hadoop ReencryptionHandler 源码
haddop ReencryptionHandler 代码
文件路径：/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/ReencryptionHandler.java
/**
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.hadoop.hdfs.server.namenode;
import org.apache.hadoop.classification.VisibleForTesting;
import org.apache.hadoop.util.Preconditions;
import org.apache.hadoop.thirdparty.com.google.common.util.concurrent.ThreadFactoryBuilder;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.crypto.key.KeyProviderCryptoExtension.EncryptedKeyVersion;
import org.apache.hadoop.fs.FileEncryptionInfo;
import org.apache.hadoop.fs.XAttr;
import org.apache.hadoop.hdfs.protocol.HdfsFileStatus;
import org.apache.hadoop.hdfs.protocol.ReencryptionStatus;
import org.apache.hadoop.hdfs.protocol.ZoneReencryptionStatus;
import org.apache.hadoop.hdfs.protocol.ZoneReencryptionStatus.State;
import org.apache.hadoop.hdfs.server.namenode.FSTreeTraverser.TraverseInfo;
import org.apache.hadoop.hdfs.server.namenode.ReencryptionUpdater.FileEdekInfo;
import org.apache.hadoop.hdfs.server.namenode.ReencryptionUpdater.ReencryptionTask;
import org.apache.hadoop.hdfs.server.namenode.ReencryptionUpdater.ZoneSubmissionTracker;
import org.apache.hadoop.ipc.RetriableException;
import org.apache.hadoop.util.Daemon;
import org.apache.hadoop.util.StopWatch;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.io.IOException;
import java.security.GeneralSecurityException;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.concurrent.BlockingQueue;
import java.util.concurrent.Callable;
import java.util.concurrent.ExecutorCompletionService;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;
import java.util.concurrent.LinkedBlockingQueue;
import java.util.concurrent.ThreadPoolExecutor;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.atomic.AtomicInteger;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_REENCRYPT_BATCH_SIZE_DEFAULT;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_REENCRYPT_BATCH_SIZE_KEY;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_REENCRYPT_SLEEP_INTERVAL_DEFAULT;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_REENCRYPT_SLEEP_INTERVAL_KEY;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_REENCRYPT_THROTTLE_LIMIT_HANDLER_RATIO_DEFAULT;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_REENCRYPT_THROTTLE_LIMIT_HANDLER_RATIO_KEY;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_REENCRYPT_EDEK_THREADS_DEFAULT;
import static org.apache.hadoop.hdfs.DFSConfigKeys.DFS_NAMENODE_REENCRYPT_EDEK_THREADS_KEY;
/**
* Class for handling re-encrypt EDEK operations.
* <p>
* For each EZ, ReencryptionHandler walks the tree in a depth-first order,
* and submits batches of (files + existing edeks) as re-encryption tasks
* to a thread pool. Each thread in the pool then contacts the KMS to
* re-encrypt the edeks. ReencryptionUpdater tracks the tasks and updates
* file xattrs with the new edeks.
* <p>
* File renames are disabled in the EZ that's being re-encrypted. Newly created
* files will have new edeks, because the edek cache is drained upon the
* submission of a re-encryption command.
* <p>
* It is assumed only 1 ReencryptionHandler will be running, because:
* 1. The bottleneck of the entire re-encryption appears to be on the KMS.
* 2. Even with multiple handlers, since updater requires writelock and is
* single-threaded, the performance gain is limited.
* <p>
* This class uses the FSDirectory lock for synchronization.
*/
@InterfaceAudience.Private
public class ReencryptionHandler implements Runnable {
public static final Logger LOG =
LoggerFactory.getLogger(ReencryptionHandler.class);
// 2000 is based on buffer size = 512 * 1024, and SetXAttr op size is
// 100 - 200 bytes (depending on the xattr value).
