org.apache.hadoop.hdfs.server.blockmanagement.DecommissionManager Maven / Gradle / Ivy
The newest version!
/**
* 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.blockmanagement;
import java.util.AbstractList;
import java.util.Iterator;
import java.util.LinkedList;
import java.util.List;
import java.util.Map;
import java.util.Queue;
import java.util.TreeMap;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;
import java.util.concurrent.ScheduledExecutorService;
import java.util.concurrent.TimeUnit;
import com.google.common.annotations.VisibleForTesting;
import com.google.common.base.Preconditions;
import com.google.common.util.concurrent.ThreadFactoryBuilder;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.hdfs.DFSConfigKeys;
import org.apache.hadoop.hdfs.protocol.Block;
import org.apache.hadoop.hdfs.protocol.DatanodeID;
import org.apache.hadoop.hdfs.server.namenode.Namesystem;
import org.apache.hadoop.hdfs.util.CyclicIteration;
import org.apache.hadoop.util.ChunkedArrayList;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import static com.google.common.base.Preconditions.checkArgument;
import static org.apache.hadoop.util.Time.monotonicNow;
/**
* Manages datanode decommissioning. A background monitor thread
* periodically checks the status of datanodes that are in-progress of
* decommissioning.
*
* A datanode can be decommissioned in a few situations:
*
* - If a DN is dead, it is decommissioned immediately.
* - If a DN is alive, it is decommissioned after all of its blocks
* are sufficiently replicated. Merely under-replicated blocks do not
* block decommissioning as long as they are above a replication
* threshold.
*
* In the second case, the datanode transitions to a
* decommission-in-progress state and is tracked by the monitor thread. The
* monitor periodically scans through the list of insufficiently replicated
* blocks on these datanodes to
* determine if they can be decommissioned. The monitor also prunes this list
* as blocks become replicated, so monitor scans will become more efficient
* over time.
*
* Decommission-in-progress nodes that become dead do not progress to
* decommissioned until they become live again. This prevents potential
* durability loss for singly-replicated blocks (see HDFS-6791).
*
* This class depends on the FSNamesystem lock for synchronization.
*/
@InterfaceAudience.Private
public class DecommissionManager {
private static final Logger LOG = LoggerFactory.getLogger(DecommissionManager
.class);
private final Namesystem namesystem;
private final BlockManager blockManager;
private final HeartbeatManager hbManager;
private final ScheduledExecutorService executor;
/**
* Map containing the decommission-in-progress datanodes that are being
* tracked so they can be be marked as decommissioned.
*
* This holds a set of references to the under-replicated blocks on the DN at
* the time the DN is added to the map, i.e. the blocks that are preventing
* the node from being marked as decommissioned. During a monitor tick, this
* list is pruned as blocks becomes replicated.
*
* Note also that the reference to the list of under-replicated blocks
* will be null on initial add
*
* However, this map can become out-of-date since it is not updated by block
* reports or other events. Before being finally marking as decommissioned,
* another check is done with the actual block map.
*/
private final TreeMap>
decomNodeBlocks;
/**
* Tracking a node in decomNodeBlocks consumes additional memory. To limit
* the impact on NN memory consumption, we limit the number of nodes in
* decomNodeBlocks. Additional nodes wait in pendingNodes.
*/
private final Queue pendingNodes;
private Monitor monitor = null;
DecommissionManager(final Namesystem namesystem,
final BlockManager blockManager, final HeartbeatManager hbManager) {
this.namesystem = namesystem;
this.blockManager = blockManager;
this.hbManager = hbManager;
executor = Executors.newScheduledThreadPool(1,
new ThreadFactoryBuilder().setNameFormat("DecommissionMonitor-%d")
.setDaemon(true).build());
decomNodeBlocks = new TreeMap<>();
pendingNodes = new LinkedList<>();
}
/**
* Start the decommission monitor thread.
* @param conf
*/
void activate(Configuration conf) {
final int intervalSecs =
conf.getInt(DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_INTERVAL_KEY,
DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_INTERVAL_DEFAULT);
checkArgument(intervalSecs >= 0, "Cannot set a negative " +
"value for " + DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_INTERVAL_KEY);
// By default, the new configuration key overrides the deprecated one.
// No # node limit is set.
int blocksPerInterval = conf.getInt(
DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_BLOCKS_PER_INTERVAL_KEY,
DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_BLOCKS_PER_INTERVAL_DEFAULT);
int nodesPerInterval = Integer.MAX_VALUE;
// If the expected key isn't present and the deprecated one is,
// use the deprecated one into the new one. This overrides the
// default.
