/**
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.hadoop.hdfs.server.namenode;
import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertNotNull;
import static org.junit.Assert.assertTrue;
import static org.junit.Assert.fail;
import static org.mockito.Matchers.anyObject;
import static org.mockito.Matchers.anyBoolean;
import static org.mockito.Mockito.doAnswer;
import static org.mockito.Mockito.doThrow;
import static org.mockito.Mockito.spy;
import java.io.File;
import java.io.IOException;
import java.io.OutputStream;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.commons.logging.impl.Log4JLogger;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.permission.FsPermission;
import org.apache.hadoop.fs.permission.PermissionStatus;
import org.apache.hadoop.hdfs.DistributedFileSystem;
import org.apache.hadoop.hdfs.MiniDFSCluster;
import org.apache.hadoop.hdfs.protocol.FSConstants.SafeModeAction;
import org.apache.hadoop.hdfs.server.common.Storage.StorageDirectory;
import org.apache.hadoop.hdfs.server.namenode.metrics.NameNodeMetrics;
import org.apache.hadoop.io.IOUtils;
import org.apache.log4j.Level;
import org.junit.Test;
import org.mockito.Mockito;
import org.mockito.invocation.InvocationOnMock;
import org.mockito.stubbing.Answer;
/**
* Test various failure scenarios during saveNamespace() operation.
* Cases covered:
* <ol>
* <li>Recover from failure while saving into the second storage directory</li>
* <li>Recover from failure while moving current into lastcheckpoint.tmp</li>
* <li>Recover from failure while moving lastcheckpoint.tmp into
* previous.checkpoint</li>
* <li>Recover from failure while rolling edits file</li>
* </ol>
*/
public class TestSaveNamespace {
static {
((Log4JLogger)FSImage.LOG).getLogger().setLevel(Level.ALL);
}
private static final Log LOG = LogFactory.getLog(TestSaveNamespace.class);
private static class FaultySaveImage implements Answer<Void> {
int count = 0;
boolean throwRTE = true;
// generate either a RuntimeException or IOException
public FaultySaveImage(boolean throwRTE) {
this.throwRTE = throwRTE;
}
public Void answer(InvocationOnMock invocation) throws Throwable {
Object[] args = invocation.getArguments();
StorageDirectory sd = ((FileImageManager)args[1]).getStorageDirectory();
if (count++ == 1) {
LOG.info("Injecting fault for sd: " + sd);
if (throwRTE) {
throw new RuntimeException("Injected fault: saveFSImage second time");
} else {
throw new IOException("Injected fault: saveFSImage second time");
}
}
LOG.info("Not injecting fault for sd: " + sd);
return (Void)invocation.callRealMethod();
}
}
private enum Fault {
SAVE_SECOND_FSIMAGE_RTE,
SAVE_SECOND_FSIMAGE_IOE,
SAVE_ALL_FSIMAGES,
WRITE_STORAGE_ALL,
WRITE_STORAGE_ONE
};
private void saveNamespaceWithInjectedFault(Fault fault) throws Exception {
Configuration conf = getConf();
NameNode.myMetrics = new NameNodeMetrics(conf, null);
NameNode.format(conf);
NameNode nn = new NameNode(conf);
FSNamesystem fsn = nn.getNamesystem();
// Replace the FSImage with a spy
FSImage originalImage = fsn.dir.fsImage;
NNStorage storage = originalImage.storage;
NNStorage spyStorage = spy(storage);
originalImage.storage = spyStorage;
FSImage spyImage = spy(originalImage);
fsn.dir.fsImage = spyImage;
boolean shouldFail = false; // should we expect the save operation to fail
// inject fault
switch(fault) {
case SAVE_SECOND_FSIMAGE_RTE:
// The spy throws a RuntimeException when writing to the second directory
doAnswer(new FaultySaveImage(true)).
when(spyImage).saveFSImage(
(SaveNamespaceContext)anyObject(),
(ImageManager)anyObject(), anyBoolean());
shouldFail = false;
break;
case SAVE_SECOND_FSIMAGE_IOE:
// The spy throws an IOException when writing to the second directory
doAnswer(new FaultySaveImage(false)).