// The buffer size is hard-coded, see outputBufferCapacity from QJM.
private static final int MAX_BATCH_SIZE_WITHOUT_FLOODING = 2000;
private final EncryptionZoneManager ezManager;
private final FSDirectory dir;
private final long interval;
private final int reencryptBatchSize;
private double throttleLimitHandlerRatio;
private final int reencryptThreadPoolSize;
// stopwatches for throttling
private final StopWatch throttleTimerAll = new StopWatch();
private final StopWatch throttleTimerLocked = new StopWatch();
private ExecutorCompletionService<ReencryptionTask> batchService;
private BlockingQueue<Runnable> taskQueue;
// protected by ReencryptionHandler object lock
private final Map<Long, ZoneSubmissionTracker> submissions = new HashMap<>();
// The current batch that the handler is working on. Handler is designed to
// be single-threaded, see class javadoc for more details.
private ReencryptionBatch currentBatch;
private final ReencryptionPendingInodeIdCollector traverser;
private final ReencryptionUpdater reencryptionUpdater;
private ExecutorService updaterExecutor;
// Vars for unit tests.
private volatile boolean shouldPauseForTesting = false;
private volatile int pauseAfterNthSubmission = 0;
/**
* Stop the re-encryption updater thread, as well as all EDEK re-encryption
* tasks submitted.
*/
void stopThreads() {
assert dir.hasWriteLock();
synchronized (this) {
for (ZoneSubmissionTracker zst : submissions.values()) {
zst.cancelAllTasks();
}
}
if (updaterExecutor != null) {
updaterExecutor.shutdownNow();
}
}
/**
* Start the re-encryption updater thread.
*/
void startUpdaterThread() {
updaterExecutor = Executors.newSingleThreadExecutor(
new ThreadFactoryBuilder().setDaemon(true)
.setNameFormat("reencryptionUpdaterThread #%d").build());
updaterExecutor.execute(reencryptionUpdater);
}
@VisibleForTesting
synchronized void pauseForTesting() {
shouldPauseForTesting = true;
LOG.info("Pausing re-encrypt handler for testing.");
notify();
}
@VisibleForTesting
synchronized void resumeForTesting() {
shouldPauseForTesting = false;
LOG.info("Resuming re-encrypt handler for testing.");
notify();
}
@VisibleForTesting
void pauseForTestingAfterNthSubmission(final int count) {
assert pauseAfterNthSubmission == 0;
pauseAfterNthSubmission = count;
}
@VisibleForTesting
void pauseUpdaterForTesting() {
reencryptionUpdater.pauseForTesting();
}
@VisibleForTesting
void resumeUpdaterForTesting() {
reencryptionUpdater.resumeForTesting();
}
@VisibleForTesting
void pauseForTestingAfterNthCheckpoint(final long zoneId, final int count) {
reencryptionUpdater.pauseForTestingAfterNthCheckpoint(zoneId, count);
}
ReencryptionHandler(final EncryptionZoneManager ezMgr,
final Configuration conf) {
this.ezManager = ezMgr;
Preconditions.checkNotNull(ezManager.getProvider(),
"No provider set, cannot re-encrypt");
this.dir = ezMgr.getFSDirectory();
this.interval =
conf.getTimeDuration(DFS_NAMENODE_REENCRYPT_SLEEP_INTERVAL_KEY,
DFS_NAMENODE_REENCRYPT_SLEEP_INTERVAL_DEFAULT,
TimeUnit.MILLISECONDS);
Preconditions.checkArgument(interval > 0,
DFS_NAMENODE_REENCRYPT_SLEEP_INTERVAL_KEY + " is not positive.");
this.reencryptBatchSize = conf.getInt(DFS_NAMENODE_REENCRYPT_BATCH_SIZE_KEY,
DFS_NAMENODE_REENCRYPT_BATCH_SIZE_DEFAULT);
Preconditions.checkArgument(reencryptBatchSize > 0,
DFS_NAMENODE_REENCRYPT_BATCH_SIZE_KEY + " is not positive.");
if (reencryptBatchSize > MAX_BATCH_SIZE_WITHOUT_FLOODING) {
LOG.warn("Re-encryption batch size is {}. It could cause edit log buffer "
+ "to be full and trigger a logSync within the writelock, greatly "
+ "impacting namenode throughput.", reencryptBatchSize);
