
}
@ -426,7 +426,7 @@ public String getUsage() {
public String getDescription() {
return isDeprecated()
? "(DEPRECATED) Same as '" + getReplacementCommand() + "'"
: getCommandField("DESCRIPTION");
: getCommandField(COMMAND_DESCRIPTION_FIELD);
}
/**

5
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/viewfs/ChRootedFs.java
View File

@ -34,6 +34,7 @@
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FsServerDefaults;
import org.apache.hadoop.fs.FsStatus;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.UnresolvedLinkException;
import org.apache.hadoop.fs.permission.FsPermission;
@ -159,11 +160,11 @@ public Path getResolvedQualifiedPath(final Path f)
public FSDataOutputStream createInternal(final Path f,
final EnumSet<CreateFlag> flag, final FsPermission absolutePermission,
final int bufferSize, final short replication, final long blockSize,
final Progressable progress, final int bytesPerChecksum,
final Progressable progress, final ChecksumOpt checksumOpt,
final boolean createParent) throws IOException, UnresolvedLinkException {
return myFs.createInternal(fullPath(f), flag,
absolutePermission, bufferSize,
replication, blockSize, progress, bytesPerChecksum, createParent);
replication, blockSize, progress, checksumOpt, createParent);
}
@Override

75
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/viewfs/ViewFileSystem.java
View File

@ -23,7 +23,7 @@
import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.HashSet;
import java.util.List;
import java.util.Set;
@ -49,11 +49,8 @@
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.fs.viewfs.InodeTree.INode;
import org.apache.hadoop.fs.viewfs.InodeTree.INodeLink;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.security.AccessControlException;
import org.apache.hadoop.security.Credentials;
import org.apache.hadoop.security.UserGroupInformation;
import org.apache.hadoop.security.token.Token;
import org.apache.hadoop.util.Progressable;
import org.apache.hadoop.util.Time;
@ -235,11 +232,6 @@ public Path getTrashCanLocation(final Path f) throws FileNotFoundException {
return res.isInternalDir() ? null : res.targetFileSystem.getHomeDirectory();
}
@Override
public String getCanonicalServiceName() {
return null;
}
@Override
public URI getUri() {
return myUri;
@ -549,6 +541,18 @@ public void setWriteChecksum(final boolean writeChecksum) {
}
}
@Override
public FileSystem[] getChildFileSystems() {
List<InodeTree.MountPoint<FileSystem>> mountPoints =
fsState.getMountPoints();
Set<FileSystem> children = new HashSet<FileSystem>();
for (InodeTree.MountPoint<FileSystem> mountPoint : mountPoints) {
FileSystem targetFs = mountPoint.target.targetFileSystem;
children.addAll(Arrays.asList(targetFs.getChildFileSystems()));
}
return children.toArray(new FileSystem[]{});
}
public MountPoint[] getMountPoints() {
List<InodeTree.MountPoint<FileSystem>> mountPoints =
fsState.getMountPoints();
@ -561,59 +565,6 @@ public MountPoint[] getMountPoints() {
return result;
}
@Override
public List<Token<?>> getDelegationTokens(String renewer) throws IOException {
List<InodeTree.MountPoint<FileSystem>> mountPoints =
fsState.getMountPoints();
int initialListSize = 0;
for (InodeTree.MountPoint<FileSystem> im : mountPoints) {
initialListSize += im.target.targetDirLinkList.length;
}
List<Token<?>> result = new ArrayList<Token<?>>(initialListSize);
for ( int i = 0; i < mountPoints.size(); ++i ) {
List<Token<?>> tokens =
mountPoints.get(i).target.targetFileSystem.getDelegationTokens(renewer);
if (tokens != null) {
result.addAll(tokens);
}
}
return result;
}
@Override
public List<Token<?>> getDelegationTokens(String renewer,
Credentials credentials) throws IOException {
List<InodeTree.MountPoint<FileSystem>> mountPoints =
fsState.getMountPoints();
int initialListSize = 0;
for (InodeTree.MountPoint<FileSystem> im : mountPoints) {
initialListSize += im.target.targetDirLinkList.length;
}
Set<String> seenServiceNames = new HashSet<String>();
List<Token<?>> result = new ArrayList<Token<?>>(initialListSize);
for (int i = 0; i < mountPoints.size(); ++i) {
String serviceName =
mountPoints.get(i).target.targetFileSystem.getCanonicalServiceName();
if (serviceName == null || seenServiceNames.contains(serviceName)) {
continue;
}
seenServiceNames.add(serviceName);
Token<?> knownToken = credentials.getToken(new Text(serviceName));
if (knownToken != null) {
result.add(knownToken);
} else {
List<Token<?>> tokens =
mountPoints.get(i).target.targetFileSystem
.getDelegationTokens(renewer);
if (tokens != null) {
result.addAll(tokens);
}
}
}
return result;
}
/*
* An instance of this class represents an internal dir of the viewFs
* that is internal dir of the mount table.

7
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/viewfs/ViewFs.java
View File

@ -42,6 +42,7 @@
import org.apache.hadoop.fs.FsConstants;
import org.apache.hadoop.fs.FsServerDefaults;
import org.apache.hadoop.fs.FsStatus;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.ParentNotDirectoryException;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.RemoteIterator;
@ -265,7 +266,7 @@ public Path resolvePath(final Path f) throws FileNotFoundException,
public FSDataOutputStream createInternal(final Path f,
final EnumSet<CreateFlag> flag, final FsPermission absolutePermission,
final int bufferSize, final short replication, final long blockSize,
final Progressable progress, final int bytesPerChecksum,
final Progressable progress, final ChecksumOpt checksumOpt,
final boolean createParent) throws AccessControlException,
FileAlreadyExistsException, FileNotFoundException,
ParentNotDirectoryException, UnsupportedFileSystemException,
@ -283,7 +284,7 @@ public FSDataOutputStream createInternal(final Path f,
assert(res.remainingPath != null);
return res.targetFileSystem.createInternal(res.remainingPath, flag,
absolutePermission, bufferSize, replication,
blockSize, progress, bytesPerChecksum,
blockSize, progress, checksumOpt,
createParent);
}
@ -632,7 +633,7 @@ static private void checkPathIsSlash(final Path f) throws IOException {
public FSDataOutputStream createInternal(final Path f,
final EnumSet<CreateFlag> flag, final FsPermission absolutePermission,
final int bufferSize, final short replication, final long blockSize,
final Progressable progress, final int bytesPerChecksum,
final Progressable progress, final ChecksumOpt checksumOpt,
final boolean createParent) throws AccessControlException,
FileAlreadyExistsException, FileNotFoundException,
ParentNotDirectoryException, UnsupportedFileSystemException,

20
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/ha/FailoverController.java
View File

@ -49,16 +49,34 @@ public class FailoverController {
private final int rpcTimeoutToNewActive;
private final Configuration conf;
/*
* Need a copy of conf for graceful fence to set
* configurable retries for IPC client.
* Refer HDFS-3561
*/
private final Configuration gracefulFenceConf;
private final RequestSource requestSource;
public FailoverController(Configuration conf,
RequestSource source) {
this.conf = conf;
this.gracefulFenceConf = new Configuration(conf);
this.requestSource = source;
this.gracefulFenceTimeout = getGracefulFenceTimeout(conf);
this.rpcTimeoutToNewActive = getRpcTimeoutToNewActive(conf);
//Configure less retries for graceful fence
int gracefulFenceConnectRetries = conf.getInt(
CommonConfigurationKeys.HA_FC_GRACEFUL_FENCE_CONNECTION_RETRIES,
CommonConfigurationKeys.HA_FC_GRACEFUL_FENCE_CONNECTION_RETRIES_DEFAULT);
gracefulFenceConf.setInt(
CommonConfigurationKeys.IPC_CLIENT_CONNECT_MAX_RETRIES_KEY,
gracefulFenceConnectRetries);
gracefulFenceConf.setInt(
CommonConfigurationKeys.IPC_CLIENT_CONNECT_MAX_RETRIES_ON_SOCKET_TIMEOUTS_KEY,
gracefulFenceConnectRetries);
}
static int getGracefulFenceTimeout(Configuration conf) {
@ -150,7 +168,7 @@ private StateChangeRequestInfo createReqInfo() {
boolean tryGracefulFence(HAServiceTarget svc) {
HAServiceProtocol proxy = null;
try {
proxy = svc.getProxy(conf, gracefulFenceTimeout);
proxy = svc.getProxy(gracefulFenceConf, gracefulFenceTimeout);
proxy.transitionToStandby(createReqInfo());
return true;
} catch (ServiceFailedException sfe) {

6
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/http/HttpServer.java
View File

@ -330,6 +330,12 @@ protected void addDefaultApps(ContextHandlerCollection parent,
Context logContext = new Context(parent, "/logs");
logContext.setResourceBase(logDir);
logContext.addServlet(AdminAuthorizedServlet.class, "/*");
if (conf.getBoolean(
CommonConfigurationKeys.HADOOP_JETTY_LOGS_SERVE_ALIASES,
CommonConfigurationKeys.DEFAULT_HADOOP_JETTY_LOGS_SERVE_ALIASES)) {
logContext.getInitParams().put(
"org.mortbay.jetty.servlet.Default.aliases", "true");
}
logContext.setDisplayName("logs");
setContextAttributes(logContext, conf);
defaultContexts.put(logContext, true);

18
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/IOUtils.java
View File

@ -206,12 +206,20 @@ public static void readFully(InputStream in, byte buf[],
* for any reason (including EOF)
*/
public static void skipFully(InputStream in, long len) throws IOException {
while (len > 0) {
long ret = in.skip(len);
if (ret < 0) {
throw new IOException( "Premature EOF from inputStream");
long amt = len;
while (amt > 0) {
long ret = in.skip(amt);
if (ret == 0) {
// skip may return 0 even if we're not at EOF. Luckily, we can
// use the read() method to figure out if we're at the end.
int b = in.read();
if (b == -1) {
throw new EOFException( "Premature EOF from inputStream after " +
"skipping " + (len - amt) + " byte(s).");
}
ret = 1;
}
len -= ret;
amt -= ret;
}
}

44
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/ipc/Server.java
View File

@ -46,11 +46,13 @@
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.HashSet;
import java.util.Iterator;
import java.util.LinkedList;
import java.util.List;
import java.util.Map;
import java.util.Random;
import java.util.Set;
import java.util.concurrent.BlockingQueue;
import java.util.concurrent.ConcurrentHashMap;
import java.util.concurrent.LinkedBlockingQueue;
@ -108,6 +110,42 @@
public abstract class Server {
private final boolean authorize;
private boolean isSecurityEnabled;
private ExceptionsHandler exceptionsHandler = new ExceptionsHandler();
public void addTerseExceptions(Class<?>... exceptionClass) {
exceptionsHandler.addTerseExceptions(exceptionClass);
}
/**
* ExceptionsHandler manages Exception groups for special handling
* e.g., terse exception group for concise logging messages
*/
static class ExceptionsHandler {
private volatile Set<String> terseExceptions = new HashSet<String>();
/**
* Add exception class so server won't log its stack trace.
* Modifying the terseException through this method is thread safe.
*
* @param exceptionClass exception classes
*/
void addTerseExceptions(Class<?>... exceptionClass) {
// Make a copy of terseException for performing modification
final HashSet<String> newSet = new HashSet<String>(terseExceptions);
// Add all class names into the HashSet
for (Class<?> name : exceptionClass) {
newSet.add(name.toString());
}
// Replace terseException set
terseExceptions = Collections.unmodifiableSet(newSet);
}
boolean isTerse(Class<?> t) {
return terseExceptions.contains(t.toString());
}
}
/**
* The first four bytes of Hadoop RPC connections
@ -1704,8 +1742,8 @@ public Writable run() throws Exception {
// on the server side, as opposed to just a normal exceptional
// result.
LOG.warn(logMsg, e);
} else if (e instanceof StandbyException) {
// Don't log the whole stack trace of these exceptions.
} else if (exceptionsHandler.isTerse(e.getClass())) {
// Don't log the whole stack trace of these exceptions.
// Way too noisy!
LOG.info(logMsg);
} else {
@ -1844,6 +1882,8 @@ protected Server(String bindAddress, int port,
if (isSecurityEnabled) {
SaslRpcServer.init(conf);
}
this.exceptionsHandler.addTerseExceptions(StandbyException.class);
}
private void closeConnection(Connection connection) {

20
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/security/Credentials.java
View File

@ -56,6 +56,20 @@ public class Credentials implements Writable {
private Map<Text, Token<? extends TokenIdentifier>> tokenMap =
new HashMap<Text, Token<? extends TokenIdentifier>>();
/**
* Create an empty credentials instance
*/
public Credentials() {
}
/**
* Create a copy of the given credentials
* @param credentials to copy
*/
public Credentials(Credentials credentials) {
this.addAll(credentials);
}
/**
* Returns the key bytes for the alias
* @param alias the alias for the key
@ -260,4 +274,10 @@ private void addAll(Credentials other, boolean overwrite) {
}
}
}
public void addTokensToUGI(UserGroupInformation ugi) {
for (Map.Entry<Text, Token<?>> token: tokenMap.entrySet()) {
ugi.addToken(token.getKey(), token.getValue());
}
}
}

63
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/security/JniBasedUnixGroupsMappingWithFallback.java Normal file
View File

@ -0,0 +1,63 @@
/**
* 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.security;
import java.io.IOException;
import java.util.List;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.util.NativeCodeLoader;
public class JniBasedUnixGroupsMappingWithFallback implements
GroupMappingServiceProvider {
private static final Log LOG = LogFactory
.getLog(JniBasedUnixGroupsMappingWithFallback.class);
private GroupMappingServiceProvider impl;
public JniBasedUnixGroupsMappingWithFallback() {
if (NativeCodeLoader.isNativeCodeLoaded()) {
this.impl = new JniBasedUnixGroupsMapping();
} else {
LOG.info("Falling back to shell based");
this.impl = new ShellBasedUnixGroupsMapping();
}
if (LOG.isDebugEnabled()){
LOG.debug("Group mapping impl=" + impl.getClass().getName());
}
}
@Override
public List<String> getGroups(String user) throws IOException {
return impl.getGroups(user);
}
@Override
public void cacheGroupsRefresh() throws IOException {
impl.cacheGroupsRefresh();
}
@Override
public void cacheGroupsAdd(List<String> groups) throws IOException {
impl.cacheGroupsAdd(groups);
}
}

63
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/security/JniBasedUnixGroupsNetgroupMappingWithFallback.java Normal file
View File

@ -0,0 +1,63 @@
/**
* 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.security;
import java.io.IOException;
import java.util.List;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.util.NativeCodeLoader;
public class JniBasedUnixGroupsNetgroupMappingWithFallback implements
GroupMappingServiceProvider {
private static final Log LOG = LogFactory
.getLog(JniBasedUnixGroupsNetgroupMappingWithFallback.class);
private GroupMappingServiceProvider impl;
public JniBasedUnixGroupsNetgroupMappingWithFallback() {
if (NativeCodeLoader.isNativeCodeLoaded()) {
this.impl = new JniBasedUnixGroupsNetgroupMapping();
} else {
LOG.info("Falling back to shell based");
this.impl = new ShellBasedUnixGroupsNetgroupMapping();
}
if (LOG.isDebugEnabled()) {
LOG.debug("Group mapping impl=" + impl.getClass().getName());
}
}
@Override
public List<String> getGroups(String user) throws IOException {
return impl.getGroups(user);
}
@Override
public void cacheGroupsRefresh() throws IOException {
impl.cacheGroupsRefresh();
}
@Override
public void cacheGroupsAdd(List<String> groups) throws IOException {
impl.cacheGroupsAdd(groups);
}
}

82
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/security/UserGroupInformation.java
View File

@ -55,6 +55,7 @@
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.metrics2.annotation.Metric;
import org.apache.hadoop.metrics2.annotation.Metrics;
import org.apache.hadoop.metrics2.lib.DefaultMetricsSystem;
@ -641,14 +642,12 @@ static UserGroupInformation getLoginUser() throws IOException {
AuthenticationMethod.SIMPLE);
loginUser = new UserGroupInformation(login.getSubject());
String fileLocation = System.getenv(HADOOP_TOKEN_FILE_LOCATION);
if (fileLocation != null && isSecurityEnabled()) {
if (fileLocation != null) {
// load the token storage file and put all of the tokens into the
// user.
Credentials cred = Credentials.readTokenStorageFile(
new Path("file:///" + fileLocation), conf);
for (Token<?> token: cred.getAllTokens()) {
loginUser.addToken(token);
}
cred.addTokensToUGI(loginUser);
}
loginUser.spawnAutoRenewalThreadForUserCreds();
} catch (LoginException le) {
@ -1177,6 +1176,41 @@ public synchronized boolean addTokenIdentifier(TokenIdentifier tokenId) {
public synchronized Set<TokenIdentifier> getTokenIdentifiers() {
return subject.getPublicCredentials(TokenIdentifier.class);
}
// wrapper to retain the creds key for the token
private class NamedToken {
Text alias;
Token<? extends TokenIdentifier> token;
NamedToken(Text alias, Token<? extends TokenIdentifier> token) {
this.alias = alias;
this.token = token;
}
@Override
public boolean equals(Object o) {
boolean equals;
if (o == this) {
equals = true;
} else if (!(o instanceof NamedToken)) {
equals = false;
} else {
Text otherAlias = ((NamedToken)o).alias;
if (alias == otherAlias) {
equals = true;
} else {
equals = (otherAlias != null && otherAlias.equals(alias));
}
}
return equals;
}
@Override
public int hashCode() {
return (alias != null) ? alias.hashCode() : -1;
}
@Override
public String toString() {
return "NamedToken: alias="+alias+" token="+token;
}
}
/**
* Add a token to this UGI
@ -1185,7 +1219,22 @@ public synchronized Set<TokenIdentifier> getTokenIdentifiers() {
* @return true on successful add of new token
*/
public synchronized boolean addToken(Token<? extends TokenIdentifier> token) {
return subject.getPrivateCredentials().add(token);
return addToken(token.getService(), token);
}
/**
* Add a named token to this UGI
*
* @param alias Name of the token
* @param token Token to be added
* @return true on successful add of new token
*/
public synchronized boolean addToken(Text alias,
Token<? extends TokenIdentifier> token) {
NamedToken namedToken = new NamedToken(alias, token);
Collection<Object> ugiCreds = subject.getPrivateCredentials();
ugiCreds.remove(namedToken); // allow token to be replaced
return ugiCreds.add(new NamedToken(alias, token));
}
/**
@ -1195,14 +1244,23 @@ public synchronized boolean addToken(Token<? extends TokenIdentifier> token) {
*/
public synchronized
Collection<Token<? extends TokenIdentifier>> getTokens() {
Set<Object> creds = subject.getPrivateCredentials();
List<Token<?>> result = new ArrayList<Token<?>>(creds.size());
for(Object o: creds) {
if (o instanceof Token<?>) {
result.add((Token<?>) o);
}
return Collections.unmodifiableList(
new ArrayList<Token<?>>(getCredentials().getAllTokens()));
}
/**
* Obtain the tokens in credentials form associated with this user.
*
* @return Credentials of tokens associated with this user
*/
public synchronized Credentials getCredentials() {
final Credentials credentials = new Credentials();
final Set<NamedToken> namedTokens =
subject.getPrivateCredentials(NamedToken.class);
for (final NamedToken namedToken : namedTokens) {
credentials.addToken(namedToken.alias, namedToken.token);
}
return Collections.unmodifiableList(result);
return credentials;
}
/**

12
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/security/token/delegation/AbstractDelegationTokenSecretManager.java
View File

@ -113,6 +113,16 @@ public void startThreads() throws IOException {
}
}
/**
* Reset all data structures and mutable state.
*/
public synchronized void reset() {
currentId = 0;
allKeys.clear();
delegationTokenSequenceNumber = 0;
currentTokens.clear();
}
/**
* Add a previously used master key to cache (when NN restarts),
* should be called before activate().
@ -190,7 +200,6 @@ private synchronized void removeExpiredKeys() {
@Override
protected synchronized byte[] createPassword(TokenIdent identifier) {
LOG.info("Creating password for identifier: "+identifier);
int sequenceNum;
long now = Time.now();
sequenceNum = ++delegationTokenSequenceNumber;
@ -198,6 +207,7 @@ protected synchronized byte[] createPassword(TokenIdent identifier) {
identifier.setMaxDate(now + tokenMaxLifetime);
identifier.setMasterKeyId(currentId);
identifier.setSequenceNumber(sequenceNum);
LOG.info("Creating password for identifier: " + identifier);
byte[] password = createPassword(identifier.getBytes(), currentKey.getKey());
currentTokens.put(identifier, new DelegationTokenInformation(now
+ tokenRenewInterval, password));

112
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/util/DataChecksum.java
View File