when(spyImage).saveFSImage(
(SaveNamespaceContext)anyObject(),
(ImageManager)anyObject(), anyBoolean());
shouldFail = false;
break;
case SAVE_ALL_FSIMAGES:
// The spy throws IOException in all directories
doThrow(new RuntimeException("Injected")).
when(spyImage).saveFSImage(
(SaveNamespaceContext)anyObject(),
(ImageManager)anyObject(), anyBoolean());
shouldFail = true;
break;
case WRITE_STORAGE_ALL:
// The spy throws an exception before writing any VERSION files
doThrow(new RuntimeException("Injected"))
.when(spyStorage).writeAll();
shouldFail = true;
break;
}
try {
doAnEdit(fsn, 1);
// Save namespace - this may fail, depending on fault injected
fsn.setSafeMode(SafeModeAction.SAFEMODE_ENTER);
try {
fsn.saveNamespace(false, false);
if (shouldFail) {
fail("Did not fail!");
}
} catch (Exception e) {
if (! shouldFail) {
throw e;
} else {
LOG.info("Test caught expected exception", e);
}
}
fsn.setSafeMode(SafeModeAction.SAFEMODE_LEAVE);
// Should still be able to perform edits
doAnEdit(fsn, 2);
// Now shut down and restart the namesystem
originalImage.close();
fsn.close();
fsn = null;
// Start a new namesystem, which should be able to recover
// the namespace from the previous incarnation.
nn.stop();
nn = new NameNode(conf);
fsn = nn.getNamesystem();
// Make sure the image loaded including our edits.
checkEditExists(fsn, 1);
checkEditExists(fsn, 2);
} finally {
if (fsn != null) {
fsn.close();
}
}
}
/**
* Verify that a saveNamespace command brings faulty directories
* in fs.name.dir and fs.edit.dir back online.
*/
@Test
public void testReinsertnamedirsInSavenamespace() throws Exception {
// create a configuration with the key to restore error
// directories in fs.name.dir
Configuration conf = getConf();
MiniDFSCluster cluster = new MiniDFSCluster(conf, 1, true, null);
cluster.waitActive();
FSNamesystem fsn = cluster.getNameNode().getNamesystem();
// Replace the FSImage with a spy
FSImage originalImage = fsn.dir.fsImage;
FSImage spyImage = spy(originalImage);
long mrctxid = originalImage.storage.getMostRecentCheckpointTxId();
spyImage.setImageDigest(mrctxid ,originalImage.getImageDigest(mrctxid));
fsn.dir.fsImage = spyImage;
File rootDir = spyImage.storage.getStorageDir(0).getRoot();
rootDir.setExecutable(false);
rootDir.setWritable(false);
rootDir.setReadable(false);
try {
doAnEdit(fsn, 1);
fsn.setSafeMode(SafeModeAction.SAFEMODE_ENTER);
// Save namespace - this injects a fault and marks one
// directory as faulty.
LOG.info("Doing the first savenamespace.");
fsn.saveNamespace(false, false);
LOG.warn("First savenamespace sucessful.");
assertTrue("Savenamespace should have marked one directory as bad." +
" But found " + spyImage.getRemovedStorageDirs().size() +
" bad directories.",
spyImage.getRemovedStorageDirs().size() == 1);
rootDir.setExecutable(true);
rootDir.setWritable(true);
rootDir.setReadable(true);
// The next call to savenamespace should try inserting the
// erroneous directory back to fs.name.dir. This command should
// be successful.
LOG.info("Doing the second savenamespace.");
fsn.saveNamespace(false, false);
LOG.warn("Second savenamespace sucessful.");
assertTrue("Savenamespace should have been successful in removing " +
" bad directories from Image." +
" But found " + originalImage.getRemovedStorageDirs().size() +
" bad directories.",
originalImage.getRemovedStorageDirs().size() == 0);
// Now shut down and restart the namesystem
LOG.info("Shutting down fsimage.");
originalImage.close();
fsn.close();
fsn = null;
cluster.shutdown();
// Start a new namesystem, which should be able to recover
// the namespace from the previous incarnation.
LOG.info("Loading new FSmage from disk.");
cluster = new MiniDFSCluster(conf, 1, false, null);
cluster.waitActive();
fsn = cluster.getNameNode().getNamesystem();
// Make sure the image loaded including our edit.