}
this.throttleLimitHandlerRatio =
conf.getDouble(DFS_NAMENODE_REENCRYPT_THROTTLE_LIMIT_HANDLER_RATIO_KEY,
DFS_NAMENODE_REENCRYPT_THROTTLE_LIMIT_HANDLER_RATIO_DEFAULT);
LOG.info("Configured throttleLimitHandlerRatio={} for re-encryption",
throttleLimitHandlerRatio);
Preconditions.checkArgument(throttleLimitHandlerRatio > 0.0f,
DFS_NAMENODE_REENCRYPT_THROTTLE_LIMIT_HANDLER_RATIO_KEY
+ " is not positive.");
this.reencryptThreadPoolSize =
conf.getInt(DFS_NAMENODE_REENCRYPT_EDEK_THREADS_KEY,
DFS_NAMENODE_REENCRYPT_EDEK_THREADS_DEFAULT);
taskQueue = new LinkedBlockingQueue<>();
ThreadPoolExecutor threadPool =
new ThreadPoolExecutor(reencryptThreadPoolSize, reencryptThreadPoolSize,
60, TimeUnit.SECONDS, taskQueue, new Daemon.DaemonFactory() {
private final AtomicInteger ind = new AtomicInteger(0);
@Override
public Thread newThread(Runnable r) {
Thread t = super.newThread(r);
t.setName("reencryption edek Thread-" + ind.getAndIncrement());
return t;
}
}, new ThreadPoolExecutor.CallerRunsPolicy() {
@Override
public void rejectedExecution(Runnable runnable,
ThreadPoolExecutor e) {
LOG.info("Execution rejected, executing in current thread");
super.rejectedExecution(runnable, e);
}
});
threadPool.allowCoreThreadTimeOut(true);
this.batchService = new ExecutorCompletionService(threadPool);
reencryptionUpdater =
new ReencryptionUpdater(dir, batchService, this, conf);
currentBatch = new ReencryptionBatch(reencryptBatchSize);
traverser = new ReencryptionPendingInodeIdCollector(dir, this, conf);
}
ReencryptionStatus getReencryptionStatus() {
return ezManager.getReencryptionStatus();
}
void cancelZone(final long zoneId, final String zoneName) throws IOException {
assert dir.hasWriteLock();
final ZoneReencryptionStatus zs =
getReencryptionStatus().getZoneStatus(zoneId);
if (zs == null || zs.getState() == State.Completed) {
throw new IOException("Zone " + zoneName + " is not under re-encryption");
}
zs.cancel();
removeZoneTrackerStopTasks(zoneId);
}
void removeZone(final long zoneId) {
assert dir.hasWriteLock();
LOG.info("Removing zone {} from re-encryption.", zoneId);
removeZoneTrackerStopTasks(zoneId);
getReencryptionStatus().removeZone(zoneId);
}
synchronized private void removeZoneTrackerStopTasks(final long zoneId) {
final ZoneSubmissionTracker zst = submissions.get(zoneId);
if (zst != null) {
zst.cancelAllTasks();
submissions.remove(zoneId);
}
}
ZoneSubmissionTracker getTracker(final long zoneId) {
assert dir.hasReadLock();
return unprotectedGetTracker(zoneId);
}
/**
* Get the tracker without holding the FSDirectory lock.
* The submissions object is protected by object lock.
*/
synchronized ZoneSubmissionTracker unprotectedGetTracker(final long zoneId) {
return submissions.get(zoneId);
}
/**
* Add a dummy tracker (with 1 task that has 0 files to re-encrypt)
* for the zone. This is necessary to complete the re-encryption in case
* no file in the entire zone needs re-encryption at all. We cannot simply
* update zone status and set zone xattrs, because in the handler we only hold
* readlock, and setting xattrs requires upgrading to a writelock.
*
* @param zoneId
*/
void addDummyTracker(final long zoneId, ZoneSubmissionTracker zst) {
assert dir.hasReadLock();
if (zst == null) {
zst = new ZoneSubmissionTracker();
}
zst.setSubmissionDone();
final Future future = batchService.submit(
new EDEKReencryptCallable(zoneId, new ReencryptionBatch(), this));
zst.addTask(future);
synchronized (this) {
submissions.put(zoneId, zst);
}
}
/**
* Main loop. It takes at most 1 zone per scan, and executes until the zone
* is completed.
* {@link #reencryptEncryptionZone(long)}.