@ -43,31 +43,48 @@ public class DataChecksum implements Checksum {
public static final int CHECKSUM_NULL = 0;
public static final int CHECKSUM_CRC32 = 1;
public static final int CHECKSUM_CRC32C = 2;
private static String[] NAMES = new String[] {
"NULL", "CRC32", "CRC32C"
};
private static final int CHECKSUM_NULL_SIZE = 0;
private static final int CHECKSUM_CRC32_SIZE = 4;
private static final int CHECKSUM_CRC32C_SIZE = 4;
public static DataChecksum newDataChecksum( int type, int bytesPerChecksum ) {
public static final int CHECKSUM_DEFAULT = 3;
public static final int CHECKSUM_MIXED = 4;
/** The checksum types */
public static enum Type {
NULL (CHECKSUM_NULL, 0),
CRC32 (CHECKSUM_CRC32, 4),
CRC32C(CHECKSUM_CRC32C, 4),
DEFAULT(CHECKSUM_DEFAULT, 0), // This cannot be used to create DataChecksum
MIXED (CHECKSUM_MIXED, 0); // This cannot be used to create DataChecksum
public final int id;
public final int size;
private Type(int id, int size) {
this.id = id;
this.size = size;
}
/** @return the type corresponding to the id. */
public static Type valueOf(int id) {
if (id < 0 || id >= values().length) {
throw new IllegalArgumentException("id=" + id
+ " out of range [0, " + values().length + ")");
}
return values()[id];
}
}
public static DataChecksum newDataChecksum(Type type, int bytesPerChecksum ) {
if ( bytesPerChecksum <= 0 ) {
return null;
}
switch ( type ) {
case CHECKSUM_NULL :
return new DataChecksum( CHECKSUM_NULL, new ChecksumNull(),
CHECKSUM_NULL_SIZE, bytesPerChecksum );
case CHECKSUM_CRC32 :
return new DataChecksum( CHECKSUM_CRC32, new PureJavaCrc32(),
CHECKSUM_CRC32_SIZE, bytesPerChecksum );
case CHECKSUM_CRC32C:
return new DataChecksum( CHECKSUM_CRC32C, new PureJavaCrc32C(),
CHECKSUM_CRC32C_SIZE, bytesPerChecksum);
case NULL :
return new DataChecksum(type, new ChecksumNull(), bytesPerChecksum );
case CRC32 :
return new DataChecksum(type, new PureJavaCrc32(), bytesPerChecksum );
case CRC32C:
return new DataChecksum(type, new PureJavaCrc32C(), bytesPerChecksum);
default:
return null;
}
@ -87,7 +104,7 @@ public static DataChecksum newDataChecksum( byte bytes[], int offset ) {
( (bytes[offset+2] & 0xff) << 16 ) |
( (bytes[offset+3] & 0xff) << 8 ) |
( (bytes[offset+4] & 0xff) );
return newDataChecksum( bytes[0], bytesPerChecksum );
return newDataChecksum( Type.valueOf(bytes[0]), bytesPerChecksum );
}
/**
@ -98,7 +115,7 @@ public static DataChecksum newDataChecksum( DataInputStream in )
throws IOException {
int type = in.readByte();
int bpc = in.readInt();
DataChecksum summer = newDataChecksum( type, bpc );
DataChecksum summer = newDataChecksum(Type.valueOf(type), bpc );
if ( summer == null ) {
throw new IOException( "Could not create DataChecksum of type " +
type + " with bytesPerChecksum " + bpc );
@ -111,13 +128,13 @@ public static DataChecksum newDataChecksum( DataInputStream in )
*/
public void writeHeader( DataOutputStream out )
throws IOException {
out.writeByte( type );
out.writeByte( type.id );
out.writeInt( bytesPerChecksum );
}
public byte[] getHeader() {
byte[] header = new byte[DataChecksum.HEADER_LEN];
header[0] = (byte) (type & 0xff);
header[0] = (byte) (type.id & 0xff);
// Writing in buffer just like DataOutput.WriteInt()
header[1+0] = (byte) ((bytesPerChecksum >>> 24) & 0xff);
header[1+1] = (byte) ((bytesPerChecksum >>> 16) & 0xff);
@ -133,11 +150,11 @@ public byte[] getHeader() {
*/
public int writeValue( DataOutputStream out, boolean reset )
throws IOException {
if ( size <= 0 ) {
if ( type.size <= 0 ) {
return 0;
}
if ( size == 4 ) {
if ( type.size == 4 ) {
out.writeInt( (int) summer.getValue() );
} else {
throw new IOException( "Unknown Checksum " + type );
@ -147,7 +164,7 @@ public int writeValue( DataOutputStream out, boolean reset )
reset();
}
return size;
return type.size;
}
/**
@ -157,11 +174,11 @@ public int writeValue( DataOutputStream out, boolean reset )
*/
public int writeValue( byte[] buf, int offset, boolean reset )
throws IOException {
if ( size <= 0 ) {
if ( type.size <= 0 ) {
return 0;
}
if ( size == 4 ) {
if ( type.size == 4 ) {
int checksum = (int) summer.getValue();
buf[offset+0] = (byte) ((checksum >>> 24) & 0xff);
buf[offset+1] = (byte) ((checksum >>> 16) & 0xff);
@ -175,7 +192,7 @@ public int writeValue( byte[] buf, int offset, boolean reset )
reset();
}
return size;
return type.size;
}
/**
@ -183,36 +200,33 @@ public int writeValue( byte[] buf, int offset, boolean reset )
* @return true if the checksum matches and false otherwise.
*/
public boolean compare( byte buf[], int offset ) {
if ( size == 4 ) {
if ( type.size == 4 ) {
int checksum = ( (buf[offset+0] & 0xff) << 24 ) |
( (buf[offset+1] & 0xff) << 16 ) |
( (buf[offset+2] & 0xff) << 8 ) |
( (buf[offset+3] & 0xff) );
return checksum == (int) summer.getValue();
}
return size == 0;
return type.size == 0;
}
private final int type;
private final int size;
private final Type type;
private final Checksum summer;
private final int bytesPerChecksum;
private int inSum = 0;
private DataChecksum( int checksumType, Checksum checksum,
int sumSize, int chunkSize ) {
type = checksumType;
private DataChecksum( Type type, Checksum checksum, int chunkSize ) {
this.type = type;
summer = checksum;
size = sumSize;
bytesPerChecksum = chunkSize;
}
// Accessors
public int getChecksumType() {
public Type getChecksumType() {
return type;
}
public int getChecksumSize() {
return size;
return type.size;
}
public int getBytesPerChecksum() {
return bytesPerChecksum;
@ -260,7 +274,7 @@ public void update( int b ) {
public void verifyChunkedSums(ByteBuffer data, ByteBuffer checksums,
String fileName, long basePos)
throws ChecksumException {
if (size == 0) return;
if (type.size == 0) return;
if (data.hasArray() && checksums.hasArray()) {
verifyChunkedSums(
@ -270,7 +284,7 @@ public void verifyChunkedSums(ByteBuffer data, ByteBuffer checksums,
return;
}
if (NativeCrc32.isAvailable()) {
NativeCrc32.verifyChunkedSums(bytesPerChecksum, type, checksums, data,
NativeCrc32.verifyChunkedSums(bytesPerChecksum, type.id, checksums, data,
fileName, basePos);
return;
}
@ -280,7 +294,7 @@ public void verifyChunkedSums(ByteBuffer data, ByteBuffer checksums,
checksums.mark();
try {
byte[] buf = new byte[bytesPerChecksum];
byte[] sum = new byte[size];
byte[] sum = new byte[type.size];
while (data.remaining() > 0) {
int n = Math.min(data.remaining(), bytesPerChecksum);
checksums.get(sum);
@ -351,7 +365,7 @@ private void verifyChunkedSums(
* buffer to put the checksums.
*/
public void calculateChunkedSums(ByteBuffer data, ByteBuffer checksums) {
if (size == 0) return;
if (type.size == 0) return;
if (data.hasArray() && checksums.hasArray()) {
calculateChunkedSums(data.array(), data.arrayOffset() + data.position(), data.remaining(),
@ -411,18 +425,12 @@ public boolean equals(Object other) {
@Override
public int hashCode() {
return (this.type + 31) * this.bytesPerChecksum;
return (this.type.id + 31) * this.bytesPerChecksum;
}
@Override
public String toString() {
String strType;
if (type < NAMES.length && type > 0) {
strType = NAMES[type];
} else {
strType = String.valueOf(type);
}
return "DataChecksum(type=" + strType +
return "DataChecksum(type=" + type +
", chunkSize=" + bytesPerChecksum + ")";
}

83
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/util/LineReader.java
View File

@ -204,11 +204,13 @@ private int readDefaultLine(Text str, int maxLineLength, int maxBytesToConsume)
int startPosn = bufferPosn; //starting from where we left off the last time
if (bufferPosn >= bufferLength) {
startPosn = bufferPosn = 0;
if (prevCharCR)
if (prevCharCR) {
++bytesConsumed; //account for CR from previous read
}
bufferLength = in.read(buffer);
if (bufferLength <= 0)
if (bufferLength <= 0) {
break; // EOF
}
}
for (; bufferPosn < bufferLength; ++bufferPosn) { //search for newline
if (buffer[bufferPosn] == LF) {
@ -223,8 +225,9 @@ private int readDefaultLine(Text str, int maxLineLength, int maxBytesToConsume)
prevCharCR = (buffer[bufferPosn] == CR);
}
int readLength = bufferPosn - startPosn;
if (prevCharCR && newlineLength == 0)
if (prevCharCR && newlineLength == 0) {
--readLength; //CR at the end of the buffer
}
bytesConsumed += readLength;
int appendLength = readLength - newlineLength;
if (appendLength > maxLineLength - txtLength) {
@ -236,8 +239,9 @@ private int readDefaultLine(Text str, int maxLineLength, int maxBytesToConsume)
}
} while (newlineLength == 0 && bytesConsumed < maxBytesToConsume);
if (bytesConsumed > (long)Integer.MAX_VALUE)
throw new IOException("Too many bytes before newline: " + bytesConsumed);
if (bytesConsumed > (long)Integer.MAX_VALUE) {
throw new IOException("Too many bytes before newline: " + bytesConsumed);
}
return (int)bytesConsumed;
}
@ -246,18 +250,56 @@ private int readDefaultLine(Text str, int maxLineLength, int maxBytesToConsume)
*/
private int readCustomLine(Text str, int maxLineLength, int maxBytesToConsume)
throws IOException {
/* We're reading data from inputStream, but the head of the stream may be
* already captured in the previous buffer, so we have several cases:
*
* 1. The buffer tail does not contain any character sequence which
* matches with the head of delimiter. We count it as a
* ambiguous byte count = 0
*
* 2. The buffer tail contains a X number of characters,
* that forms a sequence, which matches with the
* head of delimiter. We count ambiguous byte count = X
*
* // *** eg: A segment of input file is as follows
*
* " record 1792: I found this bug very interesting and
* I have completely read about it. record 1793: This bug
* can be solved easily record 1794: This ."
*
* delimiter = "record";
*
* supposing:- String at the end of buffer =
* "I found this bug very interesting and I have completely re"
* There for next buffer = "ad about it. record 179 ...."
*
* The matching characters in the input
* buffer tail and delimiter head = "re"
* Therefore, ambiguous byte count = 2 **** //
*
* 2.1 If the following bytes are the remaining characters of
* the delimiter, then we have to capture only up to the starting
* position of delimiter. That means, we need not include the
* ambiguous characters in str.
*
* 2.2 If the following bytes are not the remaining characters of
* the delimiter ( as mentioned in the example ),
* then we have to include the ambiguous characters in str.
*/
str.clear();
int txtLength = 0; // tracks str.getLength(), as an optimization
long bytesConsumed = 0;
int delPosn = 0;
int ambiguousByteCount=0; // To capture the ambiguous characters count
do {
int startPosn = bufferPosn; // starting from where we left off the last
// time
int startPosn = bufferPosn; // Start from previous end position
if (bufferPosn >= bufferLength) {
startPosn = bufferPosn = 0;
bufferLength = in.read(buffer);
if (bufferLength <= 0)
if (bufferLength <= 0) {
str.append(recordDelimiterBytes, 0, ambiguousByteCount);
break; // EOF
}
}
for (; bufferPosn < bufferLength; ++bufferPosn) {
if (buffer[bufferPosn] == recordDelimiterBytes[delPosn]) {
@ -266,7 +308,8 @@ private int readCustomLine(Text str, int maxLineLength, int maxBytesToConsume)
bufferPosn++;
break;
}
} else {
} else if (delPosn != 0) {
bufferPosn--;
delPosn = 0;
}
}
@ -277,14 +320,27 @@ private int readCustomLine(Text str, int maxLineLength, int maxBytesToConsume)
appendLength = maxLineLength - txtLength;
}
if (appendLength > 0) {
if (ambiguousByteCount > 0) {
str.append(recordDelimiterBytes, 0, ambiguousByteCount);
//appending the ambiguous characters (refer case 2.2)
bytesConsumed += ambiguousByteCount;
ambiguousByteCount=0;
}
str.append(buffer, startPosn, appendLength);
txtLength += appendLength;
}
} while (delPosn < recordDelimiterBytes.length
if (bufferPosn >= bufferLength) {
if (delPosn > 0 && delPosn < recordDelimiterBytes.length) {
ambiguousByteCount = delPosn;
bytesConsumed -= ambiguousByteCount; //to be consumed in next
}
}
} while (delPosn < recordDelimiterBytes.length
&& bytesConsumed < maxBytesToConsume);
if (bytesConsumed > (long) Integer.MAX_VALUE)
if (bytesConsumed > (long) Integer.MAX_VALUE) {
throw new IOException("Too many bytes before delimiter: " + bytesConsumed);
return (int) bytesConsumed;
}
return (int) bytesConsumed;
}
/**
@ -296,7 +352,7 @@ private int readCustomLine(Text str, int maxLineLength, int maxBytesToConsume)
*/
public int readLine(Text str, int maxLineLength) throws IOException {
return readLine(str, maxLineLength, Integer.MAX_VALUE);
}
}
/**
* Read from the InputStream into the given Text.
@ -307,5 +363,4 @@ public int readLine(Text str, int maxLineLength) throws IOException {
public int readLine(Text str) throws IOException {
return readLine(str, Integer.MAX_VALUE, Integer.MAX_VALUE);
}
}

2
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/util/NativeCodeLoader.java
View File

@ -47,7 +47,7 @@ public class NativeCodeLoader {
}
try {
System.loadLibrary("hadoop");
LOG.info("Loaded the native-hadoop library");
LOG.debug("Loaded the native-hadoop library");
nativeCodeLoaded = true;
} catch (Throwable t) {
// Ignore failure to load

2
hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Compressor.c
View File

@ -24,7 +24,7 @@
// Simple Functions
//****************************
extern int LZ4_compress (char* source, char* dest, int isize);
extern int LZ4_compress (const char* source, char* dest, int isize);
/*
LZ4_compress() :

2
hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Decompressor.c
View File

@ -20,7 +20,7 @@
#include "org_apache_hadoop.h"
#include "org_apache_hadoop_io_compress_lz4_Lz4Decompressor.h"
int LZ4_uncompress_unknownOutputSize (char* source, char* dest, int isize, int maxOutputSize);
int LZ4_uncompress_unknownOutputSize(const char* source, char* dest, int isize, int maxOutputSize);
/*
LZ4_uncompress_unknownOutputSize() :

22
hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/snappy/SnappyCompressor.c
View File

@ -25,6 +25,8 @@
#include "org_apache_hadoop_io_compress_snappy.h"
#include "org_apache_hadoop_io_compress_snappy_SnappyCompressor.h"
#define JINT_MAX 0x7fffffff
static jfieldID SnappyCompressor_clazz;
static jfieldID SnappyCompressor_uncompressedDirectBuf;
static jfieldID SnappyCompressor_uncompressedDirectBufLen;
@ -39,7 +41,7 @@ JNIEXPORT void JNICALL Java_org_apache_hadoop_io_compress_snappy_SnappyCompresso
// Load libsnappy.so
void *libsnappy = dlopen(HADOOP_SNAPPY_LIBRARY, RTLD_LAZY | RTLD_GLOBAL);
if (!libsnappy) {
char* msg = (char*)malloc(1000);
char msg[1000];
snprintf(msg, 1000, "%s (%s)!", "Cannot load " HADOOP_SNAPPY_LIBRARY, dlerror());
THROW(env, "java/lang/UnsatisfiedLinkError", msg);
return;
@ -71,6 +73,7 @@ JNIEXPORT jint JNICALL Java_org_apache_hadoop_io_compress_snappy_SnappyCompresso
jint uncompressed_direct_buf_len = (*env)->GetIntField(env, thisj, SnappyCompressor_uncompressedDirectBufLen);
jobject compressed_direct_buf = (*env)->GetObjectField(env, thisj, SnappyCompressor_compressedDirectBuf);
jint compressed_direct_buf_len = (*env)->GetIntField(env, thisj, SnappyCompressor_directBufferSize);
size_t buf_len;
// Get the input direct buffer
LOCK_CLASS(env, clazz, "SnappyCompressor");
@ -78,7 +81,7 @@ JNIEXPORT jint JNICALL Java_org_apache_hadoop_io_compress_snappy_SnappyCompresso
UNLOCK_CLASS(env, clazz, "SnappyCompressor");
if (uncompressed_bytes == 0) {
return (jint)0;
return 0;
}
// Get the output direct buffer
@ -87,15 +90,22 @@ JNIEXPORT jint JNICALL Java_org_apache_hadoop_io_compress_snappy_SnappyCompresso
UNLOCK_CLASS(env, clazz, "SnappyCompressor");
if (compressed_bytes == 0) {
return (jint)0;
return 0;
}
snappy_status ret = dlsym_snappy_compress(uncompressed_bytes, uncompressed_direct_buf_len, compressed_bytes, &compressed_direct_buf_len);
/* size_t should always be 4 bytes or larger. */
buf_len = (size_t)compressed_direct_buf_len;
snappy_status ret = dlsym_snappy_compress(uncompressed_bytes,
uncompressed_direct_buf_len, compressed_bytes, &buf_len);
if (ret != SNAPPY_OK){
THROW(env, "Ljava/lang/InternalError", "Could not compress data. Buffer length is too small.");
return 0;
}
if (buf_len > JINT_MAX) {
THROW(env, "Ljava/lang/InternalError", "Invalid return buffer length.");
return 0;
}
(*env)->SetIntField(env, thisj, SnappyCompressor_uncompressedDirectBufLen, 0);
return (jint)compressed_direct_buf_len;
return (jint)buf_len;
}

20
hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/nativeio/NativeIO.c
View File

@ -16,6 +16,8 @@
* limitations under the License.
*/
#define _GNU_SOURCE
#include <assert.h>
#include <errno.h>
#include <fcntl.h>
@ -366,23 +368,15 @@ Java_org_apache_hadoop_io_nativeio_NativeIO_chmod(
*/
static void throw_ioe(JNIEnv* env, int errnum)
{
const char* message;
char buffer[80];
char message[80];
jstring jstr_message;
buffer[0] = 0;
#ifdef STRERROR_R_CHAR_P
// GNU strerror_r
message = strerror_r(errnum, buffer, sizeof(buffer));
assert (message != NULL);
#else
int ret = strerror_r(errnum, buffer, sizeof(buffer));
if (ret == 0) {
message = buffer;
if ((errnum >= 0) && (errnum < sys_nerr)) {
snprintf(message, sizeof(message), "%s", sys_errlist[errnum]);
} else {
message = "Unknown error";
snprintf(message, sizeof(message), "Unknown error %d", errnum);
}
#endif
jobject errno_obj = errno_to_enum(env, errnum);
if ((jstr_message = (*env)->NewStringUTF(env, message)) == NULL)

6
hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/security/JniBasedUnixGroupsMapping.c
View File

@ -40,8 +40,8 @@ Java_org_apache_hadoop_security_JniBasedUnixGroupsMapping_getGroupForUser
(JNIEnv *env, jobject jobj, jstring juser) {
extern int getGroupIDList(const char *user, int *ngroups, gid_t **groups);
extern int getGroupDetails(gid_t group, char **grpBuf);
jobjectArray jgroups;
const char *cuser = NULL;
jobjectArray jgroups = NULL;
int error = -1;
if (emptyGroups == NULL) {
@ -56,7 +56,7 @@ Java_org_apache_hadoop_security_JniBasedUnixGroupsMapping_getGroupForUser
}
}
char *grpBuf = NULL;
const char *cuser = (*env)->GetStringUTFChars(env, juser, NULL);
cuser = (*env)->GetStringUTFChars(env, juser, NULL);
if (cuser == NULL) {
goto cleanup;
}

5
hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/security/JniBasedUnixGroupsNetgroupMapping.c
View File

@ -45,6 +45,8 @@ typedef struct listElement UserList;
JNIEXPORT jobjectArray JNICALL
Java_org_apache_hadoop_security_JniBasedUnixGroupsNetgroupMapping_getUsersForNetgroupJNI
(JNIEnv *env, jobject jobj, jstring jgroup) {
UserList *userListHead = NULL;
int userListSize = 0;
// pointers to free at the end
const char *cgroup = NULL;
@ -65,9 +67,6 @@ Java_org_apache_hadoop_security_JniBasedUnixGroupsNetgroupMapping_getUsersForNet
// get users
// see man pages for setnetgrent, getnetgrent and endnetgrent
UserList *userListHead = NULL;
int userListSize = 0;
// set the name of the group for subsequent calls to getnetgrent
// note that we want to end group lokup regardless whether setnetgrent
// was successful or not (as long as it was called we need to call

3
hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/util/NativeCrc32.c
View File

@ -18,6 +18,7 @@
#include <arpa/inet.h>
#include <assert.h>
#include <inttypes.h>
#include <stdlib.h>
#include <stdint.h>
#include <string.h>
@ -50,7 +51,7 @@ static void throw_checksum_exception(JNIEnv *env,
// Format error message
snprintf(message, sizeof(message),
"Checksum error: %s at %ld exp: %d got: %d",
"Checksum error: %s at %"PRId64" exp: %"PRId32" got: %"PRId32,
filename, pos, expected_crc, got_crc);
if ((jstr_message = (*env)->NewStringUTF(env, message)) == NULL) {
goto cleanup;

2
hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/util/bulk_crc32.c
View File

@ -41,7 +41,7 @@ static uint32_t crc32c_sb8(uint32_t crc, const uint8_t *buf, size_t length);
#ifdef USE_PIPELINED
static void pipelined_crc32c(uint32_t *crc1, uint32_t *crc2, uint32_t *crc3, const uint8_t *p_buf, size_t block_size, int num_blocks);
#endif USE_PIPELINED
#endif
static int cached_cpu_supports_crc32; // initialized by constructor below
static uint32_t crc32c_hardware(uint32_t crc, const uint8_t* data, size_t length);