LOG.info("Checking reloaded image.");
checkEditExists(cluster.getNameNode().namesystem, 1);
LOG.info("Reloaded image is good.");
} finally {
fsn.close();
cluster.shutdown();
}
}
@Test
public void testRTEWhileSavingSecondImage() throws Exception {
saveNamespaceWithInjectedFault(Fault.SAVE_SECOND_FSIMAGE_RTE);
}
@Test
public void testIOEWhileSavingSecondImage() throws Exception {
saveNamespaceWithInjectedFault(Fault.SAVE_SECOND_FSIMAGE_IOE);
}
@Test
public void testCrashInAllImageDirs() throws Exception {
saveNamespaceWithInjectedFault(Fault.SAVE_ALL_FSIMAGES);
}
@Test
public void testCrashWhenWritingVersionFiles() throws Exception {
saveNamespaceWithInjectedFault(Fault.WRITE_STORAGE_ALL);
}
/**
* Test case where savenamespace fails in all directories
* and then the NN shuts down. Here we should recover from the
* failed checkpoint since it only affected ".ckpt" files, not
* valid image files
*/
@Test
public void testFailedSaveNamespace() throws Exception {
doTestFailedSaveNamespace(false);
}
/**
* Test case where saveNamespace fails in all directories, but then
* the operator restores the directories and calls it again.
* This should leave the NN in a clean state for next start.
*/
@Test
public void testFailedSaveNamespaceWithRecovery() throws Exception {
doTestFailedSaveNamespace(true);
}
/**
* Injects a failure on all storage directories while saving namespace.
*
* @param restoreStorageAfterFailure if true, will try to save again after
* clearing the failure injection
*/
private void doTestFailedSaveNamespace(boolean restoreStorageAfterFailure)
throws Exception {
Configuration conf = getConf();
MiniDFSCluster cluster = new MiniDFSCluster(conf, 0, true, null);
cluster.waitActive();
FSNamesystem fsn = cluster.getNameNode().getNamesystem();
// Replace the FSImage with a spy
final FSImage originalImage = fsn.dir.fsImage;
NNStorage storage = originalImage.storage;
storage.close(); // unlock any directories that FSNamesystem's initialization may have locked
NNStorage spyStorage = spy(storage);
originalImage.storage = spyStorage;
FSImage spyImage = spy(originalImage);
fsn.dir.fsImage = spyImage;
spyImage.storage.setStorageDirectories(
NNStorageConfiguration.getNamespaceDirs(conf),
NNStorageConfiguration.getNamespaceEditsDirs(conf));
doThrow(new IOException("Injected fault: saveFSImage")).
when(spyImage).saveFSImage(
(SaveNamespaceContext)anyObject(),
(ImageManager)anyObject(), anyBoolean());
try {
doAnEdit(fsn, 1);
// Save namespace
fsn.setSafeMode(SafeModeAction.SAFEMODE_ENTER);
try {
fsn.saveNamespace(false, false);
fail("saveNamespace did not fail even when all directories failed!");
} catch (IOException ioe) {
LOG.info("Got expected exception", ioe);
}
// Ensure that, if storage dirs come back online, things work again.
if (restoreStorageAfterFailure) {
Mockito.reset(spyImage);
fsn.saveNamespace(false, false);
checkEditExists(fsn, 1);
}
// Now shut down and restart the NN
originalImage.close();
fsn.close();
fsn = null;
// Start a new namesystem, which should be able to recover
// the namespace from the previous incarnation.
cluster = new MiniDFSCluster(conf, 0, false, null);
cluster.waitActive();
fsn = cluster.getNameNode().getNamesystem();
// Make sure the image loaded including our edits.
checkEditExists(fsn, 1);
} finally {
if (fsn != null) {
fsn.close();
}
}
}
@Test
public void testSaveWhileEditsRolled() throws Exception {
Configuration conf = getConf();
NameNode.myMetrics = new NameNodeMetrics(conf, null);
NameNode.format(conf);
NameNode nn = new NameNode(conf);
FSNamesystem fsn = nn.getNamesystem();
// Replace the FSImage with a spy
final FSImage originalImage = fsn.dir.fsImage;
FSImage spyImage = spy(originalImage);
fsn.dir.fsImage = spyImage;
try {
doAnEdit(fsn, 1);
CheckpointSignature sig = fsn.rollEditLog();
LOG.warn("Checkpoint signature: " + sig);
// Do another edit
doAnEdit(fsn, 2);
// Save namespace
fsn.setSafeMode(SafeModeAction.SAFEMODE_ENTER);
fsn.saveNamespace(false, false);
// Now shut down and restart the NN
nn.stop();
nn = null;
// Start a new namesystem, which should be able to recover
// the namespace from the previous incarnation.