*/
@Override
public void run() {
LOG.info("Starting up re-encrypt thread with interval={} millisecond.",
interval);
while (true) {
try {
synchronized (this) {
wait(interval);
}
traverser.checkPauseForTesting();
} catch (InterruptedException ie) {
LOG.info("Re-encrypt handler interrupted. Exiting");
Thread.currentThread().interrupt();
return;
}
final Long zoneId;
dir.getFSNamesystem().readLock();
try {
zoneId = getReencryptionStatus().getNextUnprocessedZone();
if (zoneId == null) {
// empty queue.
continue;
}
LOG.info("Executing re-encrypt commands on zone {}. Current zones:{}",
zoneId, getReencryptionStatus());
getReencryptionStatus().markZoneStarted(zoneId);
resetSubmissionTracker(zoneId);
} finally {
dir.getFSNamesystem().readUnlock("reEncryptThread");
}
try {
reencryptEncryptionZone(zoneId);
} catch (RetriableException | SafeModeException re) {
LOG.info("Re-encryption caught exception, will retry", re);
getReencryptionStatus().markZoneForRetry(zoneId);
} catch (IOException ioe) {
LOG.warn("IOException caught when re-encrypting zone {}", zoneId, ioe);
} catch (InterruptedException ie) {
LOG.info("Re-encrypt handler interrupted. Exiting.");
Thread.currentThread().interrupt();
return;
} catch (Throwable t) {
LOG.error("Re-encrypt handler thread exiting. Exception caught when"
+ " re-encrypting zone {}.", zoneId, t);
return;
}
}
}
/**
* Re-encrypts a zone by recursively iterating all paths inside the zone,
* in lexicographic order.
* Files are re-encrypted, and subdirs are processed during iteration.
*
* @param zoneId the Zone's id.
* @throws IOException
* @throws InterruptedException
*/
void reencryptEncryptionZone(final long zoneId)
throws IOException, InterruptedException {
throttleTimerAll.reset().start();
throttleTimerLocked.reset();
final INode zoneNode;
final ZoneReencryptionStatus zs;
traverser.readLock();
try {
zoneNode = dir.getInode(zoneId);
// start re-encrypting the zone from the beginning
if (zoneNode == null) {
LOG.info("Directory with id {} removed during re-encrypt, skipping",
zoneId);
return;
}
if (!zoneNode.isDirectory()) {
LOG.info("Cannot re-encrypt directory with id {} because it's not a"
+ " directory.", zoneId);
return;
}
zs = getReencryptionStatus().getZoneStatus(zoneId);
assert zs != null;
// Only costly log FullPathName here once, and use id elsewhere.
LOG.info("Re-encrypting zone {}(id={})", zoneNode.getFullPathName(),
zoneId);
if (zs.getLastCheckpointFile() == null) {
// new re-encryption
traverser.traverseDir(zoneNode.asDirectory(), zoneId,
HdfsFileStatus.EMPTY_NAME,
new ZoneTraverseInfo(zs.getEzKeyVersionName()));
} else {
// resuming from a past re-encryption
restoreFromLastProcessedFile(zoneId, zs);
}
// save the last batch and mark complete
traverser.submitCurrentBatch(zoneId);
LOG.info("Submission completed of zone {} for re-encryption.", zoneId);
reencryptionUpdater.markZoneSubmissionDone(zoneId);
} finally {
traverser.readUnlock();
}
}
/**
* Reset the zone submission tracker for re-encryption.
* @param zoneId
*/
synchronized private void resetSubmissionTracker(final long zoneId) {
ZoneSubmissionTracker zst = submissions.get(zoneId);
if (zst == null) {
zst = new ZoneSubmissionTracker();
submissions.put(zoneId, zst);
} else {
zst.reset();
}
}
List<XAttr> completeReencryption(final INode zoneNode) throws IOException {
assert dir.hasWriteLock();
assert dir.getFSNamesystem().hasWriteLock();
final Long zoneId = zoneNode.getId();
ZoneReencryptionStatus zs = getReencryptionStatus().getZoneStatus(zoneId);
assert zs != null;
LOG.info("Re-encryption completed on zone {}. Re-encrypted {} files,"
+ " failures encountered: {}.", zoneNode.getFullPathName(),
zs.getFilesReencrypted(), zs.getNumReencryptionFailures());
synchronized (this) {
submissions.remove(zoneId);
}
return FSDirEncryptionZoneOp
.updateReencryptionFinish(dir, INodesInPath.fromINode(zoneNode), zs);
}
/**
* Restore the re-encryption from the progress inside ReencryptionStatus.