20
hadoop-common-project/hadoop-common/src/main/resources/core-default.xml
View File

@ -351,8 +351,12 @@
<name>fs.trash.interval</name>
<value>0</value>
<description>Number of minutes after which the checkpoint
gets deleted.
If zero, the trash feature is disabled.
gets deleted. If zero, the trash feature is disabled.
This option may be configured both on the server and the
client. If trash is disabled server side then the client
side configuration is checked. If trash is enabled on the
server side then the value configured on the server is
used and the client configuration value is ignored.
</description>
</property>
@ -360,7 +364,8 @@
<name>fs.trash.checkpoint.interval</name>
<value>0</value>
<description>Number of minutes between trash checkpoints.
Should be smaller or equal to fs.trash.interval.
Should be smaller or equal to fs.trash.interval. If zero,
the value is set to the value of fs.trash.interval.
Every time the checkpointer runs it creates a new checkpoint
out of current and removes checkpoints created more than
fs.trash.interval minutes ago.
@ -1083,4 +1088,13 @@
</description>
</property>
<property>
<name>hadoop.jetty.logs.serve.aliases</name>
<value>true</value>
<description>
Enable/Disable aliases serving from jetty
</description>
</property>
</configuration>

7
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/conf/TestConfiguration.java
View File

@ -39,6 +39,7 @@
import junit.framework.TestCase;
import static org.junit.Assert.assertArrayEquals;
import static org.junit.Assert.assertNotNull;
import org.apache.commons.lang.StringUtils;
import org.apache.hadoop.conf.Configuration.IntegerRanges;
@ -1157,6 +1158,12 @@ public void testSetPattern() {
configuration.getPattern("testPattern", Pattern.compile("")).pattern());
}
public void testGetClassByNameOrNull() throws Exception {
Configuration config = new Configuration();
Class<?> clazz = config.getClassByNameOrNull("java.lang.Object");
assertNotNull(clazz);
}
public static void main(String[] argv) throws Exception {
junit.textui.TestRunner.main(new String[]{
TestConfiguration.class.getName()

20
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/FSMainOperationsBaseTest.java
View File

@ -364,15 +364,17 @@ public void testListStatusFilterWithSomeMatches() throws Exception {
}
@Test
public void testGlobStatusThrowsExceptionForNonExistentFile() throws Exception {
try {
// This should throw a FileNotFoundException
fSys.globStatus(
getTestRootPath(fSys, "test/hadoopfsdf/?"));
Assert.fail("Should throw FileNotFoundException");
} catch (FileNotFoundException fnfe) {
// expected
}
public void testGlobStatusNonExistentFile() throws Exception {
FileStatus[] paths = fSys.globStatus(
getTestRootPath(fSys, "test/hadoopfsdf"));
Assert.assertNull(paths);
paths = fSys.globStatus(
getTestRootPath(fSys, "test/hadoopfsdf/?"));
Assert.assertEquals(0, paths.length);
paths = fSys.globStatus(
getTestRootPath(fSys, "test/hadoopfsdf/xyz*/?"));
Assert.assertEquals(0, paths.length);
}
@Test

20
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/FileContextMainOperationsBaseTest.java
View File

@ -360,15 +360,17 @@ public void testListStatusFilterWithSomeMatches() throws Exception {
}
@Test
public void testGlobStatusThrowsExceptionForNonExistentFile() throws Exception {
try {
// This should throw a FileNotFoundException
fc.util().globStatus(
getTestRootPath(fc, "test/hadoopfsdf/?"));
Assert.fail("Should throw FileNotFoundException");
} catch (FileNotFoundException fnfe) {
// expected
}
public void testGlobStatusNonExistentFile() throws Exception {
FileStatus[] paths = fc.util().globStatus(
getTestRootPath(fc, "test/hadoopfsdf"));
Assert.assertNull(paths);
paths = fc.util().globStatus(
getTestRootPath(fc, "test/hadoopfsdf/?"));
Assert.assertEquals(0, paths.length);
paths = fc.util().globStatus(
getTestRootPath(fc, "test/hadoopfsdf/xyz*/?"));
Assert.assertEquals(0, paths.length);
}
@Test

37
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/FileSystemTestHelper.java
View File

@ -24,8 +24,10 @@
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.security.token.Token;
import org.junit.Assert;
import static org.junit.Assert.*;
import static org.mockito.Mockito.mock;
/**
* Helper class for unit tests.
@ -218,4 +220,39 @@ public static void checkFileStatus(FileSystem aFs, String path,
}
Assert.assertEquals(aFs.makeQualified(new Path(path)), s.getPath());
}
/**
* Class to enable easier mocking of a FileSystem
* Use getRawFileSystem to retrieve the mock
*/
public static class MockFileSystem extends FilterFileSystem {
public MockFileSystem() {
// it's a bit ackward to mock ourselves, but it allows the visibility
// of methods to be increased
super(mock(MockFileSystem.class));
}
@Override
public MockFileSystem getRawFileSystem() {
return (MockFileSystem) super.getRawFileSystem();
}
// these basic methods need to directly propagate to the mock to be
// more transparent
@Override
public void initialize(URI uri, Configuration conf) throws IOException {
fs.initialize(uri, conf);
}
@Override
public String getCanonicalServiceName() {
return fs.getCanonicalServiceName();
}
@Override
public FileSystem[] getChildFileSystems() {
return fs.getChildFileSystems();
}
@Override // publicly expose for mocking
public Token<?> getDelegationToken(String renewer) throws IOException {
return fs.getDelegationToken(renewer);
}
}
}

3
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestAfsCheckPath.java
View File

@ -24,6 +24,7 @@
import java.util.EnumSet;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.security.AccessControlException;
import org.apache.hadoop.util.Progressable;
import org.junit.Test;
@ -76,7 +77,7 @@ public int getUriDefaultPort() {
@Override
public FSDataOutputStream createInternal(Path f, EnumSet<CreateFlag> flag,
FsPermission absolutePermission, int bufferSize, short replication,
long blockSize, Progressable progress, int bytesPerChecksum,
long blockSize, Progressable progress, ChecksumOpt checksumOpt,
boolean createParent) throws IOException {
// deliberately empty
return null;

12
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestFileSystemCanonicalization.java
View File

@ -18,18 +18,20 @@
package org.apache.hadoop.fs;
import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertNotNull;
import java.io.IOException;
import java.net.URI;
import junit.framework.TestCase;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.security.NetUtilsTestResolver;
import org.apache.hadoop.util.Progressable;
import org.junit.BeforeClass;
import org.junit.Test;
public class TestFileSystemCanonicalization extends TestCase {
public class TestFileSystemCanonicalization {
static String[] authorities = {
"myfs://host",
"myfs://host.a",
@ -41,8 +43,8 @@ public class TestFileSystemCanonicalization extends TestCase {
};
@Test
public void testSetupResolver() throws Exception {
@BeforeClass
public static void initialize() throws Exception {
NetUtilsTestResolver.install();
}

279
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestFileSystemTokens.java Normal file
View File

@ -0,0 +1,279 @@
/**
* 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 static org.junit.Assert.*;
import static org.mockito.Matchers.*;
import static org.mockito.Mockito.*;
import java.io.IOException;
import org.apache.hadoop.fs.FileSystemTestHelper.MockFileSystem;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.security.Credentials;
import org.apache.hadoop.security.token.Token;
import org.apache.hadoop.security.token.TokenIdentifier;
import org.junit.Test;
import org.mockito.invocation.InvocationOnMock;
import org.mockito.stubbing.Answer;
public class TestFileSystemTokens {
private static String renewer = "renewer!";
@Test
public void testFsWithNoToken() throws Exception {
MockFileSystem fs = createFileSystemForServiceName(null);
Credentials credentials = new Credentials();
fs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(fs, false);
assertEquals(0, credentials.numberOfTokens());
}
@Test
public void testFsWithToken() throws Exception {
Text service = new Text("singleTokenFs");
MockFileSystem fs = createFileSystemForServiceName(service);
Credentials credentials = new Credentials();
fs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(fs, true);
assertEquals(1, credentials.numberOfTokens());
assertNotNull(credentials.getToken(service));
}
@Test
public void testFsWithTokenExists() throws Exception {
Credentials credentials = new Credentials();
Text service = new Text("singleTokenFs");
MockFileSystem fs = createFileSystemForServiceName(service);
Token<?> token = mock(Token.class);
credentials.addToken(service, token);
fs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(fs, false);
assertEquals(1, credentials.numberOfTokens());
assertSame(token, credentials.getToken(service));
}
@Test
public void testFsWithChildTokens() throws Exception {
Credentials credentials = new Credentials();
Text service1 = new Text("singleTokenFs1");
Text service2 = new Text("singleTokenFs2");
MockFileSystem fs1 = createFileSystemForServiceName(service1);
MockFileSystem fs2 = createFileSystemForServiceName(service2);
MockFileSystem fs3 = createFileSystemForServiceName(null);
MockFileSystem multiFs =
createFileSystemForServiceName(null, fs1, fs2, fs3);
multiFs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(multiFs, false); // has no tokens of own, only child tokens
verifyTokenFetch(fs1, true);
verifyTokenFetch(fs2, true);
verifyTokenFetch(fs3, false);
assertEquals(2, credentials.numberOfTokens());
assertNotNull(credentials.getToken(service1));
assertNotNull(credentials.getToken(service2));
}
@Test
public void testFsWithDuplicateChildren() throws Exception {
Credentials credentials = new Credentials();
Text service = new Text("singleTokenFs1");
MockFileSystem fs = createFileSystemForServiceName(service);
MockFileSystem multiFs =
createFileSystemForServiceName(null, fs, new FilterFileSystem(fs));
multiFs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(multiFs, false);
verifyTokenFetch(fs, true);
assertEquals(1, credentials.numberOfTokens());
assertNotNull(credentials.getToken(service));
}
@Test
public void testFsWithDuplicateChildrenTokenExists() throws Exception {
Credentials credentials = new Credentials();
Text service = new Text("singleTokenFs1");
Token<?> token = mock(Token.class);
credentials.addToken(service, token);
MockFileSystem fs = createFileSystemForServiceName(service);
MockFileSystem multiFs =
createFileSystemForServiceName(null, fs, new FilterFileSystem(fs));
multiFs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(multiFs, false);
verifyTokenFetch(fs, false);
assertEquals(1, credentials.numberOfTokens());
assertSame(token, credentials.getToken(service));
}
@Test
public void testFsWithChildTokensOneExists() throws Exception {
Credentials credentials = new Credentials();
Text service1 = new Text("singleTokenFs1");
Text service2 = new Text("singleTokenFs2");
Token<?> token = mock(Token.class);
credentials.addToken(service2, token);
MockFileSystem fs1 = createFileSystemForServiceName(service1);
MockFileSystem fs2 = createFileSystemForServiceName(service2);
MockFileSystem fs3 = createFileSystemForServiceName(null);
MockFileSystem multiFs = createFileSystemForServiceName(null, fs1, fs2, fs3);
multiFs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(multiFs, false);
verifyTokenFetch(fs1, true);
verifyTokenFetch(fs2, false); // we had added its token to credentials
verifyTokenFetch(fs3, false);
assertEquals(2, credentials.numberOfTokens());
assertNotNull(credentials.getToken(service1));
assertSame(token, credentials.getToken(service2));
}
@Test
public void testFsWithMyOwnAndChildTokens() throws Exception {
Credentials credentials = new Credentials();
Text service1 = new Text("singleTokenFs1");
Text service2 = new Text("singleTokenFs2");
Text myService = new Text("multiTokenFs");
Token<?> token = mock(Token.class);
credentials.addToken(service2, token);
MockFileSystem fs1 = createFileSystemForServiceName(service1);
MockFileSystem fs2 = createFileSystemForServiceName(service2);
MockFileSystem multiFs = createFileSystemForServiceName(myService, fs1, fs2);
multiFs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(multiFs, true); // its own token and also of its children
verifyTokenFetch(fs1, true);
verifyTokenFetch(fs2, false); // we had added its token to credentials
assertEquals(3, credentials.numberOfTokens());
assertNotNull(credentials.getToken(myService));
assertNotNull(credentials.getToken(service1));
assertNotNull(credentials.getToken(service2));
}
@Test
public void testFsWithMyOwnExistsAndChildTokens() throws Exception {
Credentials credentials = new Credentials();
Text service1 = new Text("singleTokenFs1");
Text service2 = new Text("singleTokenFs2");
Text myService = new Text("multiTokenFs");
Token<?> token = mock(Token.class);
credentials.addToken(myService, token);
MockFileSystem fs1 = createFileSystemForServiceName(service1);
MockFileSystem fs2 = createFileSystemForServiceName(service2);
MockFileSystem multiFs = createFileSystemForServiceName(myService, fs1, fs2);
multiFs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(multiFs, false); // we had added its token to credentials
verifyTokenFetch(fs1, true);
verifyTokenFetch(fs2, true);
assertEquals(3, credentials.numberOfTokens());
assertSame(token, credentials.getToken(myService));
assertNotNull(credentials.getToken(service1));
assertNotNull(credentials.getToken(service2));
}
@Test
public void testFsWithNestedDuplicatesChildren() throws Exception {
Credentials credentials = new Credentials();
Text service1 = new Text("singleTokenFs1");
Text service2 = new Text("singleTokenFs2");
Text service4 = new Text("singleTokenFs4");
Text multiService = new Text("multiTokenFs");
Token<?> token2 = mock(Token.class);
credentials.addToken(service2, token2);
MockFileSystem fs1 = createFileSystemForServiceName(service1);
MockFileSystem fs1B = createFileSystemForServiceName(service1);
MockFileSystem fs2 = createFileSystemForServiceName(service2);
MockFileSystem fs3 = createFileSystemForServiceName(null);
MockFileSystem fs4 = createFileSystemForServiceName(service4);
// now let's get dirty! ensure dup tokens aren't fetched even when
// repeated and dupped in a nested fs. fs4 is a real test of the drill
// down: multi-filter-multi-filter-filter-fs4.
MockFileSystem multiFs = createFileSystemForServiceName(multiService,
fs1, fs1B, fs2, fs2, new FilterFileSystem(fs3),
new FilterFileSystem(new FilterFileSystem(fs4)));
MockFileSystem superMultiFs = createFileSystemForServiceName(null,
fs1, fs1B, fs1, new FilterFileSystem(fs3), new FilterFileSystem(multiFs));
superMultiFs.addDelegationTokens(renewer, credentials);
verifyTokenFetch(superMultiFs, false); // does not have its own token
verifyTokenFetch(multiFs, true); // has its own token
verifyTokenFetch(fs1, true);
verifyTokenFetch(fs2, false); // we had added its token to credentials
verifyTokenFetch(fs3, false); // has no tokens
verifyTokenFetch(fs4, true);
assertEquals(4, credentials.numberOfTokens()); //fs1+fs2+fs4+multifs (fs3=0)
assertNotNull(credentials.getToken(service1));
assertNotNull(credentials.getToken(service2));
assertSame(token2, credentials.getToken(service2));
assertNotNull(credentials.getToken(multiService));
assertNotNull(credentials.getToken(service4));
}
public static MockFileSystem createFileSystemForServiceName(
final Text service, final FileSystem... children) throws IOException {
final MockFileSystem fs = new MockFileSystem();
final MockFileSystem mockFs = fs.getRawFileSystem();
if (service != null) {
when(mockFs.getCanonicalServiceName()).thenReturn(service.toString());
when(mockFs.getDelegationToken(any(String.class))).thenAnswer(
new Answer<Token<?>>() {
@Override
public Token<?> answer(InvocationOnMock invocation) throws Throwable {
Token<?> token = new Token<TokenIdentifier>();
token.setService(service);
return token;
}
});
}
when(mockFs.getChildFileSystems()).thenReturn(children);
return fs;
}
// check that canonical name was requested, if renewer is not null that
// a token was requested, and that child fs was invoked
private void verifyTokenFetch(MockFileSystem fs, boolean expected) throws IOException {
verify(fs.getRawFileSystem(), atLeast(1)).getCanonicalServiceName();
if (expected) {
verify(fs.getRawFileSystem()).getDelegationToken(renewer);
} else {
verify(fs.getRawFileSystem(), never()).getDelegationToken(any(String.class));
}
verify(fs.getRawFileSystem(), atLeast(1)).getChildFileSystems();
}
}

21
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestFilterFileSystem.java
View File

@ -32,8 +32,10 @@
import org.apache.commons.logging.Log;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.Options.CreateOpts;
import org.apache.hadoop.fs.Options.Rename;
import org.apache.hadoop.security.Credentials;
import org.apache.hadoop.security.token.Token;
import org.apache.hadoop.util.Progressable;
import org.junit.BeforeClass;
@ -79,6 +81,11 @@ public FSDataOutputStream createNonRecursive(Path f, FsPermission permission,
Progressable progress) throws IOException {
return null;
}
public FSDataOutputStream createNonRecursive(Path f, FsPermission permission,
EnumSet<CreateFlag> flags, int bufferSize, short replication, long blockSize,
Progressable progress, ChecksumOpt checksumOpt) throws IOException {
return null;
}
public boolean mkdirs(Path f) { return false; }
public FSDataInputStream open(Path f) { return null; }
public FSDataOutputStream create(Path f) { return null; }
@ -137,6 +144,16 @@ public FSDataOutputStream create(Path f,
Progressable progress) throws IOException {
return null;
}
public FSDataOutputStream create(Path f,
FsPermission permission,
EnumSet<CreateFlag> flags,
int bufferSize,
short replication,
long blockSize,
Progressable progress,
ChecksumOpt checksumOpt) throws IOException {
return null;
}
public String getName() { return null; }
public boolean delete(Path f) { return false; }
public short getReplication(Path src) { return 0 ; }
@ -185,6 +202,10 @@ public boolean deleteOnExit(Path f) throws IOException {
public boolean cancelDeleteOnExit(Path f) throws IOException {
return false;
}
public Token<?>[] addDelegationTokens(String renewer, Credentials creds)
throws IOException {
return null;
}
public String getScheme() {
return "dontcheck";
}

70
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestFsOptions.java Normal file
View File

@ -0,0 +1,70 @@
/**
* 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 static org.junit.Assert.*;
import java.io.IOException;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.util.DataChecksum;
import org.junit.Test;
public class TestFsOptions {
@Test
public void testProcessChecksumOpt() {
ChecksumOpt defaultOpt = new ChecksumOpt(DataChecksum.Type.CRC32, 512);
ChecksumOpt finalOpt;
// Give a null
finalOpt = ChecksumOpt.processChecksumOpt(defaultOpt, null);
checkParams(defaultOpt, finalOpt);
// null with bpc
finalOpt = ChecksumOpt.processChecksumOpt(defaultOpt, null, 1024);
checkParams(DataChecksum.Type.CRC32, 1024, finalOpt);
ChecksumOpt myOpt = new ChecksumOpt();
// custom with unspecified parameters
finalOpt = ChecksumOpt.processChecksumOpt(defaultOpt, myOpt);
checkParams(defaultOpt, finalOpt);
myOpt = new ChecksumOpt(DataChecksum.Type.CRC32C, 2048);
// custom config
finalOpt = ChecksumOpt.processChecksumOpt(defaultOpt, myOpt);
checkParams(DataChecksum.Type.CRC32C, 2048, finalOpt);
// custom config + bpc
finalOpt = ChecksumOpt.processChecksumOpt(defaultOpt, myOpt, 4096);
checkParams(DataChecksum.Type.CRC32C, 4096, finalOpt);
}
private void checkParams(ChecksumOpt expected, ChecksumOpt obtained) {
assertEquals(expected.getChecksumType(), obtained.getChecksumType());
assertEquals(expected.getBytesPerChecksum(), obtained.getBytesPerChecksum());
}
private void checkParams(DataChecksum.Type type, int bpc, ChecksumOpt obtained) {
assertEquals(type, obtained.getChecksumType());
assertEquals(bpc, obtained.getBytesPerChecksum());
}
}

1
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestFsShellReturnCode.java
View File

@ -32,7 +32,6 @@
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.ftpserver.command.impl.STAT;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.shell.CommandFactory;
import org.apache.hadoop.fs.shell.FsCommand;

20
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestLocalDirAllocator.java
View File

@ -343,14 +343,18 @@ public void testGetLocalPathToRead() throws IOException {
@Test
public void testRemoveContext() throws IOException {
String dir = buildBufferDir(ROOT, 0);
String contextCfgItemName = "application_1340842292563_0004.app.cache.dirs";
conf.set(contextCfgItemName, dir);
LocalDirAllocator localDirAllocator = new LocalDirAllocator(
contextCfgItemName);
localDirAllocator.getLocalPathForWrite("p1/x", SMALL_FILE_SIZE, conf);
assertTrue(LocalDirAllocator.isContextValid(contextCfgItemName));
LocalDirAllocator.removeContext(contextCfgItemName);
assertFalse(LocalDirAllocator.isContextValid(contextCfgItemName));
try {
String contextCfgItemName = "application_1340842292563_0004.app.cache.dirs";
conf.set(contextCfgItemName, dir);
LocalDirAllocator localDirAllocator = new LocalDirAllocator(
contextCfgItemName);
localDirAllocator.getLocalPathForWrite("p1/x", SMALL_FILE_SIZE, conf);
assertTrue(LocalDirAllocator.isContextValid(contextCfgItemName));
LocalDirAllocator.removeContext(contextCfgItemName);
assertFalse(LocalDirAllocator.isContextValid(contextCfgItemName));
} finally {
rmBufferDirs();
}
}
}

4
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestLocalFileSystem.java
View File

@ -33,8 +33,8 @@
* This class tests the local file system via the FileSystem abstraction.
*/
public class TestLocalFileSystem {
private static String TEST_ROOT_DIR
= System.getProperty("test.build.data","build/test/data/work-dir/localfs");
private static final String TEST_ROOT_DIR
= System.getProperty("test.build.data","build/test/data") + "/work-dir/localfs";
private Configuration conf;
private FileSystem fileSys;