nn = new NameNode(conf);
fsn = nn.getNamesystem();
// Make sure the image loaded including our edits.
checkEditExists(fsn, 1);
checkEditExists(fsn, 2);
} finally {
if (nn != null) {
nn.stop();
}
}
}
@Test
public void testTxIdPersistence() throws Exception {
Configuration conf = getConf();
MiniDFSCluster cluster = new MiniDFSCluster(conf, 0, true, null);
cluster.waitActive();
FSNamesystem fsn = cluster.getNameNode().getNamesystem();
try {
// We have a BEGIN_LOG_SEGMENT txn to start
assertEquals(0, fsn.getEditLog().getLastWrittenTxId());
doAnEdit(fsn, 1);
assertEquals(1, fsn.getEditLog().getLastWrittenTxId());
fsn.setSafeMode(SafeModeAction.SAFEMODE_ENTER);
fsn.saveNamespace(false, false);
// 2 more txns: END the first segment, BEGIN a new one
assertEquals(3, fsn.getEditLog().getLastWrittenTxId());
// Shut down and restart
fsn.getFSImage().close();
fsn.close();
// 1 more txn to END that segment
assertEquals(4, fsn.getEditLog().getLastWrittenTxId());
fsn = null;
cluster = new MiniDFSCluster(conf, 0, false, null);
cluster.waitActive();
fsn = cluster.getNameNode().getNamesystem();
// 1 more txn to start new segment on restart
assertEquals(5, fsn.getEditLog().getLastWrittenTxId());
} finally {
if (fsn != null) {
fsn.close();
}
}
}
/**
* Test for save namespace should succeed when parent directory renamed with
* open lease and destination directory exist.
* This test is a regression for HDFS-2827
*/
@Test
public void testSaveNamespaceWithRenamedLease() throws Exception {
MiniDFSCluster cluster = new MiniDFSCluster(new Configuration(), 1, true, (String[]) null);
cluster.waitActive();
DistributedFileSystem fs = (DistributedFileSystem) cluster.getFileSystem();
OutputStream out = null;
try {
fs.mkdirs(new Path("/test-target"));
out = fs.create(new Path("/test-source/foo")); // don't close
fs.rename(new Path("/test-source/"), new Path("/test-target/"));
fs.setSafeMode(SafeModeAction.SAFEMODE_ENTER);
cluster.getNameNode().saveNamespace();
fs.setSafeMode(SafeModeAction.SAFEMODE_LEAVE);
} finally {
IOUtils.cleanup(LOG, out, fs);
if (cluster != null) {
cluster.shutdown();
}
}
}
private void doAnEdit(FSNamesystem fsn, int id) throws IOException {
// Make an edit
fsn.mkdirs(
"/test" + id,
new PermissionStatus("test", "Test",
new FsPermission((short)0777)));
}
private void checkEditExists(FSNamesystem fsn, int id) throws IOException {
// Make sure the image loaded including our edit.
assertNotNull(fsn.getFileInfo("/test" + id));
}
private Configuration getConf() throws IOException {
String baseDir = System.getProperty("test.build.data", "build/test/data/dfs/");
String nameDirs = baseDir + "name1" + "," + baseDir + "name2";
String editsDirs = baseDir + "edits1" + "," + baseDir + "edits2";
Configuration conf = new Configuration();
FileSystem.setDefaultUri(conf, "hdfs://localhost:0");
conf.set("dfs.http.address", "0.0.0.0:0");
conf.set("dfs.name.dir", nameDirs);
conf.set("dfs.name.edits.dir", editsDirs);
conf.set("dfs.secondary.http.address", "0.0.0.0:0");
conf.setBoolean("dfs.permissions", false);
return conf;
}
}