* This means start from exactly the lastProcessedFile (LPF), skipping all
* earlier paths in lexicographic order. Lexicographically-later directories
* on the LPF parent paths are added to subdirs.
*/
private void restoreFromLastProcessedFile(final long zoneId,
final ZoneReencryptionStatus zs)
throws IOException, InterruptedException {
final INodeDirectory parent;
final byte[] startAfter;
final INodesInPath lpfIIP =
dir.getINodesInPath(zs.getLastCheckpointFile(), FSDirectory.DirOp.READ);
parent = lpfIIP.getLastINode().getParent();
startAfter = lpfIIP.getLastINode().getLocalNameBytes();
traverser.traverseDir(parent, zoneId, startAfter,
new ZoneTraverseInfo(zs.getEzKeyVersionName()));
}
final class ReencryptionBatch {
// First file's path, for logging purpose.
private String firstFilePath;
private final List<FileEdekInfo> batch;
ReencryptionBatch() {
this(reencryptBatchSize);
}
ReencryptionBatch(int initialCapacity) {
batch = new ArrayList<>(initialCapacity);
}
void add(final INodeFile inode) throws IOException {
assert dir.hasReadLock();
Preconditions.checkNotNull(inode, "INodeFile is null");
if (batch.isEmpty()) {
firstFilePath = inode.getFullPathName();
}
batch.add(new FileEdekInfo(dir, inode));
}
String getFirstFilePath() {
return firstFilePath;
}
boolean isEmpty() {
return batch.isEmpty();
}
int size() {
return batch.size();
}
void clear() {
batch.clear();
}
List<FileEdekInfo> getBatch() {
return batch;
}
}
/**
* Simply contacts the KMS for re-encryption. No NN locks held.
*/
private static class EDEKReencryptCallable
implements Callable<ReencryptionTask> {
private final long zoneNodeId;
private final ReencryptionBatch batch;
private final ReencryptionHandler handler;
EDEKReencryptCallable(final long zoneId,
final ReencryptionBatch currentBatch, final ReencryptionHandler rh) {
zoneNodeId = zoneId;
batch = currentBatch;
handler = rh;
}
@Override
public ReencryptionTask call() {
LOG.info("Processing batched re-encryption for zone {}, batch size {},"
+ " start:{}", zoneNodeId, batch.size(), batch.getFirstFilePath());
if (batch.isEmpty()) {
return new ReencryptionTask(zoneNodeId, 0, batch);
}
final StopWatch kmsSW = new StopWatch().start();
int numFailures = 0;
String result = "Completed";
if (!reencryptEdeks()) {
numFailures += batch.size();
result = "Failed to";
}
LOG.info("{} re-encrypting one batch of {} edeks from KMS,"
+ " time consumed: {}, start: {}.", result,
batch.size(), kmsSW.stop(), batch.getFirstFilePath());
return new ReencryptionTask(zoneNodeId, numFailures, batch);
}
private boolean reencryptEdeks() {
// communicate with the kms out of lock
final List<EncryptedKeyVersion> edeks = new ArrayList<>(batch.size());
for (FileEdekInfo entry : batch.getBatch()) {
edeks.add(entry.getExistingEdek());
}
// provider already has LoadBalancingKMSClientProvider's reties. It that
// fails, just fail this callable.
try {
handler.ezManager.getProvider().reencryptEncryptedKeys(edeks);
EncryptionFaultInjector.getInstance().reencryptEncryptedKeys();
} catch (GeneralSecurityException | IOException ex) {
LOG.warn("Failed to re-encrypt one batch of {} edeks, start:{}",
batch.size(), batch.getFirstFilePath(), ex);
return false;
}
int i = 0;
for (FileEdekInfo entry : batch.getBatch()) {
assert i < edeks.size();
entry.setEdek(edeks.get(i++));
}
return true;
}
}
/**
* Called when a new zone is submitted for re-encryption. This will interrupt
* the background thread if it's waiting for the next
* DFS_NAMENODE_REENCRYPT_SLEEP_INTERVAL_KEY.
*/
synchronized void notifyNewSubmission() {
LOG.debug("Notifying handler for new re-encryption command.");
this.notify();
}
public ReencryptionPendingInodeIdCollector getTraverser() {
return traverser;
}
/**
* ReencryptionPendingInodeIdCollector which throttle based on configured
* throttle ratio.