8
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/TestTrash.java
View File

@ -111,10 +111,10 @@ public static void trashShell(final Configuration conf, final Path base,
throws IOException {
FileSystem fs = FileSystem.get(conf);
conf.set(FS_TRASH_INTERVAL_KEY, "0"); // disabled
conf.setLong(FS_TRASH_INTERVAL_KEY, 0); // disabled
assertFalse(new Trash(conf).isEnabled());
conf.set(FS_TRASH_INTERVAL_KEY, "10"); // 10 minute
conf.setLong(FS_TRASH_INTERVAL_KEY, 10); // 10 minute
assertTrue(new Trash(conf).isEnabled());
FsShell shell = new FsShell();
@ -435,7 +435,7 @@ public static void trashShell(final Configuration conf, final Path base,
}
public static void trashNonDefaultFS(Configuration conf) throws IOException {
conf.set(FS_TRASH_INTERVAL_KEY, "10"); // 10 minute
conf.setLong(FS_TRASH_INTERVAL_KEY, 10); // 10 minute
// attempt non-default FileSystem trash
{
final FileSystem lfs = FileSystem.getLocal(conf);
@ -580,7 +580,7 @@ public static void performanceTestDeleteSameFile() throws IOException{
FileSystem fs = FileSystem.getLocal(conf);
conf.set("fs.defaultFS", fs.getUri().toString());
conf.set(FS_TRASH_INTERVAL_KEY, "10"); //minutes..
conf.setLong(FS_TRASH_INTERVAL_KEY, 10); //minutes..
FsShell shell = new FsShell();
shell.setConf(conf);
//Path trashRoot = null;

21
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/shell/TestCommandFactory.java
View File

@ -48,6 +48,10 @@ public void testRegistration() {
factory.addClass(TestCommand3.class, "tc3");
names = factory.getNames();
assertArrayEquals(new String []{"tc1", "tc2", "tc2.1", "tc3"}, names);
factory.addClass(TestCommand4.class, (new TestCommand4()).getName());
names = factory.getNames();
assertArrayEquals(new String[]{"tc1", "tc2", "tc2.1", "tc3", "tc4"}, names);
}
@Test
@ -72,8 +76,17 @@ public void testGetInstances() {
assertNotNull(instance);
assertEquals(TestCommand2.class, instance.getClass());
assertEquals("tc2.1", instance.getCommandName());
factory.addClass(TestCommand4.class, "tc4");
instance = factory.getInstance("tc4");
assertNotNull(instance);
assertEquals(TestCommand4.class, instance.getClass());
assertEquals("tc4", instance.getCommandName());
String usage = instance.getUsage();
assertEquals("-tc4 tc4_usage", usage);
assertEquals("tc4_description", instance.getDescription());
}
static class TestRegistrar {
public static void registerCommands(CommandFactory factory) {
factory.addClass(TestCommand1.class, "tc1");
@ -84,4 +97,10 @@ public static void registerCommands(CommandFactory factory) {
static class TestCommand1 extends FsCommand {}
static class TestCommand2 extends FsCommand {}
static class TestCommand3 extends FsCommand {}
static class TestCommand4 extends FsCommand {
static final String NAME = "tc4";
static final String USAGE = "tc4_usage";
static final String DESCRIPTION = "tc4_description";
}
}

78
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/shell/TestPathData.java
View File

@ -26,23 +26,17 @@
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.junit.BeforeClass;
import org.junit.After;
import org.junit.Before;
import org.junit.Test;
public class TestPathData {
protected static Configuration conf;
protected static FileSystem fs;
protected static String dirString;
protected static Path testDir;
protected static PathData item;
protected static String[] d1Paths =
new String[] { "d1/f1", "d1/f1.1", "d1/f2" };
protected static String[] d2Paths =
new String[] { "d2/f3" };
@BeforeClass
public static void initialize() throws Exception {
protected Configuration conf;
protected FileSystem fs;
protected Path testDir;
@Before
public void initialize() throws Exception {
conf = new Configuration();
fs = FileSystem.getLocal(conf);
testDir = new Path(
@ -60,23 +54,28 @@ public static void initialize() throws Exception {
fs.create(new Path("d2","f3"));
}
@After
public void cleanup() throws Exception {
fs.close();
}
@Test
public void testWithDirStringAndConf() throws Exception {
dirString = "d1";
item = new PathData(dirString, conf);
checkPathData();
String dirString = "d1";
PathData item = new PathData(dirString, conf);
checkPathData(dirString, item);
// properly implementing symlink support in various commands will require
// trailing slashes to be retained
dirString = "d1/";
item = new PathData(dirString, conf);
checkPathData();
checkPathData(dirString, item);
}
@Test
public void testUnqualifiedUriContents() throws Exception {
dirString = "d1";
item = new PathData(dirString, conf);
String dirString = "d1";
PathData item = new PathData(dirString, conf);
PathData[] items = item.getDirectoryContents();
assertEquals(
sortedString("d1/f1", "d1/f1.1", "d1/f2"),
@ -86,8 +85,8 @@ public void testUnqualifiedUriContents() throws Exception {
@Test
public void testQualifiedUriContents() throws Exception {
dirString = fs.makeQualified(new Path("d1")).toString();
item = new PathData(dirString, conf);
String dirString = fs.makeQualified(new Path("d1")).toString();
PathData item = new PathData(dirString, conf);
PathData[] items = item.getDirectoryContents();
assertEquals(
sortedString(dirString+"/f1", dirString+"/f1.1", dirString+"/f2"),
@ -97,8 +96,8 @@ public void testQualifiedUriContents() throws Exception {
@Test
public void testCwdContents() throws Exception {
dirString = Path.CUR_DIR;
item = new PathData(dirString, conf);
String dirString = Path.CUR_DIR;
PathData item = new PathData(dirString, conf);
PathData[] items = item.getDirectoryContents();
assertEquals(
sortedString("d1", "d2"),
@ -106,17 +105,16 @@ public void testCwdContents() throws Exception {
);
}
@Test
public void testToFile() throws Exception {
item = new PathData(".", conf);
@Test
public void testToFile() throws Exception {
PathData item = new PathData(".", conf);
assertEquals(new File(testDir.toString()), item.toFile());
item = new PathData("d1/f1", conf);
assertEquals(new File(testDir+"/d1/f1"), item.toFile());
item = new PathData(testDir+"/d1/f1", conf);
assertEquals(new File(testDir+"/d1/f1"), item.toFile());
}
item = new PathData("d1/f1", conf);
assertEquals(new File(testDir + "/d1/f1"), item.toFile());
item = new PathData(testDir + "/d1/f1", conf);
assertEquals(new File(testDir + "/d1/f1"), item.toFile());
}
@Test
public void testAbsoluteGlob() throws Exception {
PathData[] items = PathData.expandAsGlob(testDir+"/d1/f1*", conf);
@ -147,18 +145,18 @@ public void testRelativeGlobBack() throws Exception {
@Test
public void testWithStringAndConfForBuggyPath() throws Exception {
dirString = "file:///tmp";
testDir = new Path(dirString);
item = new PathData(dirString, conf);
String dirString = "file:///tmp";
Path tmpDir = new Path(dirString);
PathData item = new PathData(dirString, conf);
// this may fail some day if Path is fixed to not crunch the uri
// if the authority is null, however we need to test that the PathData
// toString() returns the given string, while Path toString() does
// the crunching
assertEquals("file:/tmp", testDir.toString());
checkPathData();
assertEquals("file:/tmp", tmpDir.toString());
checkPathData(dirString, item);
}
public void checkPathData() throws Exception {
public void checkPathData(String dirString, PathData item) throws Exception {
assertEquals("checking fs", fs, item.fs);
assertEquals("checking string", dirString, item.toString());
assertEquals("checking path",

121
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/viewfs/TestViewFileSystemDelegationTokenSupport.java
View File

@ -22,10 +22,18 @@
import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;
import java.util.Arrays;
import java.util.List;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.FsConstants;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.RawLocalFileSystem;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.security.Credentials;
import org.apache.hadoop.security.token.Token;
import org.apache.hadoop.security.token.TokenIdentifier;
import org.junit.BeforeClass;
import org.junit.Test;
/**
@ -38,6 +46,29 @@
public class TestViewFileSystemDelegationTokenSupport {
private static final String MOUNT_TABLE_NAME = "vfs-cluster";
static Configuration conf;
static FileSystem viewFs;
static FakeFileSystem fs1;
static FakeFileSystem fs2;
@BeforeClass
public static void setup() throws Exception {
conf = ViewFileSystemTestSetup.createConfig();
fs1 = setupFileSystem(new URI("fs1:///"), FakeFileSystem.class);
fs2 = setupFileSystem(new URI("fs2:///"), FakeFileSystem.class);
viewFs = FileSystem.get(FsConstants.VIEWFS_URI, conf);
}
static FakeFileSystem setupFileSystem(URI uri, Class<? extends FileSystem> clazz)
throws Exception {
String scheme = uri.getScheme();
conf.set("fs."+scheme+".impl", clazz.getName());
FakeFileSystem fs = (FakeFileSystem)FileSystem.get(uri, conf);
// mount each fs twice, will later ensure 1 token/fs
ConfigUtil.addLink(conf, "/mounts/"+scheme+"-one", fs.getUri());
ConfigUtil.addLink(conf, "/mounts/"+scheme+"-two", fs.getUri());
return fs;
}
/**
* Regression test for HADOOP-8408.
@ -69,4 +100,92 @@ public void testGetCanonicalServiceNameWithDefaultMountTable()
assertNull(serviceName);
}
@Test
public void testGetChildFileSystems() throws Exception {
assertNull(fs1.getChildFileSystems());
assertNull(fs2.getChildFileSystems());
List<FileSystem> children = Arrays.asList(viewFs.getChildFileSystems());
assertEquals(2, children.size());
assertTrue(children.contains(fs1));
assertTrue(children.contains(fs2));
}
@Test
public void testAddDelegationTokens() throws Exception {
Credentials creds = new Credentials();
Token<?> fs1Tokens[] = addTokensWithCreds(fs1, creds);
assertEquals(1, fs1Tokens.length);
assertEquals(1, creds.numberOfTokens());
Token<?> fs2Tokens[] = addTokensWithCreds(fs2, creds);
assertEquals(1, fs2Tokens.length);
assertEquals(2, creds.numberOfTokens());
Credentials savedCreds = creds;
creds = new Credentials();
// should get the same set of tokens as explicitly fetched above
Token<?> viewFsTokens[] = viewFs.addDelegationTokens("me", creds);
assertEquals(2, viewFsTokens.length);
assertTrue(creds.getAllTokens().containsAll(savedCreds.getAllTokens()));
assertEquals(savedCreds.numberOfTokens(), creds.numberOfTokens());
// should get none, already have all tokens
viewFsTokens = viewFs.addDelegationTokens("me", creds);
assertEquals(0, viewFsTokens.length);
assertTrue(creds.getAllTokens().containsAll(savedCreds.getAllTokens()));
assertEquals(savedCreds.numberOfTokens(), creds.numberOfTokens());
}
Token<?>[] addTokensWithCreds(FileSystem fs, Credentials creds) throws Exception {
Credentials savedCreds;
savedCreds = new Credentials(creds);
Token<?> tokens[] = fs.addDelegationTokens("me", creds);
// test that we got the token we wanted, and that creds were modified
assertEquals(1, tokens.length);
assertEquals(fs.getCanonicalServiceName(), tokens[0].getService().toString());
assertTrue(creds.getAllTokens().contains(tokens[0]));
assertTrue(creds.getAllTokens().containsAll(savedCreds.getAllTokens()));
assertEquals(savedCreds.numberOfTokens()+1, creds.numberOfTokens());
// shouldn't get any new tokens since already in creds
savedCreds = new Credentials(creds);
Token<?> tokenRefetch[] = fs.addDelegationTokens("me", creds);
assertEquals(0, tokenRefetch.length);
assertTrue(creds.getAllTokens().containsAll(savedCreds.getAllTokens()));
assertEquals(savedCreds.numberOfTokens(), creds.numberOfTokens());
return tokens;
}
static class FakeFileSystem extends RawLocalFileSystem {
URI uri;
public void initialize(URI name, Configuration conf) throws IOException {
this.uri = name;
}
@Override
public Path getInitialWorkingDirectory() {
return new Path("/"); // ctor calls getUri before the uri is inited...
}
public URI getUri() {
return uri;
}
@Override
public String getCanonicalServiceName() {
return String.valueOf(this.getUri()+"/"+this.hashCode());
}
@Override
public Token<?> getDelegationToken(String renewer) throws IOException {
Token<?> token = new Token<TokenIdentifier>();
token.setService(new Text(getCanonicalServiceName()));
return token;
}
@Override
public void close() {}
}
}

26
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/fs/viewfs/ViewFileSystemBaseTest.java
View File

@ -19,6 +19,7 @@
import java.io.FileNotFoundException;
import java.io.IOException;
import java.util.Arrays;
import java.util.List;
@ -137,9 +138,9 @@ int getExpectedMountPoints() {
*/
@Test
public void testGetDelegationTokens() throws IOException {
List<Token<?>> delTokens =
fsView.getDelegationTokens("sanjay");
Assert.assertEquals(getExpectedDelegationTokenCount(), delTokens.size());
Token<?>[] delTokens =
fsView.addDelegationTokens("sanjay", new Credentials());
Assert.assertEquals(getExpectedDelegationTokenCount(), delTokens.length);
}
int getExpectedDelegationTokenCount() {
@ -150,29 +151,20 @@ int getExpectedDelegationTokenCount() {
public void testGetDelegationTokensWithCredentials() throws IOException {
Credentials credentials = new Credentials();
List<Token<?>> delTokens =
fsView.getDelegationTokens("sanjay", credentials);
Arrays.asList(fsView.addDelegationTokens("sanjay", credentials));
int expectedTokenCount = getExpectedDelegationTokenCountWithCredentials();
Assert.assertEquals(expectedTokenCount, delTokens.size());
Credentials newCredentials = new Credentials();
for (int i = 0; i < expectedTokenCount / 2; i++) {
Token<?> token = delTokens.get(i);
credentials.addToken(token.getService(), token);
newCredentials.addToken(token.getService(), token);
}
List<Token<?>> delTokens2 =
fsView.getDelegationTokens("sanjay", credentials);
Assert.assertEquals(expectedTokenCount, delTokens2.size());
for (int i = 0; i < delTokens2.size(); i++) {
for (int j = 0; j < delTokens.size(); j++) {
if (delTokens.get(j) == delTokens2.get(i)) {
delTokens.remove(j);
break;
}
}
}
Assert.assertEquals((expectedTokenCount + 1) / 2, delTokens.size());
Arrays.asList(fsView.addDelegationTokens("sanjay", newCredentials));
Assert.assertEquals((expectedTokenCount + 1) / 2, delTokens2.size());
}
int getExpectedDelegationTokenCountWithCredentials() {

39
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/io/TestIOUtils.java
View File

@ -21,6 +21,8 @@
import static org.junit.Assert.assertEquals;
import static org.junit.Assert.fail;
import java.io.ByteArrayInputStream;
import java.io.EOFException;
import java.io.File;
import java.io.IOException;
import java.io.InputStream;
@ -175,4 +177,41 @@ public void testWrappedReadForCompressedData() throws IOException {
"Error while reading compressed data", ioe);
}
}
@Test
public void testSkipFully() throws IOException {
byte inArray[] = new byte[] {0, 1, 2, 3, 4};
ByteArrayInputStream in = new ByteArrayInputStream(inArray);
try {
in.mark(inArray.length);
IOUtils.skipFully(in, 2);
IOUtils.skipFully(in, 2);
try {
IOUtils.skipFully(in, 2);
fail("expected to get a PrematureEOFException");
} catch (EOFException e) {
assertEquals(e.getMessage(), "Premature EOF from inputStream " +
"after skipping 1 byte(s).");
}
in.reset();
try {
IOUtils.skipFully(in, 20);
fail("expected to get a PrematureEOFException");
} catch (EOFException e) {
assertEquals(e.getMessage(), "Premature EOF from inputStream " +
"after skipping 5 byte(s).");
}
in.reset();
IOUtils.skipFully(in, 5);
try {
IOUtils.skipFully(in, 10);
fail("expected to get a PrematureEOFException");
} catch (EOFException e) {
assertEquals(e.getMessage(), "Premature EOF from inputStream " +
"after skipping 0 byte(s).");
}
} finally {
in.close();
}
}
}

1
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/io/TestWritableName.java
View File

@ -83,6 +83,7 @@ public void testAddName() throws Exception {
Configuration conf = new Configuration();
String altName = testName + ".alt";
WritableName.setName(SimpleWritable.class, testName);
WritableName.addName(SimpleWritable.class, altName);
Class<?> test = WritableName.getClass(altName, conf);

13
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/ipc/TestServer.java
View File

@ -20,6 +20,7 @@
import static org.junit.Assert.*;
import java.io.IOException;
import java.net.BindException;
import java.net.InetSocketAddress;
import java.net.ServerSocket;
@ -115,4 +116,16 @@ public void testBindError() throws Exception {
socket.close();
}
}
@Test
public void testExceptionsHandler() throws IOException {
Server.ExceptionsHandler handler = new Server.ExceptionsHandler();
handler.addTerseExceptions(IOException.class);
handler.addTerseExceptions(RpcServerException.class, IpcException.class);
assertTrue(handler.isTerse(IOException.class));
assertTrue(handler.isTerse(RpcServerException.class));
assertTrue(handler.isTerse(IpcException.class));
assertFalse(handler.isTerse(RpcClientException.class));
}
}

19
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/security/TestCredentials.java
View File

@ -213,5 +213,22 @@ public void mergeAll() {
// new token & secret should be added
assertEquals(token[2], creds.getToken(service[2]));
assertEquals(secret[2], new Text(creds.getSecretKey(secret[2])));
}
}
@Test
public void testAddTokensToUGI() {
UserGroupInformation ugi = UserGroupInformation.createRemoteUser("someone");
Credentials creds = new Credentials();
for (int i=0; i < service.length; i++) {
creds.addToken(service[i], token[i]);
}
creds.addTokensToUGI(ugi);
creds = ugi.getCredentials();
for (int i=0; i < service.length; i++) {
assertSame(token[i], creds.getToken(service[i]));
}
assertEquals(service.length, creds.numberOfTokens());
}
}

105
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/security/TestGroupFallback.java Normal file
View File

@ -0,0 +1,105 @@
/**
* 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.security;
import static org.junit.Assert.assertTrue;
import java.util.List;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.CommonConfigurationKeys;
import org.apache.log4j.Level;
import org.apache.log4j.Logger;
import org.junit.Test;
public class TestGroupFallback {
public static final Log LOG = LogFactory.getLog(TestGroupFallback.class);
@Test
public void testGroupShell() throws Exception {
Logger.getRootLogger().setLevel(Level.DEBUG);
Configuration conf = new Configuration();
conf.set(CommonConfigurationKeys.HADOOP_SECURITY_GROUP_MAPPING,
"org.apache.hadoop.security.ShellBasedUnixGroupsMapping");
Groups groups = new Groups(conf);
String username = System.getProperty("user.name");
List<String> groupList = groups.getGroups(username);
LOG.info(username + " has GROUPS: " + groupList.toString());
assertTrue(groupList.size() > 0);
}
@Test
public void testNetgroupShell() throws Exception {
Logger.getRootLogger().setLevel(Level.DEBUG);
Configuration conf = new Configuration();
conf.set(CommonConfigurationKeys.HADOOP_SECURITY_GROUP_MAPPING,
"org.apache.hadoop.security.ShellBasedUnixGroupsNetgroupMapping");
Groups groups = new Groups(conf);
String username = System.getProperty("user.name");
List<String> groupList = groups.getGroups(username);
LOG.info(username + " has GROUPS: " + groupList.toString());
assertTrue(groupList.size() > 0);
}
@Test
public void testGroupWithFallback() throws Exception {
LOG.info("running 'mvn -Pnative -DTestGroupFallback clear test' will " +
"test the normal path and 'mvn -DTestGroupFallback clear test' will" +
" test the fall back functionality");
Logger.getRootLogger().setLevel(Level.DEBUG);
Configuration conf = new Configuration();
conf.set(CommonConfigurationKeys.HADOOP_SECURITY_GROUP_MAPPING,
"org.apache.hadoop.security.JniBasedUnixGroupsMappingWithFallback");
Groups groups = new Groups(conf);
String username = System.getProperty("user.name");
List<String> groupList = groups.getGroups(username);
LOG.info(username + " has GROUPS: " + groupList.toString());
assertTrue(groupList.size() > 0);
}
@Test
public void testNetgroupWithFallback() throws Exception {
LOG.info("running 'mvn -Pnative -DTestGroupFallback clear test' will " +
"test the normal path and 'mvn -DTestGroupFallback clear test' will" +
" test the fall back functionality");
Logger.getRootLogger().setLevel(Level.DEBUG);
Configuration conf = new Configuration();
conf.set(CommonConfigurationKeys.HADOOP_SECURITY_GROUP_MAPPING,
"org.apache.hadoop.security.JniBasedUnixGroupsNetgroupMappingWithFallback");
Groups groups = new Groups(conf);
String username = System.getProperty("user.name");
List<String> groupList = groups.getGroups(username);
LOG.info(username + " has GROUPS: " + groupList.toString());
assertTrue(groupList.size() > 0);
}
}

2
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/security/TestSecurityUtil.java
View File

@ -137,6 +137,7 @@ public void testGetHostFromPrincipal() {
@Test
public void testBuildDTServiceName() {
SecurityUtil.setTokenServiceUseIp(true);
assertEquals("127.0.0.1:123",
SecurityUtil.buildDTServiceName(URI.create("test://LocalHost"), 123)
);
@ -153,6 +154,7 @@ public void testBuildDTServiceName() {
@Test
public void testBuildTokenServiceSockAddr() {
SecurityUtil.setTokenServiceUseIp(true);
assertEquals("127.0.0.1:123",
SecurityUtil.buildTokenService(new InetSocketAddress("LocalHost", 123)).toString()
);

83
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/security/TestUserGroupInformation.java
View File

@ -19,8 +19,7 @@
import static org.junit.Assert.*;
import org.junit.*;
import org.mockito.Mockito;
import static org.mockito.Mockito.mock;
import static org.mockito.Mockito.*;
import java.io.BufferedReader;
import java.io.IOException;
@ -35,6 +34,7 @@
import javax.security.auth.login.LoginContext;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.metrics2.MetricsRecordBuilder;
import org.apache.hadoop.security.UserGroupInformation.AuthenticationMethod;
import org.apache.hadoop.security.token.Token;
@ -194,8 +194,6 @@ public void testEquals() throws Exception {
public void testEqualsWithRealUser() throws Exception {
UserGroupInformation realUgi1 = UserGroupInformation.createUserForTesting(
"RealUser", GROUP_NAMES);
UserGroupInformation realUgi2 = UserGroupInformation.createUserForTesting(
"RealUser", GROUP_NAMES);
UserGroupInformation proxyUgi1 = UserGroupInformation.createProxyUser(
USER_NAME, realUgi1);
UserGroupInformation proxyUgi2 =
@ -213,7 +211,82 @@ public void testGettingGroups() throws Exception {
assertArrayEquals(new String[]{GROUP1_NAME, GROUP2_NAME, GROUP3_NAME},
uugi.getGroupNames());
}
@SuppressWarnings("unchecked") // from Mockito mocks
@Test
public <T extends TokenIdentifier> void testAddToken() throws Exception {
UserGroupInformation ugi =
UserGroupInformation.createRemoteUser("someone");
Token<T> t1 = mock(Token.class);
Token<T> t2 = mock(Token.class);
Token<T> t3 = mock(Token.class);
// add token to ugi
ugi.addToken(t1);
checkTokens(ugi, t1);
// replace token t1 with t2 - with same key (null)
ugi.addToken(t2);
checkTokens(ugi, t2);
// change t1 service and add token
when(t1.getService()).thenReturn(new Text("t1"));
ugi.addToken(t1);
checkTokens(ugi, t1, t2);
// overwrite t1 token with t3 - same key (!null)
when(t3.getService()).thenReturn(new Text("t1"));
ugi.addToken(t3);
checkTokens(ugi, t2, t3);
// just try to re-add with new name
when(t1.getService()).thenReturn(new Text("t1.1"));
ugi.addToken(t1);
checkTokens(ugi, t1, t2, t3);
// just try to re-add with new name again
ugi.addToken(t1);
checkTokens(ugi, t1, t2, t3);
}
private void checkTokens(UserGroupInformation ugi, Token<?> ... tokens) {
// check the ugi's token collection
Collection<Token<?>> ugiTokens = ugi.getTokens();
for (Token<?> t : tokens) {
assertTrue(ugiTokens.contains(t));
}
assertEquals(tokens.length, ugiTokens.size());
// check the ugi's credentials
Credentials ugiCreds = ugi.getCredentials();
for (Token<?> t : tokens) {
assertSame(t, ugiCreds.getToken(t.getService()));
}
assertEquals(tokens.length, ugiCreds.numberOfTokens());
}
@SuppressWarnings("unchecked") // from Mockito mocks
@Test
public <T extends TokenIdentifier> void testAddNamedToken() throws Exception {
UserGroupInformation ugi =
UserGroupInformation.createRemoteUser("someone");
Token<T> t1 = mock(Token.class);
Text service1 = new Text("t1");
Text service2 = new Text("t2");
when(t1.getService()).thenReturn(service1);
// add token
ugi.addToken(service1, t1);
assertSame(t1, ugi.getCredentials().getToken(service1));
// add token with another name
ugi.addToken(service2, t1);
assertSame(t1, ugi.getCredentials().getToken(service1));
assertSame(t1, ugi.getCredentials().getToken(service2));
}
@SuppressWarnings("unchecked") // from Mockito mocks
@Test
public <T extends TokenIdentifier> void testUGITokens() throws Exception {
@ -221,7 +294,9 @@ public <T extends TokenIdentifier> void testUGITokens() throws Exception {
UserGroupInformation.createUserForTesting("TheDoctor",
new String [] { "TheTARDIS"});
Token<T> t1 = mock(Token.class);
when(t1.getService()).thenReturn(new Text("t1"));
Token<T> t2 = mock(Token.class);
when(t2.getService()).thenReturn(new Text("t2"));
ugi.addToken(t1);
ugi.addToken(t2);

21
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/util/TestDataChecksum.java
View File

@ -35,13 +35,13 @@ public class TestDataChecksum {
private static final int DATA_TRAILER_IN_BUFFER = 3;
private static final int BYTES_PER_CHUNK = 512;
private static final int CHECKSUM_TYPES[] = new int[] {
DataChecksum.CHECKSUM_CRC32, DataChecksum.CHECKSUM_CRC32C
private static final DataChecksum.Type CHECKSUM_TYPES[] = {
DataChecksum.Type.CRC32, DataChecksum.Type.CRC32C
};
@Test
public void testBulkOps() throws Exception {
for (int type : CHECKSUM_TYPES) {
for (DataChecksum.Type type : CHECKSUM_TYPES) {
System.err.println(
"---- beginning tests with checksum type " + type + "----");
DataChecksum checksum = DataChecksum.newDataChecksum(
@ -118,21 +118,20 @@ private void doBulkTest(DataChecksum checksum, int dataLength,
@Test
public void testEquality() {
assertEquals(
DataChecksum.newDataChecksum(DataChecksum.CHECKSUM_CRC32, 512),
DataChecksum.newDataChecksum(DataChecksum.CHECKSUM_CRC32, 512));
DataChecksum.newDataChecksum(DataChecksum.Type.CRC32, 512),
DataChecksum.newDataChecksum(DataChecksum.Type.CRC32, 512));
assertFalse(
DataChecksum.newDataChecksum(DataChecksum.CHECKSUM_CRC32, 512).equals(
DataChecksum.newDataChecksum(DataChecksum.CHECKSUM_CRC32, 1024)));
DataChecksum.newDataChecksum(DataChecksum.Type.CRC32, 512).equals(
DataChecksum.newDataChecksum(DataChecksum.Type.CRC32, 1024)));
assertFalse(
DataChecksum.newDataChecksum(DataChecksum.CHECKSUM_CRC32, 512).equals(
DataChecksum.newDataChecksum(DataChecksum.CHECKSUM_CRC32C, 512)));
DataChecksum.newDataChecksum(DataChecksum.Type.CRC32, 512).equals(
DataChecksum.newDataChecksum(DataChecksum.Type.CRC32C, 512)));
}
@Test
public void testToString() {
assertEquals("DataChecksum(type=CRC32, chunkSize=512)",
DataChecksum.newDataChecksum(DataChecksum.CHECKSUM_CRC32, 512)
.toString());
DataChecksum.newDataChecksum(DataChecksum.Type.CRC32, 512).toString());
}
private static void corruptBufferOffset(ByteBuffer buf, int offset) {

141
hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/util/TestLineReader.java Normal file
View File

@ -0,0 +1,141 @@
/**
* 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.util;
import java.io.ByteArrayInputStream;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.util.LineReader;
import org.junit.Test;
import junit.framework.Assert;
public class TestLineReader {
private LineReader lineReader;
private String TestData;
private String Delimiter;
private Text line;
@Test
public void testCustomDelimiter() throws Exception {
/* TEST_1
* The test scenario is the tail of the buffer
* equals the starting character/s of delimiter
*
* The Test Data is such that,
*
* 1) we will have "</entity>" as delimiter
*
* 2) The tail of the current buffer would be "</"
* which matches with the starting character sequence of delimiter.
*
* 3) The Head of the next buffer would be "id>"
* which does NOT match with the remaining characters of delimiter.
*
* 4) Input data would be prefixed by char 'a'
* about numberOfCharToFillTheBuffer times.
* So that, one iteration to buffer the input data,
* would end at '</' ie equals starting 2 char of delimiter
*
* 5) For this we would take BufferSize as 64 * 1024;
*
* Check Condition
* In the second key value pair, the value should contain
* "</" from currentToken and
* "id>" from next token
*/
Delimiter="</entity>";
String CurrentBufferTailToken=
"</entity><entity><id>Gelesh</";
// Ending part of Input Data Buffer
// It contains '</' ie delimiter character
String NextBufferHeadToken=
"id><name>Omathil</name></entity>";
// Supposing the start of next buffer is this
String Expected =
(CurrentBufferTailToken+NextBufferHeadToken)
.replace(Delimiter, "");
// Expected ,must capture from both the buffer, excluding Delimiter
String TestPartOfInput = CurrentBufferTailToken+NextBufferHeadToken;
int BufferSize=64 * 1024;
int numberOfCharToFillTheBuffer=BufferSize-CurrentBufferTailToken.length();
StringBuilder fillerString=new StringBuilder();
for (int i=0;i<numberOfCharToFillTheBuffer;i++) {
fillerString.append('a'); // char 'a' as a filler for the test string
}
TestData = fillerString + TestPartOfInput;
lineReader = new LineReader(
new ByteArrayInputStream(TestData.getBytes()),Delimiter.getBytes());
line = new Text();
lineReader.readLine(line);
Assert.assertEquals(fillerString.toString(),line.toString());
lineReader.readLine(line);
Assert.assertEquals(Expected, line.toString());
/*TEST_2
* The test scenario is such that,
* the character/s preceding the delimiter,
* equals the starting character/s of delimiter
*/
Delimiter = "record";
StringBuilder TestStringBuilder = new StringBuilder();
TestStringBuilder.append(Delimiter+"Kerala ");
TestStringBuilder.append(Delimiter+"Bangalore");
TestStringBuilder.append(Delimiter+" North Korea");
TestStringBuilder.append(Delimiter+Delimiter+
"Guantanamo");
TestStringBuilder.append(Delimiter+"ecord"+"recor"+"core"); //~EOF with 're'
TestData=TestStringBuilder.toString();
lineReader = new LineReader(
new ByteArrayInputStream(TestData.getBytes()),Delimiter.getBytes());
lineReader.readLine(line);
Assert.assertEquals("",line.toString());
lineReader.readLine(line);
Assert.assertEquals("Kerala ",line.toString());
lineReader.readLine(line);
Assert.assertEquals("Bangalore",line.toString());
lineReader.readLine(line);
Assert.assertEquals(" North Korea",line.toString());
lineReader.readLine(line);
Assert.assertEquals("",line.toString());
lineReader.readLine(line);
Assert.assertEquals("Guantanamo",line.toString());
lineReader.readLine(line);
Assert.assertEquals(("ecord"+"recor"+"core"),line.toString());
}
}

30
hadoop-hdfs-project/hadoop-hdfs-httpfs/pom.xml
View File

@ -60,8 +60,13 @@
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-annotations</artifactId>
<scope>provided</scope>
<artifactId>hadoop-auth</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>com.sun.jersey</groupId>
<artifactId>jersey-core</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>com.sun.jersey</groupId>
@ -74,18 +79,8 @@
<scope>provided</scope>
</dependency>
<dependency>
<groupId>javax.servlet.jsp</groupId>
<artifactId>jsp-api</artifactId>
<scope>provided</scope>
</dependency>
<dependency>
<groupId>commons-codec</groupId>
<artifactId>commons-codec</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.jdom</groupId>
<artifactId>jdom</artifactId>
<groupId>com.google.guava</groupId>
<artifactId>guava</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
@ -93,6 +88,11 @@
<artifactId>json-simple</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.mortbay.jetty</groupId>
<artifactId>jetty</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-common</artifactId>
@ -248,7 +248,7 @@
<dependency>
<groupId>org.slf4j</groupId>
<artifactId>slf4j-log4j12</artifactId>
<scope>compile</scope>
<scope>runtime</scope>
</dependency>
</dependencies>

14
hadoop-hdfs-project/hadoop-hdfs-httpfs/src/main/java/org/apache/hadoop/fs/http/client/HttpFSFileSystem.java
View File

@ -863,7 +863,6 @@ public void readFields(DataInput in) throws IOException {
@Override
@SuppressWarnings("deprecation")
public Token<?> getDelegationToken(final String renewer)
throws IOException {
return doAsRealUserIfNecessary(new Callable<Token<?>>() {
@ -875,19 +874,6 @@ public Token<?> call() throws Exception {
});
}
@Override
public List<Token<?>> getDelegationTokens(final String renewer)
throws IOException {
return doAsRealUserIfNecessary(new Callable<List<Token<?>>>() {
@Override
public List<Token<?>> call() throws Exception {
return HttpFSKerberosAuthenticator.
getDelegationTokens(uri, httpFSAddr, authToken, renewer);
}
});
}
public long renewDelegationToken(final Token<?> token) throws IOException {
return doAsRealUserIfNecessary(new Callable<Long>() {
@Override

65
hadoop-hdfs-project/hadoop-hdfs-httpfs/src/main/java/org/apache/hadoop/fs/http/client/HttpFSKerberosAuthenticator.java
View File

@ -66,7 +66,6 @@ protected Authenticator getFallBackAuthenticator() {
public static final String RENEWER_PARAM = "renewer";
public static final String TOKEN_KIND = "HTTPFS_DELEGATION_TOKEN";
public static final String DELEGATION_TOKEN_JSON = "Token";
public static final String DELEGATION_TOKENS_JSON = "Tokens";
public static final String DELEGATION_TOKEN_URL_STRING_JSON = "urlString";
public static final String RENEW_DELEGATION_TOKEN_JSON = "long";
@ -76,7 +75,6 @@ protected Authenticator getFallBackAuthenticator() {
@InterfaceAudience.Private
public static enum DelegationTokenOperation {
GETDELEGATIONTOKEN(HTTP_GET, true),
GETDELEGATIONTOKENS(HTTP_GET, true),
RENEWDELEGATIONTOKEN(HTTP_PUT, true),
CANCELDELEGATIONTOKEN(HTTP_PUT, false);
@ -121,10 +119,11 @@ public void authenticate(URL url, AuthenticatedURL.Token token)
public static final String OP_PARAM = "op";
private static List<Token<?>> getDelegationTokens(URI fsURI,
InetSocketAddress httpFSAddr, DelegationTokenOperation op,
AuthenticatedURL.Token token, String renewer)
throws IOException {
public static Token<?> getDelegationToken(URI fsURI,
InetSocketAddress httpFSAddr, AuthenticatedURL.Token token,
String renewer) throws IOException {
DelegationTokenOperation op =
DelegationTokenOperation.GETDELEGATIONTOKEN;
Map<String, String> params = new HashMap<String, String>();
params.put(OP_PARAM, op.toString());
params.put(RENEWER_PARAM,renewer);
@ -135,56 +134,20 @@ private static List<Token<?>> getDelegationTokens(URI fsURI,
HttpURLConnection conn = aUrl.openConnection(url, token);
conn.setRequestMethod(op.getHttpMethod());
HttpFSUtils.validateResponse(conn, HttpURLConnection.HTTP_OK);
List<String> list = new ArrayList<String>();
if (op == DelegationTokenOperation.GETDELEGATIONTOKEN) {
JSONObject json = (JSONObject) ((JSONObject)
HttpFSUtils.jsonParse(conn)).get(DELEGATION_TOKEN_JSON);
String tokenStr = (String)
json.get(DELEGATION_TOKEN_URL_STRING_JSON);
list.add(tokenStr);
}
else if (op == DelegationTokenOperation.GETDELEGATIONTOKENS) {
JSONObject json = (JSONObject) ((JSONObject)
HttpFSUtils.jsonParse(conn)).get(DELEGATION_TOKENS_JSON);
JSONArray array = (JSONArray) json.get(DELEGATION_TOKEN_JSON);
for (Object element : array) {
String tokenStr = (String)
((Map) element).get(DELEGATION_TOKEN_URL_STRING_JSON);
list.add(tokenStr);
}
} else {
throw new IllegalArgumentException("Invalid operation: " +
op.toString());
}
List<Token<?>> dTokens = new ArrayList<Token<?>>();
for (String tokenStr : list) {
Token<AbstractDelegationTokenIdentifier> dToken =
new Token<AbstractDelegationTokenIdentifier>();
dToken.decodeFromUrlString(tokenStr);
dTokens.add(dToken);
SecurityUtil.setTokenService(dToken, httpFSAddr);
}
return dTokens;
JSONObject json = (JSONObject) ((JSONObject)
HttpFSUtils.jsonParse(conn)).get(DELEGATION_TOKEN_JSON);
String tokenStr = (String)
json.get(DELEGATION_TOKEN_URL_STRING_JSON);
Token<AbstractDelegationTokenIdentifier> dToken =
new Token<AbstractDelegationTokenIdentifier>();
dToken.decodeFromUrlString(tokenStr);
SecurityUtil.setTokenService(dToken, httpFSAddr);
return dToken;
} catch (AuthenticationException ex) {
throw new IOException(ex.toString(), ex);
}
}
public static List<Token<?>> getDelegationTokens(URI fsURI,
InetSocketAddress httpFSAddr, AuthenticatedURL.Token token,
String renewer) throws IOException {
return getDelegationTokens(fsURI, httpFSAddr,
DelegationTokenOperation.GETDELEGATIONTOKENS, token, renewer);
}
public static Token<?> getDelegationToken(URI fsURI,
InetSocketAddress httpFSAddr, AuthenticatedURL.Token token,
String renewer) throws IOException {
return getDelegationTokens(fsURI, httpFSAddr,
DelegationTokenOperation.GETDELEGATIONTOKENS, token, renewer).get(0);
}
public static long renewDelegationToken(URI fsURI,
AuthenticatedURL.Token token, Token<?> dToken) throws IOException {
Map<String, String> params = new HashMap<String, String>();

26
hadoop-hdfs-project/hadoop-hdfs-httpfs/src/main/java/org/apache/hadoop/fs/http/server/HttpFSKerberosAuthenticationHandler.java
View File

@ -63,8 +63,6 @@ public class HttpFSKerberosAuthenticationHandler
static {
DELEGATION_TOKEN_OPS.add(
DelegationTokenOperation.GETDELEGATIONTOKEN.toString());
DELEGATION_TOKEN_OPS.add(
DelegationTokenOperation.GETDELEGATIONTOKENS.toString());
DELEGATION_TOKEN_OPS.add(
DelegationTokenOperation.RENEWDELEGATIONTOKEN.toString());
DELEGATION_TOKEN_OPS.add(
@ -111,7 +109,6 @@ public boolean managementOperation(AuthenticationToken token,
Map map = null;
switch (dtOp) {
case GETDELEGATIONTOKEN:
case GETDELEGATIONTOKENS:
String renewerParam =
request.getParameter(HttpFSKerberosAuthenticator.RENEWER_PARAM);
if (renewerParam == null) {
@ -119,11 +116,7 @@ public boolean managementOperation(AuthenticationToken token,
}
Token<?> dToken = tokenManager.createToken(
UserGroupInformation.getCurrentUser(), renewerParam);
if (dtOp == DelegationTokenOperation.GETDELEGATIONTOKEN) {
map = delegationTokenToJSON(dToken);
} else {
map = delegationTokensToJSON(Arrays.asList((Token)dToken));
}
map = delegationTokenToJSON(dToken);
break;
case RENEWDELEGATIONTOKEN:
case CANCELDELEGATIONTOKEN:
@ -191,23 +184,6 @@ private static Map delegationTokenToJSON(Token token) throws IOException {
return response;
}
@SuppressWarnings("unchecked")
private static Map delegationTokensToJSON(List<Token> tokens)
throws IOException {
List list = new ArrayList();
for (Token token : tokens) {
Map map = new HashMap();
map.put(HttpFSKerberosAuthenticator.DELEGATION_TOKEN_URL_STRING_JSON,
token.encodeToUrlString());
list.add(map);
}
Map map = new HashMap();
map.put(HttpFSKerberosAuthenticator.DELEGATION_TOKEN_JSON, list);
Map response = new LinkedHashMap();
response.put(HttpFSKerberosAuthenticator.DELEGATION_TOKENS_JSON, map);
return response;
}
/**
* Authenticates a request looking for the <code>delegation</code>
* query-string parameter and verifying it is a valid token. If there is not

27
hadoop-hdfs-project/hadoop-hdfs-httpfs/src/test/java/org/apache/hadoop/fs/http/server/TestHttpFSKerberosAuthenticationHandler.java
View File