*/
class ReencryptionPendingInodeIdCollector extends FSTreeTraverser {
private final ReencryptionHandler reencryptionHandler;
ReencryptionPendingInodeIdCollector(FSDirectory dir,
ReencryptionHandler rHandler, Configuration conf) {
super(dir, conf);
this.reencryptionHandler = rHandler;
}
@Override
protected void checkPauseForTesting()
throws InterruptedException {
assert !dir.hasReadLock();
assert !dir.getFSNamesystem().hasReadLock();
while (shouldPauseForTesting) {
LOG.info("Sleeping in the re-encrypt handler for unit test.");
synchronized (reencryptionHandler) {
if (shouldPauseForTesting) {
reencryptionHandler.wait(30000);
}
}
LOG.info("Continuing re-encrypt handler after pausing.");
}
}
/**
* Process an Inode for re-encryption. Add to current batch if it's a file,
* no-op otherwise.
*
* @param inode
* the inode
* @return true if inode is added to currentBatch and should be
* re-encrypted. false otherwise: could be inode is not a file, or
* inode's edek's key version is not changed.
* @throws IOException
* @throws InterruptedException
*/
@Override
public boolean processFileInode(INode inode, TraverseInfo traverseInfo)
throws IOException, InterruptedException {
assert dir.hasReadLock();
if (LOG.isTraceEnabled()) {
LOG.trace("Processing {} for re-encryption", inode.getFullPathName());
}
if (!inode.isFile()) {
return false;
}
FileEncryptionInfo feInfo = FSDirEncryptionZoneOp.getFileEncryptionInfo(
dir, INodesInPath.fromINode(inode));
if (feInfo == null) {
LOG.warn("File {} skipped re-encryption because it is not encrypted! "
+ "This is very likely a bug.", inode.getId());
return false;
}
if (traverseInfo instanceof ZoneTraverseInfo
&& ((ZoneTraverseInfo) traverseInfo).getEzKeyVerName().equals(
feInfo.getEzKeyVersionName())) {
if (LOG.isDebugEnabled()) {
LOG.debug("File {} skipped re-encryption because edek's key version"
+ " name is not changed.", inode.getFullPathName());
}
return false;
}
currentBatch.add(inode.asFile());
return true;
}
/**
* Check whether zone is ready for re-encryption. Throws IOE if it's not. 1.
* If EZ is deleted. 2. if the re-encryption is canceled. 3. If NN is not
* active or is in safe mode.
*
* @throws IOException
* if zone does not exist / is cancelled, or if NN is not ready
* for write.
*/
@Override
protected void checkINodeReady(long zoneId) throws IOException {
final ZoneReencryptionStatus zs = getReencryptionStatus().getZoneStatus(
zoneId);
if (zs == null) {
throw new IOException("Zone " + zoneId + " status cannot be found.");
}
if (zs.isCanceled()) {
throw new IOException("Re-encryption is canceled for zone " + zoneId);
}
dir.getFSNamesystem().checkNameNodeSafeMode(
"NN is in safe mode, cannot re-encrypt.");
// re-encryption should be cancelled when NN goes to standby. Just
// double checking for sanity.
dir.getFSNamesystem().checkOperation(NameNode.OperationCategory.WRITE);
}
/**
* Submit the current batch to the thread pool.
*
* @param zoneId
* Id of the EZ INode
* @throws IOException
* @throws InterruptedException
*/
@Override
protected void submitCurrentBatch(final Long zoneId) throws IOException,
InterruptedException {
if (currentBatch.isEmpty()) {
return;
}
ZoneSubmissionTracker zst;
synchronized (ReencryptionHandler.this) {
zst = submissions.get(zoneId);
if (zst == null) {
zst = new ZoneSubmissionTracker();
submissions.put(zoneId, zst);
}
Future future = batchService.submit(new EDEKReencryptCallable(zoneId,
currentBatch, reencryptionHandler));
zst.addTask(future);
}
LOG.info("Submitted batch (start:{}, size:{}) of zone {} to re-encrypt.",
currentBatch.getFirstFilePath(), currentBatch.size(), zoneId);
currentBatch = new ReencryptionBatch(reencryptBatchSize);
// flip the pause flag if this is nth submission.
// The actual pause need to happen outside of the lock.
if (pauseAfterNthSubmission > 0) {
if (--pauseAfterNthSubmission == 0) {
shouldPauseForTesting = true;
}
}
}
/**
* Throttles the ReencryptionHandler in 3 aspects:
* 1. Prevents generating more Callables than the CPU could possibly
* handle.
* 2. Prevents generating more Callables than the ReencryptionUpdater
* can handle, under its own throttling.
* 3. Prevents contending FSN/FSD read locks. This is done based
* on the DFS_NAMENODE_REENCRYPT_THROTTLE_LIMIT_RATIO_KEY configuration.
* <p>
* Item 1 and 2 are to control NN heap usage.
*
* @throws InterruptedException
*/
@VisibleForTesting
@Override
protected void throttle() throws InterruptedException {
assert !dir.hasReadLock();
assert !dir.getFSNamesystem().hasReadLock();
final int numCores = Runtime.getRuntime().availableProcessors();
if (taskQueue.size() >= numCores) {
LOG.debug("Re-encryption handler throttling because queue size {} is"
+ "larger than number of cores {}", taskQueue.size(), numCores);
while (taskQueue.size() >= numCores) {
Thread.sleep(100);
}
}
// 2. if tasks are piling up on the updater, don't create new callables
// until the queue size goes down.
final int maxTasksPiled = Runtime.getRuntime().availableProcessors() * 2;
int numTasks = numTasksSubmitted();
if (numTasks >= maxTasksPiled) {
LOG.debug("Re-encryption handler throttling because total tasks pending"
+ " re-encryption updater is {}", numTasks);
while (numTasks >= maxTasksPiled) {
Thread.sleep(500);
numTasks = numTasksSubmitted();
}
}
// 3.
if (throttleLimitHandlerRatio >= 1.0) {
return;
}
final long expect = (long) (throttleTimerAll.now(TimeUnit.MILLISECONDS)
* throttleLimitHandlerRatio);
final long actual = throttleTimerLocked.now(TimeUnit.MILLISECONDS);
if (LOG.isDebugEnabled()) {
LOG.debug("Re-encryption handler throttling expect: {}, actual: {},"
+ " throttleTimerAll:{}", expect, actual,
throttleTimerAll.now(TimeUnit.MILLISECONDS));
}
if (expect - actual < 0) {
// in case throttleLimitHandlerRatio is very small, expect will be 0.
// so sleepMs should not be calculated from expect, to really meet the
// ratio. e.g. if ratio is 0.001, expect = 0 and actual = 1, sleepMs
// should be 1000 - throttleTimerAll.now()
final long sleepMs = (long) (actual / throttleLimitHandlerRatio)
- throttleTimerAll.now(TimeUnit.MILLISECONDS);
LOG.debug("Throttling re-encryption, sleeping for {} ms", sleepMs);
Thread.sleep(sleepMs);
}
throttleTimerAll.reset().start();
throttleTimerLocked.reset();
}
private int numTasksSubmitted() {
int ret = 0;
synchronized (ReencryptionHandler.this) {
for (ZoneSubmissionTracker zst : submissions.values()) {
ret += zst.getTasks().size();
}
}
return ret;
}
@Override
public boolean shouldSubmitCurrentBatch() {
return currentBatch.size() >= reencryptBatchSize;
}
@Override
public boolean canTraverseDir(INode inode) throws IOException {
if (ezManager.isEncryptionZoneRoot(inode, inode.getFullPathName())) {
// nested EZ, ignore.
LOG.info("{}({}) is a nested EZ, skipping for re-encryption",
inode.getFullPathName(), inode.getId());
return false;
}
return true;
}
@Override
protected void readLock() {
super.readLock();
throttleTimerLocked.start();
}
@Override
protected void readUnlock() {
super.readUnlock();
throttleTimerLocked.stop();
}
}
private static class ZoneTraverseInfo extends TraverseInfo {
private String ezKeyVerName;
ZoneTraverseInfo(String ezKeyVerName) {
this.ezKeyVerName = ezKeyVerName;
}
public String getEzKeyVerName() {
return ezKeyVerName;
}
}
}
相关信息
相关文章
hadoop AclEntryStatusFormat 源码
0
赞
热门推荐
-
2、 - 优质文章
-
3、 gate.io
-
7、 golang
-
9、 openharmony
-
10、 Vue中input框自动聚焦