@ -68,10 +68,8 @@ public void testManagementOperations() throws Exception {
testNonManagementOperation(handler);
testManagementOperationErrors(handler);
testGetToken(handler, false, null);
testGetToken(handler, true, null);
testGetToken(handler, false, "foo");
testGetToken(handler, true, "foo");
testGetToken(handler, null);
testGetToken(handler, "foo");
testCancelToken(handler);
testRenewToken(handler);
@ -115,12 +113,9 @@ private void testManagementOperationErrors(AuthenticationHandler handler)
Mockito.contains("requires SPNEGO"));
}
private void testGetToken(AuthenticationHandler handler, boolean tokens,
String renewer)
private void testGetToken(AuthenticationHandler handler, String renewer)
throws Exception {
DelegationTokenOperation op =
(tokens) ? DelegationTokenOperation.GETDELEGATIONTOKENS
: DelegationTokenOperation.GETDELEGATIONTOKEN;
DelegationTokenOperation op = DelegationTokenOperation.GETDELEGATIONTOKEN;
HttpServletRequest request = Mockito.mock(HttpServletRequest.class);
HttpServletResponse response = Mockito.mock(HttpServletResponse.class);
Mockito.when(request.getParameter(HttpFSFileSystem.OP_PARAM)).
@ -148,23 +143,13 @@ private void testGetToken(AuthenticationHandler handler, boolean tokens,
Mockito.verify(response).setContentType(MediaType.APPLICATION_JSON);
pwriter.close();
String responseOutput = writer.toString();
String tokenLabel = (tokens)
? HttpFSKerberosAuthenticator.DELEGATION_TOKENS_JSON
: HttpFSKerberosAuthenticator.DELEGATION_TOKEN_JSON;
if (tokens) {
Assert.assertTrue(responseOutput.contains(tokenLabel));
} else {
Assert.assertTrue(responseOutput.contains(tokenLabel));
}
String tokenLabel = HttpFSKerberosAuthenticator.DELEGATION_TOKEN_JSON;
Assert.assertTrue(responseOutput.contains(tokenLabel));
Assert.assertTrue(responseOutput.contains(
HttpFSKerberosAuthenticator.DELEGATION_TOKEN_URL_STRING_JSON));
JSONObject json = (JSONObject) new JSONParser().parse(responseOutput);
json = (JSONObject) json.get(tokenLabel);
String tokenStr;
if (tokens) {
json = (JSONObject) ((JSONArray)
json.get(HttpFSKerberosAuthenticator.DELEGATION_TOKEN_JSON)).get(0);
}
tokenStr = (String)
json.get(HttpFSKerberosAuthenticator.DELEGATION_TOKEN_URL_STRING_JSON);
Token<DelegationTokenIdentifier> dt = new Token<DelegationTokenIdentifier>();

5
hadoop-hdfs-project/hadoop-hdfs-httpfs/src/test/java/org/apache/hadoop/fs/http/server/TestHttpFSWithKerberos.java
View File

@ -222,10 +222,11 @@ private void testDelegationTokenWithFS(Class fileSystemClass)
URI uri = new URI( "webhdfs://" +
TestJettyHelper.getJettyURL().toURI().getAuthority());
FileSystem fs = FileSystem.get(uri, conf);
Token<?> token = fs.getDelegationToken("foo");
Token<?> tokens[] = fs.addDelegationTokens("foo", null);
fs.close();
Assert.assertEquals(1, tokens.length);
fs = FileSystem.get(uri, conf);
((DelegationTokenRenewer.Renewable) fs).setDelegationToken(token);
((DelegationTokenRenewer.Renewable) fs).setDelegationToken(tokens[0]);
fs.listStatus(new Path("/"));
fs.close();
}

59
hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
View File

@ -120,6 +120,12 @@ Trunk (unreleased changes)
HDFS-3803. Change BlockPoolSliceScanner chatty INFO log to DEBUG.
(Andrew Purtell via suresh)
HDFS-3817. Avoid printing SafeModeException stack trace.
(Brandon Li via suresh)
HDFS-3819. Should check whether invalidate work percentage default value is
not greater than 1.0f. (Jing Zhao via jitendra)
OPTIMIZATIONS
BUG FIXES
@ -191,6 +197,14 @@ Trunk (unreleased changes)
HDFS-3792. Fix two findbugs introduced by HDFS-3695 (todd)
HDFS-3827. TestHASafeMode#assertSafemode method should be made static.
(Jing Zhao via suresh)
HDFS-3834. Remove unused static fields NAME, DESCRIPTION and Usage from
Command. (Jing Zhao via suresh)
HDFS-3678. Edit log files are never being purged from 2NN. (atm)
Branch-2 ( Unreleased changes )
INCOMPATIBLE CHANGES
@ -399,6 +413,22 @@ Branch-2 ( Unreleased changes )
HDFS-3796. Speed up edit log tests by avoiding fsync() (todd)
HDFS-2963. Console Output is confusing while executing metasave
(dfsadmin command). (Andrew Wang via eli)
HDFS-3672. Expose disk-location information for blocks to enable better
scheduling. (Andrew Wang via atm)
HDFS-2727. libhdfs should get the default block size from the server.
(Colin Patrick McCabe via eli)
HDFS-2686. Remove DistributedUpgrade related code. (suresh)
HDFS-3832. Remove protocol methods related to DistributedUpgrade. (suresh)
HDFS-3177. Update DFSClient and DataXceiver to handle different checkum
types in file checksum computation. (Kihwal Lee via szetszwo)
OPTIMIZATIONS
HDFS-2982. Startup performance suffers when there are many edit log
@ -412,6 +442,9 @@ Branch-2 ( Unreleased changes )
HDFS-3697. Enable fadvise readahead by default. (todd)
HDFS-2421. Improve the concurrency of SerialNumberMap in NameNode.
(Jing Zhao and Weiyan Wang via szetszwo)
BUG FIXES
HDFS-3385. The last block of INodeFileUnderConstruction is not
@ -614,6 +647,32 @@ Branch-2 ( Unreleased changes )
header when offset is specified and length is omitted.
(Ravi Prakash via szetszwo)
HDFS-3048. Small race in BlockManager#close. (Andy Isaacson via eli)
HDFS-3194. DataNode block scanner is running too frequently.
(Andy Isaacson via eli)
HDFS-3808. fuse_dfs: postpone libhdfs intialization until after fork.
(Colin Patrick McCabe via atm)
HDFS-3788. ByteRangeInputStream should not expect HTTP Content-Length header
when chunked transfer-encoding is used. (szetszwo)
HDFS-3816. Invalidate work percentage default value should be 0.32f
instead of 32. (Jing Zhao via suresh)
HDFS-3707. TestFSInputChecker: improper use of skip.
(Colin Patrick McCabe via eli)
HDFS-3830. test_libhdfs_threaded: use forceNewInstance.
(Colin Patrick McCabe via eli)
HDFS-3835. Long-lived 2NN cannot perform a checkpoint if security is
enabled and the NN restarts with outstanding delegation tokens. (atm)
HDFS-3715. Fix TestFileCreation#testFileCreationNamenodeRestart.
(Andrew Whang via eli)
BREAKDOWN OF HDFS-3042 SUBTASKS
HDFS-2185. HDFS portion of ZK-based FailoverController (todd)

108
hadoop-hdfs-project/hadoop-hdfs/pom.xml
View File

@ -39,18 +39,13 @@ http://maven.apache.org/xsd/maven-4.0.0.xsd">
<dependencies>
<dependency>
<groupId>org.aspectj</groupId>
<artifactId>aspectjtools</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.aspectj</groupId>
<artifactId>aspectjrt</artifactId>
<scope>test</scope>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-annotations</artifactId>
<scope>provided</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-annotations</artifactId>
<artifactId>hadoop-auth</artifactId>
<scope>provided</scope>
</dependency>
<dependency>
@ -64,6 +59,58 @@ http://maven.apache.org/xsd/maven-4.0.0.xsd">
<scope>test</scope>
<type>test-jar</type>
</dependency>
<dependency>
<groupId>org.apache.zookeeper</groupId>
<artifactId>zookeeper</artifactId>
<version>3.4.2</version>
<type>test-jar</type>
<scope>test</scope>
</dependency>
<dependency>
<groupId>com.google.guava</groupId>
<artifactId>guava</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.mortbay.jetty</groupId>
<artifactId>jetty</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.mortbay.jetty</groupId>
<artifactId>jetty-util</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>com.sun.jersey</groupId>
<artifactId>jersey-core</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>com.sun.jersey</groupId>
<artifactId>jersey-server</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-cli</groupId>
<artifactId>commons-cli</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-codec</groupId>
<artifactId>commons-codec</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-io</groupId>
<artifactId>commons-io</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-lang</groupId>
<artifactId>commons-lang</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>commons-logging</groupId>
<artifactId>commons-logging</artifactId>
@ -74,6 +121,11 @@ http://maven.apache.org/xsd/maven-4.0.0.xsd">
<artifactId>commons-daemon</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>javax.servlet.jsp</groupId>
<artifactId>jsp-api</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>log4j</groupId>
<artifactId>log4j</artifactId>
@ -85,8 +137,8 @@ http://maven.apache.org/xsd/maven-4.0.0.xsd">
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.apache.avro</groupId>
<artifactId>avro</artifactId>
<groupId>javax.servlet</groupId>
<artifactId>servlet-api</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
@ -99,28 +151,30 @@ http://maven.apache.org/xsd/maven-4.0.0.xsd">
<artifactId>mockito-all</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.ant</groupId>
<artifactId>ant</artifactId>
<scope>provided</scope>
</dependency>
<dependency>
<groupId>org.apache.zookeeper</groupId>
<artifactId>zookeeper</artifactId>
<version>3.4.2</version>
<scope>provided</scope>
</dependency>
<dependency>
<groupId>org.slf4j</groupId>
<artifactId>slf4j-log4j12</artifactId>
<scope>provided</scope>
</dependency>
<dependency>
<groupId>org.apache.zookeeper</groupId>
<artifactId>zookeeper</artifactId>
<version>3.4.2</version>
<type>test-jar</type>
<scope>test</scope>
<groupId>org.codehaus.jackson</groupId>
<artifactId>jackson-core-asl</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.codehaus.jackson</groupId>
<artifactId>jackson-mapper-asl</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>tomcat</groupId>
<artifactId>jasper-runtime</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>xmlenc</groupId>
<artifactId>xmlenc</artifactId>
<scope>compile</scope>
</dependency>
</dependencies>

16
hadoop-hdfs-project/hadoop-hdfs/src/contrib/bkjournal/pom.xml
View File

@ -37,9 +37,9 @@ http://maven.apache.org/xsd/maven-4.0.0.xsd">
<dependencies>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-annotations</artifactId>
<scope>provided</scope>
<groupId>commons-logging</groupId>
<artifactId>commons-logging</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
@ -68,6 +68,16 @@ http://maven.apache.org/xsd/maven-4.0.0.xsd">
<artifactId>bookkeeper-server</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.apache.zookeeper</groupId>
<artifactId>zookeeper</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>com.google.guava</groupId>
<artifactId>guava</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>

51
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/fs/BlockStorageLocation.java Normal file
View File

@ -0,0 +1,51 @@
/**
* 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 java.io.IOException;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
/**
* Wrapper for {@link BlockLocation} that also adds {@link VolumeId} volume
* location information for each replica.
*/
@InterfaceStability.Unstable
@InterfaceAudience.Public
public class BlockStorageLocation extends BlockLocation {
private final VolumeId[] volumeIds;
public BlockStorageLocation(BlockLocation loc, VolumeId[] volumeIds)
throws IOException {
// Initialize with data from passed in BlockLocation
super(loc.getNames(), loc.getHosts(), loc.getTopologyPaths(), loc
.getOffset(), loc.getLength(), loc.isCorrupt());
this.volumeIds = volumeIds;
}
/**
* Gets the list of {@link VolumeId} corresponding to the block's replicas.
*
* @return volumeIds list of VolumeId for the block's replicas
*/
public VolumeId[] getVolumeIds() {
return volumeIds;
}
}

5
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/fs/Hdfs.java
View File

@ -31,6 +31,7 @@
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.hdfs.CorruptFileBlockIterator;
import org.apache.hadoop.hdfs.DFSClient;
import org.apache.hadoop.hdfs.DFSUtil;
@ -93,10 +94,10 @@ public int getUriDefaultPort() {
public HdfsDataOutputStream createInternal(Path f,
EnumSet<CreateFlag> createFlag, FsPermission absolutePermission,
int bufferSize, short replication, long blockSize, Progressable progress,
int bytesPerChecksum, boolean createParent) throws IOException {
ChecksumOpt checksumOpt, boolean createParent) throws IOException {
return new HdfsDataOutputStream(dfs.primitiveCreate(getUriPath(f),
absolutePermission, createFlag, createParent, replication, blockSize,
progress, bufferSize, bytesPerChecksum), getStatistics());
progress, bufferSize, checksumOpt), getStatistics());
}
@Override

48
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/fs/HdfsBlockLocation.java Normal file
View File

@ -0,0 +1,48 @@
/**
* 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 java.io.IOException;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.hdfs.protocol.LocatedBlock;
/**
* Wrapper for {@link BlockLocation} that also includes a {@link LocatedBlock},
* allowing more detailed queries to the datanode about a block.
*
*/
@InterfaceAudience.Private
@InterfaceStability.Unstable
public class HdfsBlockLocation extends BlockLocation {
private final LocatedBlock block;
public HdfsBlockLocation(BlockLocation loc, LocatedBlock block)
throws IOException {
// Initialize with data from passed in BlockLocation
super(loc.getNames(), loc.getHosts(), loc.getTopologyPaths(),
loc.getOffset(), loc.getLength(), loc.isCorrupt());
this.block = block;
}
public LocatedBlock getLocatedBlock() {
return block;
}
}

74
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/fs/HdfsVolumeId.java Normal file
View File

@ -0,0 +1,74 @@
/**
* 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.commons.lang.builder.EqualsBuilder;
import org.apache.commons.lang.builder.HashCodeBuilder;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
/**
* HDFS-specific volume identifier which implements {@link VolumeId}. Can be
* used to differentiate between the data directories on a single datanode. This
* identifier is only unique on a per-datanode basis.
*/
@InterfaceStability.Unstable
@InterfaceAudience.Public
public class HdfsVolumeId implements VolumeId {
private final byte id;
private final boolean isValid;
public HdfsVolumeId(byte id, boolean isValid) {
this.id = id;
this.isValid = isValid;
}
@Override
public boolean isValid() {
return isValid;
}
@Override
public int compareTo(VolumeId arg0) {
return hashCode() - arg0.hashCode();
}
@Override
public int hashCode() {
return new HashCodeBuilder().append(id).toHashCode();
}
@Override
public boolean equals(Object obj) {
if (obj == null || obj.getClass() != getClass()) {
return false;
}
if (obj == this) {
return true;
}
HdfsVolumeId that = (HdfsVolumeId) obj;
return new EqualsBuilder().append(this.id, that.id).isEquals();
}
@Override
public String toString() {
return Byte.toString(id);
}
}

49
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/fs/VolumeId.java Normal file
View File

@ -0,0 +1,49 @@
/**
* 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;
/**
* Opaque interface that identifies a disk location. Subclasses
* should implement {@link Comparable} and override both equals and hashCode.
*/
@InterfaceStability.Unstable
@InterfaceAudience.Public
public interface VolumeId extends Comparable<VolumeId> {
/**
* Indicates if the disk identifier is valid. Invalid identifiers indicate
* that the block was not present, or the location could otherwise not be
* determined.
*
* @return true if the disk identifier is valid
*/
public boolean isValid();
@Override
abstract public int compareTo(VolumeId arg0);
@Override
abstract public int hashCode();
@Override
abstract public boolean equals(Object obj);
}

2
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/BlockReaderLocal.java
View File

@ -290,7 +290,7 @@ private BlockReaderLocal(Configuration conf, String hdfsfile,
long length, BlockLocalPathInfo pathinfo, FileInputStream dataIn)
throws IOException {
this(conf, hdfsfile, block, token, startOffset, length, pathinfo,
DataChecksum.newDataChecksum(DataChecksum.CHECKSUM_NULL, 4), false,
DataChecksum.newDataChecksum(DataChecksum.Type.NULL, 4), false,
dataIn, startOffset, null);
}

337
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/BlockStorageLocationUtil.java Normal file
View File

@ -0,0 +1,337 @@
/**
* 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;
import java.io.IOException;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
import java.util.concurrent.Callable;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Future;
import java.util.concurrent.ScheduledThreadPoolExecutor;
import java.util.concurrent.TimeUnit;
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.conf.Configuration;
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.BlockStorageLocation;
import org.apache.hadoop.fs.HdfsVolumeId;
import org.apache.hadoop.fs.VolumeId;
import org.apache.hadoop.hdfs.protocol.ClientDatanodeProtocol;
import org.apache.hadoop.hdfs.protocol.DatanodeInfo;
import org.apache.hadoop.hdfs.protocol.ExtendedBlock;
import org.apache.hadoop.hdfs.protocol.HdfsBlocksMetadata;
import org.apache.hadoop.hdfs.protocol.LocatedBlock;
import org.apache.hadoop.hdfs.security.token.block.BlockTokenIdentifier;
import org.apache.hadoop.hdfs.security.token.block.InvalidBlockTokenException;
import org.apache.hadoop.ipc.RPC;
import org.apache.hadoop.security.token.Token;
@InterfaceAudience.Private
@InterfaceStability.Unstable
class BlockStorageLocationUtil {
private static final Log LOG = LogFactory
.getLog(BlockStorageLocationUtil.class);
/**
* Create a list of {@link VolumeBlockLocationCallable} corresponding to a set
* of datanodes and blocks.
*
* @param datanodeBlocks
* Map of datanodes to block replicas at each datanode
* @return callables Used to query each datanode for location information on
* the block replicas at the datanode
*/
private static List<VolumeBlockLocationCallable> createVolumeBlockLocationCallables(
Configuration conf, Map<DatanodeInfo, List<LocatedBlock>> datanodeBlocks,
int timeout, boolean connectToDnViaHostname) {
// Construct the callables, one per datanode
List<VolumeBlockLocationCallable> callables =
new ArrayList<VolumeBlockLocationCallable>();
for (Map.Entry<DatanodeInfo, List<LocatedBlock>> entry : datanodeBlocks
.entrySet()) {
// Construct RPC parameters
DatanodeInfo datanode = entry.getKey();
List<LocatedBlock> locatedBlocks = entry.getValue();
List<ExtendedBlock> extendedBlocks =
new ArrayList<ExtendedBlock>(locatedBlocks.size());
List<Token<BlockTokenIdentifier>> dnTokens =
new ArrayList<Token<BlockTokenIdentifier>>(
locatedBlocks.size());
for (LocatedBlock b : locatedBlocks) {
extendedBlocks.add(b.getBlock());
dnTokens.add(b.getBlockToken());
}
VolumeBlockLocationCallable callable = new VolumeBlockLocationCallable(
conf, datanode, extendedBlocks, dnTokens, timeout,
connectToDnViaHostname);
callables.add(callable);
}
return callables;
}
/**
* Queries datanodes for the blocks specified in <code>datanodeBlocks</code>,
* making one RPC to each datanode. These RPCs are made in parallel using a
* threadpool.
*
* @param datanodeBlocks
* Map of datanodes to the blocks present on the DN
* @return metadatas List of block metadata for each datanode, specifying
* volume locations for each block
* @throws InvalidBlockTokenException
* if client does not have read access on a requested block
*/
static List<HdfsBlocksMetadata> queryDatanodesForHdfsBlocksMetadata(
Configuration conf, Map<DatanodeInfo, List<LocatedBlock>> datanodeBlocks,
int poolsize, int timeout, boolean connectToDnViaHostname)
throws InvalidBlockTokenException {
List<VolumeBlockLocationCallable> callables =
createVolumeBlockLocationCallables(conf, datanodeBlocks, timeout,
connectToDnViaHostname);
// Use a thread pool to execute the Callables in parallel
List<Future<HdfsBlocksMetadata>> futures =
new ArrayList<Future<HdfsBlocksMetadata>>();
ExecutorService executor = new ScheduledThreadPoolExecutor(poolsize);
try {
futures = executor.invokeAll(callables, timeout, TimeUnit.SECONDS);
} catch (InterruptedException e) {
// Swallow the exception here, because we can return partial results
}
executor.shutdown();
// Initialize metadatas list with nulls
// This is used to later indicate if we didn't get a response from a DN
List<HdfsBlocksMetadata> metadatas = new ArrayList<HdfsBlocksMetadata>();
for (int i = 0; i < futures.size(); i++) {
metadatas.add(null);
}
// Fill in metadatas with results from DN RPCs, where possible
for (int i = 0; i < futures.size(); i++) {
Future<HdfsBlocksMetadata> future = futures.get(i);
try {
HdfsBlocksMetadata metadata = future.get();
metadatas.set(i, metadata);
} catch (ExecutionException e) {
VolumeBlockLocationCallable callable = callables.get(i);
DatanodeInfo datanode = callable.getDatanodeInfo();
Throwable t = e.getCause();
if (t instanceof InvalidBlockTokenException) {
LOG.warn("Invalid access token when trying to retrieve "
+ "information from datanode " + datanode.getIpcAddr(false));
throw (InvalidBlockTokenException) t;
}
else if (t instanceof UnsupportedOperationException) {
LOG.info("Datanode " + datanode.getIpcAddr(false) + " does not support"
+ " required #getHdfsBlocksMetadata() API");
throw (UnsupportedOperationException) t;
} else {
LOG.info("Failed to connect to datanode " +
datanode.getIpcAddr(false));
}
if (LOG.isDebugEnabled()) {
LOG.debug("Could not fetch information from datanode", t);
}
} catch (InterruptedException e) {
// Shouldn't happen, because invokeAll waits for all Futures to be ready
LOG.info("Interrupted while fetching HdfsBlocksMetadata");
}
}
return metadatas;
}
/**
* Group the per-replica {@link VolumeId} info returned from
* {@link DFSClient#queryDatanodesForHdfsBlocksMetadata(Map)} to be associated
* with the corresponding {@link LocatedBlock}.
*
* @param blocks
* Original LocatedBlock array
* @param datanodeBlocks
* Mapping from datanodes to the list of replicas on each datanode
* @param metadatas
* VolumeId information for the replicas on each datanode
* @return blockVolumeIds per-replica VolumeId information associated with the
* parent LocatedBlock
*/
static Map<LocatedBlock, List<VolumeId>> associateVolumeIdsWithBlocks(
List<LocatedBlock> blocks, Map<DatanodeInfo,
List<LocatedBlock>> datanodeBlocks, List<HdfsBlocksMetadata> metadatas) {
// Initialize mapping of ExtendedBlock to LocatedBlock.
// Used to associate results from DN RPCs to the parent LocatedBlock
Map<ExtendedBlock, LocatedBlock> extBlockToLocBlock =
new HashMap<ExtendedBlock, LocatedBlock>();
for (LocatedBlock b : blocks) {
extBlockToLocBlock.put(b.getBlock(), b);
}
// Initialize the mapping of blocks -> list of VolumeIds, one per replica
// This is filled out with real values from the DN RPCs
Map<LocatedBlock, List<VolumeId>> blockVolumeIds =
new HashMap<LocatedBlock, List<VolumeId>>();
for (LocatedBlock b : blocks) {
ArrayList<VolumeId> l = new ArrayList<VolumeId>(b.getLocations().length);
// Start off all IDs as invalid, fill it in later with results from RPCs
for (int i = 0; i < b.getLocations().length; i++) {
l.add(new HdfsVolumeId((byte)-1, false));
}
blockVolumeIds.put(b, l);
}
// Iterate through the list of metadatas (one per datanode).
// For each metadata, if it's valid, insert its volume location information
// into the Map returned to the caller
Iterator<HdfsBlocksMetadata> metadatasIter = metadatas.iterator();
Iterator<DatanodeInfo> datanodeIter = datanodeBlocks.keySet().iterator();
while (metadatasIter.hasNext()) {
HdfsBlocksMetadata metadata = metadatasIter.next();
DatanodeInfo datanode = datanodeIter.next();
// Check if metadata is valid
if (metadata == null) {
continue;
}
ExtendedBlock[] metaBlocks = metadata.getBlocks();
List<byte[]> metaVolumeIds = metadata.getVolumeIds();
List<Integer> metaVolumeIndexes = metadata.getVolumeIndexes();
// Add VolumeId for each replica in the HdfsBlocksMetadata
for (int j = 0; j < metaBlocks.length; j++) {
int volumeIndex = metaVolumeIndexes.get(j);
ExtendedBlock extBlock = metaBlocks[j];
// Skip if block wasn't found, or not a valid index into metaVolumeIds
// Also skip if the DN responded with a block we didn't ask for
if (volumeIndex == Integer.MAX_VALUE
|| volumeIndex >= metaVolumeIds.size()
|| !extBlockToLocBlock.containsKey(extBlock)) {
continue;
}
// Get the VolumeId by indexing into the list of VolumeIds
// provided by the datanode
HdfsVolumeId id = new HdfsVolumeId(metaVolumeIds.get(volumeIndex)[0],
true);
// Find out which index we are in the LocatedBlock's replicas
LocatedBlock locBlock = extBlockToLocBlock.get(extBlock);
DatanodeInfo[] dnInfos = locBlock.getLocations();
int index = -1;
for (int k = 0; k < dnInfos.length; k++) {
if (dnInfos[k].equals(datanode)) {
index = k;
break;
}
}
if (index < 0) {
if (LOG.isDebugEnabled()) {
LOG.debug("Datanode responded with a block volume id we did" +
" not request, omitting.");
}
continue;
}
// Place VolumeId at the same index as the DN's index in the list of
// replicas
List<VolumeId> VolumeIds = blockVolumeIds.get(locBlock);
VolumeIds.set(index, id);
}
}
return blockVolumeIds;
}
/**
* Helper method to combine a list of {@link LocatedBlock} with associated
* {@link VolumeId} information to form a list of {@link BlockStorageLocation}
* .
*/
static BlockStorageLocation[] convertToVolumeBlockLocations(
List<LocatedBlock> blocks,
Map<LocatedBlock, List<VolumeId>> blockVolumeIds) throws IOException {
// Construct the final return value of VolumeBlockLocation[]
BlockLocation[] locations = DFSUtil.locatedBlocks2Locations(blocks);
List<BlockStorageLocation> volumeBlockLocs =
new ArrayList<BlockStorageLocation>(locations.length);
for (int i = 0; i < locations.length; i++) {
LocatedBlock locBlock = blocks.get(i);
List<VolumeId> volumeIds = blockVolumeIds.get(locBlock);
BlockStorageLocation bsLoc = new BlockStorageLocation(locations[i],
volumeIds.toArray(new VolumeId[0]));
volumeBlockLocs.add(bsLoc);
}
return volumeBlockLocs.toArray(new BlockStorageLocation[] {});
}
/**
* Callable that sets up an RPC proxy to a datanode and queries it for
* volume location information for a list of ExtendedBlocks.
*/
private static class VolumeBlockLocationCallable implements
Callable<HdfsBlocksMetadata> {
private Configuration configuration;
private int timeout;
private DatanodeInfo datanode;
private List<ExtendedBlock> extendedBlocks;
private List<Token<BlockTokenIdentifier>> dnTokens;
private boolean connectToDnViaHostname;
VolumeBlockLocationCallable(Configuration configuration,
DatanodeInfo datanode, List<ExtendedBlock> extendedBlocks,
List<Token<BlockTokenIdentifier>> dnTokens, int timeout,
boolean connectToDnViaHostname) {
this.configuration = configuration;
this.timeout = timeout;
this.datanode = datanode;
this.extendedBlocks = extendedBlocks;
this.dnTokens = dnTokens;
this.connectToDnViaHostname = connectToDnViaHostname;
}
public DatanodeInfo getDatanodeInfo() {
return datanode;
}
@Override
public HdfsBlocksMetadata call() throws Exception {
HdfsBlocksMetadata metadata = null;
// Create the RPC proxy and make the RPC
ClientDatanodeProtocol cdp = null;
try {
cdp = DFSUtil.createClientDatanodeProtocolProxy(datanode, configuration,
timeout, connectToDnViaHostname);
metadata = cdp.getHdfsBlocksMetadata(extendedBlocks, dnTokens);
} catch (IOException e) {
// Bubble this up to the caller, handle with the Future
throw e;
} finally {
if (cdp != null) {
RPC.stopProxy(cdp);
}
}
return metadata;
}
}
}

69
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/ByteRangeInputStream.java
View File

@ -22,12 +22,15 @@
import java.io.InputStream;
import java.net.HttpURLConnection;
import java.net.URL;
import java.util.List;
import java.util.Map;
import java.util.StringTokenizer;
import org.apache.commons.io.input.BoundedInputStream;
import org.apache.hadoop.fs.FSInputStream;
import org.apache.hadoop.hdfs.server.namenode.StreamFile;
import com.google.common.annotations.VisibleForTesting;
import com.google.common.net.HttpHeaders;
/**
* To support HTTP byte streams, a new connection to an HTTP server needs to be
@ -70,7 +73,7 @@ enum StreamStatus {
protected URLOpener resolvedURL;
protected long startPos = 0;
protected long currentPos = 0;
protected long filelength;
protected Long fileLength = null;
StreamStatus status = StreamStatus.SEEK;
@ -114,28 +117,60 @@ protected InputStream openInputStream() throws IOException {
final URLOpener opener = resolved? resolvedURL: originalURL;
final HttpURLConnection connection = opener.connect(startPos, resolved);
final String cl = connection.getHeaderField(StreamFile.CONTENT_LENGTH);
if (cl == null) {
throw new IOException(StreamFile.CONTENT_LENGTH+" header is missing");
}
final long streamlength = Long.parseLong(cl);
filelength = startPos + streamlength;
// Java has a bug with >2GB request streams. It won't bounds check
// the reads so the transfer blocks until the server times out
InputStream is =
new BoundedInputStream(connection.getInputStream(), streamlength);
resolvedURL.setURL(getResolvedUrl(connection));
return is;
InputStream in = connection.getInputStream();
final Map<String, List<String>> headers = connection.getHeaderFields();
if (isChunkedTransferEncoding(headers)) {
// file length is not known
fileLength = null;
} else {
// for non-chunked transfer-encoding, get content-length
final String cl = connection.getHeaderField(HttpHeaders.CONTENT_LENGTH);
if (cl == null) {
throw new IOException(HttpHeaders.CONTENT_LENGTH + " is missing: "
+ headers);
}
final long streamlength = Long.parseLong(cl);
fileLength = startPos + streamlength;
// Java has a bug with >2GB request streams. It won't bounds check
// the reads so the transfer blocks until the server times out
in = new BoundedInputStream(in, streamlength);
}
return in;
}
private static boolean isChunkedTransferEncoding(
final Map<String, List<String>> headers) {
return contains(headers, HttpHeaders.TRANSFER_ENCODING, "chunked")
|| contains(headers, HttpHeaders.TE, "chunked");
}
/** Does the HTTP header map contain the given key, value pair? */
private static boolean contains(final Map<String, List<String>> headers,
final String key, final String value) {
final List<String> values = headers.get(key);
if (values != null) {
for(String v : values) {
for(final StringTokenizer t = new StringTokenizer(v, ",");
t.hasMoreTokens(); ) {
if (value.equalsIgnoreCase(t.nextToken())) {
return true;
}
}
}
}
return false;
}
private int update(final int n) throws IOException {
if (n != -1) {
currentPos += n;
} else if (currentPos < filelength) {
} else if (fileLength != null && currentPos < fileLength) {
throw new IOException("Got EOF but currentPos = " + currentPos
+ " < filelength = " + filelength);
+ " < filelength = " + fileLength);
}
return n;
}

218
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSClient.java
View File

@ -45,8 +45,6 @@
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_CLIENT_WRITE_PACKET_SIZE_DEFAULT;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_CLIENT_WRITE_PACKET_SIZE_KEY;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_DATANODE_SOCKET_WRITE_TIMEOUT_KEY;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_ENCRYPT_DATA_TRANSFER_DEFAULT;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_ENCRYPT_DATA_TRANSFER_KEY;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_REPLICATION_DEFAULT;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_REPLICATION_KEY;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_CLIENT_USE_DN_HOSTNAME;
@ -69,6 +67,7 @@
import java.util.Collections;
import java.util.EnumSet;
import java.util.HashMap;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;
import java.util.Random;
@ -80,6 +79,7 @@
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.BlockStorageLocation;
import org.apache.hadoop.fs.CommonConfigurationKeysPublic;
import org.apache.hadoop.fs.ContentSummary;
import org.apache.hadoop.fs.CreateFlag;
@ -87,12 +87,17 @@
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.FsServerDefaults;
import org.apache.hadoop.fs.FsStatus;
import org.apache.hadoop.fs.HdfsBlockLocation;
import org.apache.hadoop.fs.InvalidPathException;
import org.apache.hadoop.fs.MD5MD5CRC32CastagnoliFileChecksum;
import org.apache.hadoop.fs.MD5MD5CRC32FileChecksum;
import org.apache.hadoop.fs.MD5MD5CRC32GzipFileChecksum;
import org.apache.hadoop.fs.Options;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.ParentNotDirectoryException;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.UnresolvedLinkException;
import org.apache.hadoop.fs.VolumeId;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.hdfs.client.HdfsDataInputStream;
import org.apache.hadoop.hdfs.client.HdfsDataOutputStream;
@ -102,10 +107,10 @@
import org.apache.hadoop.hdfs.protocol.DatanodeInfo;
import org.apache.hadoop.hdfs.protocol.DirectoryListing;
import org.apache.hadoop.hdfs.protocol.ExtendedBlock;
import org.apache.hadoop.hdfs.protocol.HdfsBlocksMetadata;
import org.apache.hadoop.hdfs.protocol.HdfsConstants;
import org.apache.hadoop.hdfs.protocol.HdfsConstants.DatanodeReportType;
import org.apache.hadoop.hdfs.protocol.HdfsConstants.SafeModeAction;
import org.apache.hadoop.hdfs.protocol.HdfsConstants.UpgradeAction;
import org.apache.hadoop.hdfs.protocol.HdfsFileStatus;
import org.apache.hadoop.hdfs.protocol.HdfsProtoUtil;
import org.apache.hadoop.hdfs.protocol.LocatedBlock;
@ -120,11 +125,11 @@
import org.apache.hadoop.hdfs.protocol.proto.DataTransferProtos.BlockOpResponseProto;
import org.apache.hadoop.hdfs.protocol.proto.DataTransferProtos.OpBlockChecksumResponseProto;
import org.apache.hadoop.hdfs.protocol.proto.DataTransferProtos.Status;
import org.apache.hadoop.hdfs.security.token.block.DataEncryptionKey;
import org.apache.hadoop.hdfs.security.token.block.BlockTokenIdentifier;
import org.apache.hadoop.hdfs.security.token.block.DataEncryptionKey;
import org.apache.hadoop.hdfs.security.token.block.InvalidBlockTokenException;
import org.apache.hadoop.hdfs.security.token.delegation.DelegationTokenIdentifier;
import org.apache.hadoop.hdfs.server.common.HdfsServerConstants;
import org.apache.hadoop.hdfs.server.common.UpgradeStatusReport;
import org.apache.hadoop.hdfs.server.namenode.NameNode;
import org.apache.hadoop.hdfs.server.namenode.SafeModeException;
import org.apache.hadoop.io.DataOutputBuffer;
@ -199,8 +204,7 @@ static class Conf {
final int maxBlockAcquireFailures;
final int confTime;
final int ioBufferSize;
final int checksumType;
final int bytesPerChecksum;
final ChecksumOpt defaultChecksumOpt;
final int writePacketSize;
final int socketTimeout;
final int socketCacheCapacity;
@ -216,6 +220,9 @@ static class Conf {
final FsPermission uMask;
final boolean useLegacyBlockReader;
final boolean connectToDnViaHostname;
final boolean getHdfsBlocksMetadataEnabled;
final int getFileBlockStorageLocationsNumThreads;
final int getFileBlockStorageLocationsTimeout;
Conf(Configuration conf) {
maxFailoverAttempts = conf.getInt(
@ -236,9 +243,7 @@ static class Conf {
ioBufferSize = conf.getInt(
CommonConfigurationKeysPublic.IO_FILE_BUFFER_SIZE_KEY,
CommonConfigurationKeysPublic.IO_FILE_BUFFER_SIZE_DEFAULT);
checksumType = getChecksumType(conf);
bytesPerChecksum = conf.getInt(DFS_BYTES_PER_CHECKSUM_KEY,
DFS_BYTES_PER_CHECKSUM_DEFAULT);
defaultChecksumOpt = getChecksumOptFromConf(conf);
socketTimeout = conf.getInt(DFS_CLIENT_SOCKET_TIMEOUT_KEY,
HdfsServerConstants.READ_TIMEOUT);
/** dfs.write.packet.size is an internal config variable */
@ -268,26 +273,57 @@ static class Conf {
DFS_CLIENT_USE_LEGACY_BLOCKREADER_DEFAULT);
connectToDnViaHostname = conf.getBoolean(DFS_CLIENT_USE_DN_HOSTNAME,
DFS_CLIENT_USE_DN_HOSTNAME_DEFAULT);
getHdfsBlocksMetadataEnabled = conf.getBoolean(
DFSConfigKeys.DFS_HDFS_BLOCKS_METADATA_ENABLED,
DFSConfigKeys.DFS_HDFS_BLOCKS_METADATA_ENABLED_DEFAULT);
getFileBlockStorageLocationsNumThreads = conf.getInt(
DFSConfigKeys.DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS,
DFSConfigKeys.DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS_DEFAULT);
getFileBlockStorageLocationsTimeout = conf.getInt(
DFSConfigKeys.DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT,
DFSConfigKeys.DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_DEFAULT);
}
private int getChecksumType(Configuration conf) {
String checksum = conf.get(DFSConfigKeys.DFS_CHECKSUM_TYPE_KEY,
private DataChecksum.Type getChecksumType(Configuration conf) {
final String checksum = conf.get(
DFSConfigKeys.DFS_CHECKSUM_TYPE_KEY,
DFSConfigKeys.DFS_CHECKSUM_TYPE_DEFAULT);
if ("CRC32".equals(checksum)) {
return DataChecksum.CHECKSUM_CRC32;
} else if ("CRC32C".equals(checksum)) {
return DataChecksum.CHECKSUM_CRC32C;
} else if ("NULL".equals(checksum)) {
return DataChecksum.CHECKSUM_NULL;
} else {
LOG.warn("Bad checksum type: " + checksum + ". Using default.");
return DataChecksum.CHECKSUM_CRC32C;
try {
return DataChecksum.Type.valueOf(checksum);
} catch(IllegalArgumentException iae) {
LOG.warn("Bad checksum type: " + checksum + ". Using default "
+ DFSConfigKeys.DFS_CHECKSUM_TYPE_DEFAULT);
return DataChecksum.Type.valueOf(
DFSConfigKeys.DFS_CHECKSUM_TYPE_DEFAULT);
}
}
private DataChecksum createChecksum() {
return DataChecksum.newDataChecksum(
checksumType, bytesPerChecksum);
// Construct a checksum option from conf
private ChecksumOpt getChecksumOptFromConf(Configuration conf) {
DataChecksum.Type type = getChecksumType(conf);
int bytesPerChecksum = conf.getInt(DFS_BYTES_PER_CHECKSUM_KEY,
DFS_BYTES_PER_CHECKSUM_DEFAULT);
return new ChecksumOpt(type, bytesPerChecksum);
}
// create a DataChecksum with the default option.
private DataChecksum createChecksum() throws IOException {
return createChecksum(null);
}
private DataChecksum createChecksum(ChecksumOpt userOpt)
throws IOException {
// Fill in any missing field with the default.
ChecksumOpt myOpt = ChecksumOpt.processChecksumOpt(
defaultChecksumOpt, userOpt);
DataChecksum dataChecksum = DataChecksum.newDataChecksum(
myOpt.getChecksumType(),
myOpt.getBytesPerChecksum());
if (dataChecksum == null) {
throw new IOException("Invalid checksum type specified: "
+ myOpt.getChecksumType().name());
}
return dataChecksum;
}
}
@ -943,7 +979,81 @@ boolean recoverLease(String src) throws IOException {
public BlockLocation[] getBlockLocations(String src, long start,
long length) throws IOException, UnresolvedLinkException {
LocatedBlocks blocks = getLocatedBlocks(src, start, length);
return DFSUtil.locatedBlocks2Locations(blocks);
BlockLocation[] locations = DFSUtil.locatedBlocks2Locations(blocks);
HdfsBlockLocation[] hdfsLocations = new HdfsBlockLocation[locations.length];
for (int i = 0; i < locations.length; i++) {
hdfsLocations[i] = new HdfsBlockLocation(locations[i], blocks.get(i));
}
return hdfsLocations;
}
/**
* Get block location information about a list of {@link HdfsBlockLocation}.
* Used by {@link DistributedFileSystem#getFileBlockStorageLocations(List)} to
* get {@link BlockStorageLocation}s for blocks returned by
* {@link DistributedFileSystem#getFileBlockLocations(org.apache.hadoop.fs.FileStatus, long, long)}
* .
*
* This is done by making a round of RPCs to the associated datanodes, asking
* the volume of each block replica. The returned array of
* {@link BlockStorageLocation} expose this information as a
* {@link VolumeId}.
*
* @param blockLocations
* target blocks on which to query volume location information
* @return volumeBlockLocations original block array augmented with additional
* volume location information for each replica.
*/
public BlockStorageLocation[] getBlockStorageLocations(
List<BlockLocation> blockLocations) throws IOException,
UnsupportedOperationException, InvalidBlockTokenException {
if (!getConf().getHdfsBlocksMetadataEnabled) {
throw new UnsupportedOperationException("Datanode-side support for " +
"getVolumeBlockLocations() must also be enabled in the client " +
"configuration.");
}
// Downcast blockLocations and fetch out required LocatedBlock(s)
List<LocatedBlock> blocks = new ArrayList<LocatedBlock>();
for (BlockLocation loc : blockLocations) {
if (!(loc instanceof HdfsBlockLocation)) {
throw new ClassCastException("DFSClient#getVolumeBlockLocations " +
"expected to be passed HdfsBlockLocations");
}
HdfsBlockLocation hdfsLoc = (HdfsBlockLocation) loc;
blocks.add(hdfsLoc.getLocatedBlock());
}
// Re-group the LocatedBlocks to be grouped by datanodes, with the values
// a list of the LocatedBlocks on the datanode.
Map<DatanodeInfo, List<LocatedBlock>> datanodeBlocks =
new LinkedHashMap<DatanodeInfo, List<LocatedBlock>>();
for (LocatedBlock b : blocks) {
for (DatanodeInfo info : b.getLocations()) {
if (!datanodeBlocks.containsKey(info)) {
datanodeBlocks.put(info, new ArrayList<LocatedBlock>());
}
List<LocatedBlock> l = datanodeBlocks.get(info);
l.add(b);
}
}
// Make RPCs to the datanodes to get volume locations for its replicas
List<HdfsBlocksMetadata> metadatas = BlockStorageLocationUtil
.queryDatanodesForHdfsBlocksMetadata(conf, datanodeBlocks,
getConf().getFileBlockStorageLocationsNumThreads,
getConf().getFileBlockStorageLocationsTimeout,
getConf().connectToDnViaHostname);
// Regroup the returned VolumeId metadata to again be grouped by
// LocatedBlock rather than by datanode
Map<LocatedBlock, List<VolumeId>> blockVolumeIds = BlockStorageLocationUtil
.associateVolumeIdsWithBlocks(blocks, datanodeBlocks, metadatas);
// Combine original BlockLocations with new VolumeId information
BlockStorageLocation[] volumeBlockLocations = BlockStorageLocationUtil
.convertToVolumeBlockLocations(blocks, blockVolumeIds);
return volumeBlockLocations;
}
public DFSInputStream open(String src)
@ -1054,12 +1164,13 @@ public OutputStream create(String src,
return create(src, FsPermission.getDefault(),
overwrite ? EnumSet.of(CreateFlag.CREATE, CreateFlag.OVERWRITE)
: EnumSet.of(CreateFlag.CREATE), replication, blockSize, progress,
buffersize);
buffersize, null);
}
/**
* Call {@link #create(String, FsPermission, EnumSet, boolean, short,
* long, Progressable, int)} with <code>createParent</code> set to true.
* long, Progressable, int, ChecksumOpt)} with <code>createParent</code>
* set to true.
*/
public DFSOutputStream create(String src,
FsPermission permission,
@ -1067,10 +1178,11 @@ public DFSOutputStream create(String src,
short replication,
long blockSize,
Progressable progress,
int buffersize)
int buffersize,
ChecksumOpt checksumOpt)
throws IOException {
return create(src, permission, flag, true,
replication, blockSize, progress, buffersize);
replication, blockSize, progress, buffersize, checksumOpt);
}
/**
@ -1088,6 +1200,7 @@ public DFSOutputStream create(String src,
* @param blockSize maximum block size
* @param progress interface for reporting client progress
* @param buffersize underlying buffer size
* @param checksumOpt checksum options
*
* @return output stream
*
@ -1101,8 +1214,8 @@ public DFSOutputStream create(String src,
short replication,
long blockSize,
Progressable progress,
int buffersize)
throws IOException {
int buffersize,
ChecksumOpt checksumOpt) throws IOException {
checkOpen();
if (permission == null) {
permission = FsPermission.getDefault();
@ -1113,7 +1226,7 @@ public DFSOutputStream create(String src,
}
final DFSOutputStream result = DFSOutputStream.newStreamForCreate(this,
src, masked, flag, createParent, replication, blockSize, progress,
buffersize, dfsClientConf.createChecksum());
buffersize, dfsClientConf.createChecksum(checksumOpt));
beginFileLease(src, result);
return result;
}
@ -1151,15 +1264,13 @@ public DFSOutputStream primitiveCreate(String src,
long blockSize,
Progressable progress,
int buffersize,
int bytesPerChecksum)
ChecksumOpt checksumOpt)
throws IOException, UnresolvedLinkException {
checkOpen();
CreateFlag.validate(flag);
DFSOutputStream result = primitiveAppend(src, flag, buffersize, progress);
if (result == null) {
DataChecksum checksum = DataChecksum.newDataChecksum(
dfsClientConf.checksumType,
bytesPerChecksum);
DataChecksum checksum = dfsClientConf.createChecksum(checksumOpt);
result = DFSOutputStream.newStreamForCreate(this, src, absPermission,
flag, createParent, replication, blockSize, progress, buffersize,
checksum);
@ -1494,7 +1605,8 @@ public static MD5MD5CRC32FileChecksum getFileChecksum(String src,
}
List<LocatedBlock> locatedblocks = blockLocations.getLocatedBlocks();
final DataOutputBuffer md5out = new DataOutputBuffer();
int bytesPerCRC = 0;
int bytesPerCRC = -1;
DataChecksum.Type crcType = DataChecksum.Type.DEFAULT;
long crcPerBlock = 0;
boolean refetchBlocks = false;
int lastRetriedIndex = -1;
@ -1598,6 +1710,17 @@ else if (bpc != bytesPerCRC) {
checksumData.getMd5().toByteArray());
md5.write(md5out);
// read crc-type
final DataChecksum.Type ct = HdfsProtoUtil.
fromProto(checksumData.getCrcType());
if (i == 0) { // first block
crcType = ct;
} else if (crcType != DataChecksum.Type.MIXED
&& crcType != ct) {
// if crc types are mixed in a file
crcType = DataChecksum.Type.MIXED;
}
done = true;
if (LOG.isDebugEnabled()) {
@ -1623,7 +1746,18 @@ else if (bpc != bytesPerCRC) {
//compute file MD5
final MD5Hash fileMD5 = MD5Hash.digest(md5out.getData());
return new MD5MD5CRC32FileChecksum(bytesPerCRC, crcPerBlock, fileMD5);
switch (crcType) {
case CRC32:
return new MD5MD5CRC32GzipFileChecksum(bytesPerCRC,
crcPerBlock, fileMD5);
case CRC32C:
return new MD5MD5CRC32CastagnoliFileChecksum(bytesPerCRC,
crcPerBlock, fileMD5);
default:
// we should never get here since the validity was checked
// when getCrcType() was called above.
return null;
}
}
/**
@ -1786,14 +1920,6 @@ public void finalizeUpgrade() throws IOException {
namenode.finalizeUpgrade();
}
/**
* @see ClientProtocol#distributedUpgradeProgress(HdfsConstants.UpgradeAction)
*/
public UpgradeStatusReport distributedUpgradeProgress(UpgradeAction action)
throws IOException {
return namenode.distributedUpgradeProgress(action);
}
/**
*/
@Deprecated

10
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSConfigKeys.java
View File

@ -54,6 +54,12 @@ public class DFSConfigKeys extends CommonConfigurationKeys {
public static final int DFS_CLIENT_SOCKET_CACHE_CAPACITY_DEFAULT = 16;
public static final String DFS_CLIENT_USE_DN_HOSTNAME = "dfs.client.use.datanode.hostname";
public static final boolean DFS_CLIENT_USE_DN_HOSTNAME_DEFAULT = false;
public static final String DFS_HDFS_BLOCKS_METADATA_ENABLED = "dfs.datanode.hdfs-blocks-metadata.enabled";
public static final boolean DFS_HDFS_BLOCKS_METADATA_ENABLED_DEFAULT = false;
public static final String DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS = "dfs.client.file-block-storage-locations.num-threads";
public static final int DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_NUM_THREADS_DEFAULT = 10;
public static final String DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT = "dfs.client.file-block-storage-locations.timeout";
public static final int DFS_CLIENT_FILE_BLOCK_STORAGE_LOCATIONS_TIMEOUT_DEFAULT = 60;
// HA related configuration
public static final String DFS_CLIENT_FAILOVER_PROXY_PROVIDER_KEY_PREFIX = "dfs.client.failover.proxy.provider";
@ -172,7 +178,7 @@ public class DFSConfigKeys extends CommonConfigurationKeys {
// Replication monitoring related keys
public static final String DFS_NAMENODE_INVALIDATE_WORK_PCT_PER_ITERATION =
"dfs.namenode.invalidate.work.pct.per.iteration";
public static final int DFS_NAMENODE_INVALIDATE_WORK_PCT_PER_ITERATION_DEFAULT = 32;
public static final float DFS_NAMENODE_INVALIDATE_WORK_PCT_PER_ITERATION_DEFAULT = 0.32f;
public static final String DFS_NAMENODE_REPLICATION_WORK_MULTIPLIER_PER_ITERATION =
"dfs.namenode.replication.work.multiplier.per.iteration";
public static final int DFS_NAMENODE_REPLICATION_WORK_MULTIPLIER_PER_ITERATION_DEFAULT = 2;
@ -245,7 +251,7 @@ public class DFSConfigKeys extends CommonConfigurationKeys {
public static final String DFS_DATANODE_DU_RESERVED_KEY = "dfs.datanode.du.reserved";
public static final long DFS_DATANODE_DU_RESERVED_DEFAULT = 0;
public static final String DFS_DATANODE_HANDLER_COUNT_KEY = "dfs.datanode.handler.count";
public static final int DFS_DATANODE_HANDLER_COUNT_DEFAULT = 3;
public static final int DFS_DATANODE_HANDLER_COUNT_DEFAULT = 10;
public static final String DFS_DATANODE_HTTP_ADDRESS_KEY = "dfs.datanode.http.address";
public static final int DFS_DATANODE_HTTP_DEFAULT_PORT = 50075;
public static final String DFS_DATANODE_HTTP_ADDRESS_DEFAULT = "0.0.0.0:" + DFS_DATANODE_HTTP_DEFAULT_PORT;

55
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSUtil.java
View File

@ -80,6 +80,7 @@
import org.apache.hadoop.util.ToolRunner;
import com.google.common.base.Joiner;
import com.google.common.base.Preconditions;
import com.google.common.collect.Lists;
import com.google.common.collect.Maps;
import com.google.protobuf.BlockingService;
@ -282,13 +283,25 @@ public static BlockLocation[] locatedBlocks2Locations(LocatedBlocks blocks) {
if (blocks == null) {
return new BlockLocation[0];
}
int nrBlocks = blocks.locatedBlockCount();
return locatedBlocks2Locations(blocks.getLocatedBlocks());
}
/**
* Convert a List<LocatedBlock> to BlockLocation[]
* @param blocks A List<LocatedBlock> to be converted
* @return converted array of BlockLocation
*/
public static BlockLocation[] locatedBlocks2Locations(List<LocatedBlock> blocks) {
if (blocks == null) {
return new BlockLocation[0];
}
int nrBlocks = blocks.size();
BlockLocation[] blkLocations = new BlockLocation[nrBlocks];
if (nrBlocks == 0) {
return blkLocations;
}
int idx = 0;
for (LocatedBlock blk : blocks.getLocatedBlocks()) {
for (LocatedBlock blk : blocks) {
assert idx < nrBlocks : "Incorrect index";
DatanodeInfo[] locations = blk.getLocations();
String[] hosts = new String[locations.length];
@ -1131,4 +1144,42 @@ public static boolean parseHelpArgument(String[] args,
}
return false;
}
/**
* Get DFS_NAMENODE_INVALIDATE_WORK_PCT_PER_ITERATION from configuration.
*
* @param conf Configuration
* @return Value of DFS_NAMENODE_INVALIDATE_WORK_PCT_PER_ITERATION
*/
public static float getInvalidateWorkPctPerIteration(Configuration conf) {
float blocksInvalidateWorkPct = conf.getFloat(
DFSConfigKeys.DFS_NAMENODE_INVALIDATE_WORK_PCT_PER_ITERATION,
DFSConfigKeys.DFS_NAMENODE_INVALIDATE_WORK_PCT_PER_ITERATION_DEFAULT);
Preconditions.checkArgument(
(blocksInvalidateWorkPct > 0 && blocksInvalidateWorkPct <= 1.0f),
DFSConfigKeys.DFS_NAMENODE_INVALIDATE_WORK_PCT_PER_ITERATION +
" = '" + blocksInvalidateWorkPct + "' is invalid. " +
"It should be a positive, non-zero float value, not greater than 1.0f, " +
"to indicate a percentage.");
return blocksInvalidateWorkPct;
}
/**
* Get DFS_NAMENODE_REPLICATION_WORK_MULTIPLIER_PER_ITERATION from
* configuration.
*
* @param conf Configuration
* @return Value of DFS_NAMENODE_REPLICATION_WORK_MULTIPLIER_PER_ITERATION
*/
public static int getReplWorkMultiplier(Configuration conf) {
int blocksReplWorkMultiplier = conf.getInt(
DFSConfigKeys.DFS_NAMENODE_REPLICATION_WORK_MULTIPLIER_PER_ITERATION,
DFSConfigKeys.DFS_NAMENODE_REPLICATION_WORK_MULTIPLIER_PER_ITERATION_DEFAULT);
Preconditions.checkArgument(
(blocksReplWorkMultiplier > 0),
DFSConfigKeys.DFS_NAMENODE_REPLICATION_WORK_MULTIPLIER_PER_ITERATION +
" = '" + blocksReplWorkMultiplier + "' is invalid. " +
"It should be a positive, non-zero integer value.");
return blocksReplWorkMultiplier;
}
}

64
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DistributedFileSystem.java
View File

@ -32,6 +32,8 @@
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.ContentSummary;
import org.apache.hadoop.fs.CreateFlag;
import org.apache.hadoop.fs.BlockStorageLocation;
import org.apache.hadoop.fs.VolumeId;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
@ -40,6 +42,7 @@
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.MD5MD5CRC32FileChecksum;
import org.apache.hadoop.fs.Options;
import org.apache.hadoop.fs.Options.ChecksumOpt;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.PathFilter;
import org.apache.hadoop.fs.RemoteIterator;
@ -52,12 +55,11 @@
import org.apache.hadoop.hdfs.protocol.HdfsConstants;
import org.apache.hadoop.hdfs.protocol.HdfsConstants.DatanodeReportType;
import org.apache.hadoop.hdfs.protocol.HdfsConstants.SafeModeAction;
import org.apache.hadoop.hdfs.protocol.HdfsConstants.UpgradeAction;
import org.apache.hadoop.hdfs.protocol.HdfsFileStatus;
import org.apache.hadoop.hdfs.protocol.HdfsLocatedFileStatus;
import org.apache.hadoop.hdfs.protocol.LocatedBlock;
import org.apache.hadoop.hdfs.security.token.block.InvalidBlockTokenException;
import org.apache.hadoop.hdfs.security.token.delegation.DelegationTokenIdentifier;
import org.apache.hadoop.hdfs.server.common.UpgradeStatusReport;
import org.apache.hadoop.hdfs.server.namenode.NameNode;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.security.AccessControlException;
@ -189,6 +191,36 @@ public BlockLocation[] getFileBlockLocations(Path p,
}
/**
* Used to query storage location information for a list of blocks. This list
* of blocks is normally constructed via a series of calls to
* {@link DistributedFileSystem#getFileBlockLocations(Path, long, long)} to
* get the blocks for ranges of a file.
*
* The returned array of {@link BlockStorageLocation} augments
* {@link BlockLocation} with a {@link VolumeId} per block replica. The
* VolumeId specifies the volume on the datanode on which the replica resides.
* The VolumeId has to be checked via {@link VolumeId#isValid()} before being
* used because volume information can be unavailable if the corresponding
* datanode is down or if the requested block is not found.
*
* This API is unstable, and datanode-side support is disabled by default. It
* can be enabled by setting "dfs.datanode.hdfs-blocks-metadata.enabled" to
* true.
*
* @param blocks
* List of target BlockLocations to query volume location information
* @return volumeBlockLocations Augmented array of
* {@link BlockStorageLocation}s containing additional volume location
* information for each replica of each block.
*/
@InterfaceStability.Unstable
public BlockStorageLocation[] getFileBlockStorageLocations(
List<BlockLocation> blocks) throws IOException,
UnsupportedOperationException, InvalidBlockTokenException {
return dfs.getBlockStorageLocations(blocks);
}
@Override
public void setVerifyChecksum(boolean verifyChecksum) {
this.verifyChecksum = verifyChecksum;
@ -225,19 +257,19 @@ public HdfsDataOutputStream append(Path f, int bufferSize,
public HdfsDataOutputStream create(Path f, FsPermission permission,
boolean overwrite, int bufferSize, short replication, long blockSize,
Progressable progress) throws IOException {
return create(f, permission,
return this.create(f, permission,
overwrite ? EnumSet.of(CreateFlag.CREATE, CreateFlag.OVERWRITE)
: EnumSet.of(CreateFlag.CREATE), bufferSize, replication,
blockSize, progress);
blockSize, progress, null);
}
@Override
public HdfsDataOutputStream create(Path f, FsPermission permission,
EnumSet<CreateFlag> cflags, int bufferSize, short replication, long blockSize,
Progressable progress) throws IOException {
Progressable progress, ChecksumOpt checksumOpt) throws IOException {
statistics.incrementWriteOps(1);
final DFSOutputStream out = dfs.create(getPathName(f), permission, cflags,
replication, blockSize, progress, bufferSize);
replication, blockSize, progress, bufferSize, checksumOpt);
return new HdfsDataOutputStream(out, statistics);
}
@ -246,11 +278,11 @@ public HdfsDataOutputStream create(Path f, FsPermission permission,
protected HdfsDataOutputStream primitiveCreate(Path f,
FsPermission absolutePermission, EnumSet<CreateFlag> flag, int bufferSize,
short replication, long blockSize, Progressable progress,
int bytesPerChecksum) throws IOException {
ChecksumOpt checksumOpt) throws IOException {
statistics.incrementWriteOps(1);
return new HdfsDataOutputStream(dfs.primitiveCreate(getPathName(f),
absolutePermission, flag, true, replication, blockSize,
progress, bufferSize, bytesPerChecksum),statistics);
progress, bufferSize, checksumOpt),statistics);
}
/**
@ -265,7 +297,8 @@ public HdfsDataOutputStream createNonRecursive(Path f, FsPermission permission,
flag.add(CreateFlag.CREATE);
}
return new HdfsDataOutputStream(dfs.create(getPathName(f), permission, flag,
false, replication, blockSize, progress, bufferSize), statistics);
false, replication, blockSize, progress,
bufferSize, null), statistics);
}
@Override
@ -619,11 +652,6 @@ public void finalizeUpgrade() throws IOException {
dfs.finalizeUpgrade();
}
public UpgradeStatusReport distributedUpgradeProgress(UpgradeAction action
) throws IOException {
return dfs.distributedUpgradeProgress(action);
}
/*
* Requests the namenode to dump data strcutures into specified
* file.
@ -765,14 +793,6 @@ public Token<DelegationTokenIdentifier> getDelegationToken(Text renewer)
return getDelegationToken(renewer.toString());
}
@Override // FileSystem
public List<Token<?>> getDelegationTokens(String renewer) throws IOException {
List<Token<?>> tokenList = new ArrayList<Token<?>>();
Token<DelegationTokenIdentifier> token = this.getDelegationToken(renewer);
tokenList.add(token);
return tokenList;
}
/**
* Renew an existing delegation token.
*

18
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/ClientDatanodeProtocol.java
View File

@ -18,6 +18,7 @@
package org.apache.hadoop.hdfs.protocol;
import java.io.IOException;
import java.util.List;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
@ -106,4 +107,21 @@ public interface ClientDatanodeProtocol {
*/
BlockLocalPathInfo getBlockLocalPathInfo(ExtendedBlock block,
Token<BlockTokenIdentifier> token) throws IOException;
/**
* Retrieves volume location information about a list of blocks on a datanode.
* This is in the form of an opaque {@link VolumeId} for each configured
* data directory, which is not guaranteed to be the same across DN restarts.
*
* @param blocks
* list of blocks on the local datanode
* @param tokens
* block access tokens corresponding to the requested blocks
* @return an HdfsBlocksMetadata that associates {@link ExtendedBlock}s with
* data directories
* @throws IOException
* if datanode is unreachable, or replica is not found on datanode
*/
HdfsBlocksMetadata getHdfsBlocksMetadata(List<ExtendedBlock> blocks,
List<Token<BlockTokenIdentifier>> tokens) throws IOException;
}

12
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/ClientProtocol.java
View File

@ -33,8 +33,6 @@
import org.apache.hadoop.fs.Options.Rename;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.hdfs.DFSConfigKeys;
import org.apache.hadoop.hdfs.protocol.HdfsConstants.UpgradeAction;
import org.apache.hadoop.hdfs.server.common.UpgradeStatusReport;
import org.apache.hadoop.hdfs.server.namenode.NotReplicatedYetException;
import org.apache.hadoop.hdfs.server.namenode.SafeModeException;
import org.apache.hadoop.io.EnumSetWritable;
@ -694,16 +692,6 @@ public boolean restoreFailedStorage(String arg)
*/
public void finalizeUpgrade() throws IOException;
/**
* Report distributed upgrade progress or force current upgrade to proceed.
*
* @param action {@link HdfsConstants.UpgradeAction} to perform
* @return upgrade status information or null if no upgrades are in progress
* @throws IOException
*/
public UpgradeStatusReport distributedUpgradeProgress(UpgradeAction action)
throws IOException;
/**
* @return CorruptFileBlocks, containing a list of corrupt files (with
* duplicates if there is more than one corrupt block in a file)

94
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/HdfsBlocksMetadata.java Normal file
View File

@ -0,0 +1,94 @@
/**
* 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.protocol;
import java.util.List;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceStability;
/**
* Augments an array of blocks on a datanode with additional information about
* where the block is stored.
*/
@InterfaceAudience.Private
@InterfaceStability.Unstable
public class HdfsBlocksMetadata {
/**
* List of blocks
*/
private final ExtendedBlock[] blocks;
/**
* List of volumes
*/
private final List<byte[]> volumeIds;
/**
* List of indexes into <code>volumeIds</code>, one per block in
* <code>blocks</code>. A value of Integer.MAX_VALUE indicates that the
* block was not found.
*/
private final List<Integer> volumeIndexes;
/**
* Constructs HdfsBlocksMetadata.
*
* @param blocks
* List of blocks described
* @param volumeIds
* List of potential volume identifiers, specifying volumes where
* blocks may be stored
* @param volumeIndexes
* Indexes into the list of volume identifiers, one per block
*/
public HdfsBlocksMetadata(ExtendedBlock[] blocks, List<byte[]> volumeIds,
List<Integer> volumeIndexes) {
this.blocks = blocks;
this.volumeIds = volumeIds;
this.volumeIndexes = volumeIndexes;
}
/**
* Get the array of blocks.
*
* @return array of blocks
*/
public ExtendedBlock[] getBlocks() {
return blocks;
}
/**
* Get the list of volume identifiers in raw byte form.
*
* @return list of ids
*/
public List<byte[]> getVolumeIds() {
return volumeIds;
}
/**
* Get a list of indexes into the array of {@link VolumeId}s, one per block.
*
* @return list of indexes
*/
public List<Integer> getVolumeIndexes() {
return volumeIndexes;
}
}

12
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/HdfsConstants.java
View File

@ -60,7 +60,7 @@ protected HdfsConstants() {
public static int MAX_PATH_LENGTH = 8000;
public static int MAX_PATH_DEPTH = 1000;
// TODO mb@media-style.com: should be conf injected?
// TODO should be conf injected?
public static final int DEFAULT_DATA_SOCKET_SIZE = 128 * 1024;
public static final int IO_FILE_BUFFER_SIZE = new HdfsConfiguration().getInt(
DFSConfigKeys.IO_FILE_BUFFER_SIZE_KEY,
@ -84,16 +84,6 @@ public static enum DatanodeReportType {
// An invalid transaction ID that will never be seen in a real namesystem.
public static final long INVALID_TXID = -12345;
/**
* Distributed upgrade actions:
*
* 1. Get upgrade status. 2. Get detailed upgrade status. 3. Proceed with the
* upgrade if it is stuck, no matter what the status is.
*/
public static enum UpgradeAction {
GET_STATUS, DETAILED_STATUS, FORCE_PROCEED;
}
/**
* URI Scheme for hdfs://namenode/ URIs.
*/

Some files were not shown because too many files have changed in this diff Show More