//
// Also print a deprecation warning.
final String deprecatedKey =
"dfs.namenode.decommission.nodes.per.interval";
final String strNodes = conf.get(deprecatedKey);
if (strNodes != null) {
nodesPerInterval = Integer.parseInt(strNodes);
blocksPerInterval = Integer.MAX_VALUE;
LOG.warn("Using deprecated configuration key {} value of {}.",
deprecatedKey, nodesPerInterval);
LOG.warn("Please update your configuration to use {} instead.",
DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_BLOCKS_PER_INTERVAL_KEY);
}
checkArgument(blocksPerInterval > 0,
"Must set a positive value for "
+ DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_BLOCKS_PER_INTERVAL_KEY);
final int maxConcurrentTrackedNodes = conf.getInt(
DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_MAX_CONCURRENT_TRACKED_NODES,
DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_MAX_CONCURRENT_TRACKED_NODES_DEFAULT);
checkArgument(maxConcurrentTrackedNodes >= 0, "Cannot set a negative " +
"value for "
+ DFSConfigKeys.DFS_NAMENODE_DECOMMISSION_MAX_CONCURRENT_TRACKED_NODES);
monitor = new Monitor(blocksPerInterval,
nodesPerInterval, maxConcurrentTrackedNodes);
executor.scheduleAtFixedRate(monitor, intervalSecs, intervalSecs,
TimeUnit.SECONDS);
LOG.debug("Activating DecommissionManager with interval {} seconds, " +
"{} max blocks per interval, {} max nodes per interval, " +
"{} max concurrently tracked nodes.", intervalSecs,
blocksPerInterval, nodesPerInterval, maxConcurrentTrackedNodes);
}
/**
* Stop the decommission monitor thread, waiting briefly for it to terminate.
*/
void close() {
executor.shutdownNow();
try {
executor.awaitTermination(3000, TimeUnit.MILLISECONDS);
} catch (InterruptedException e) {}
}
/**
* Start decommissioning the specified datanode.
* @param node
*/
@VisibleForTesting
public void startDecommission(DatanodeDescriptor node) {
if (!node.isDecommissionInProgress() && !node.isDecommissioned()) {
// Update DN stats maintained by HeartbeatManager
hbManager.startDecommission(node);
// hbManager.startDecommission will set dead node to decommissioned.
if (node.isDecommissionInProgress()) {
for (DatanodeStorageInfo storage : node.getStorageInfos()) {
LOG.info("Starting decommission of {} {} with {} blocks",
node, storage, storage.numBlocks());
}
node.decommissioningStatus.setStartTime(monotonicNow());
pendingNodes.add(node);
}
} else {
LOG.trace("startDecommission: Node {} in {}, nothing to do." +
node, node.getAdminState());
}
}
/**
* Stop decommissioning the specified datanode.
* @param node
*/
@VisibleForTesting
public void stopDecommission(DatanodeDescriptor node) {
if (node.isDecommissionInProgress() || node.isDecommissioned()) {
// Update DN stats maintained by HeartbeatManager
hbManager.stopDecommission(node);
// Over-replicated blocks will be detected and processed when
// the dead node comes back and send in its full block report.
if (node.isAlive) {
blockManager.processOverReplicatedBlocksOnReCommission(node);
}
// Remove from tracking in DecommissionManager
pendingNodes.remove(node);
decomNodeBlocks.remove(node);
} else {
LOG.trace("stopDecommission: Node {} in {}, nothing to do." +
node, node.getAdminState());
}
}
private void setDecommissioned(DatanodeDescriptor dn) {
dn.setDecommissioned();
LOG.info("Decommissioning complete for node {}", dn);
}
/**
* Checks whether a block is sufficiently replicated for decommissioning.
* Full-strength replication is not always necessary, hence "sufficient".
* @return true if sufficient, else false.
*/
private boolean isSufficientlyReplicated(BlockInfoContiguous block,
BlockCollection bc,
NumberReplicas numberReplicas) {
final int numExpected = bc.getBlockReplication();
final int numLive = numberReplicas.liveReplicas();
if (numLive >= numExpected
&& blockManager.isPlacementPolicySatisfied(block)) {
// Block has enough replica, skip
LOG.trace("Block {} does not need replication.", block);
return true;
}
// Block is under-replicated
LOG.trace("Block {} numExpected={}, numLive={}", block, numExpected,
numLive);
if (numExpected > numLive) {
if (bc.isUnderConstruction() && block.equals(bc.getLastBlock())) {
// Can decom a UC block as long as there will still be minReplicas
if (numLive >= blockManager.minReplication) {
LOG.trace("UC block {} sufficiently-replicated since numLive ({}) "
+ ">= minR ({})", block, numLive, blockManager.minReplication);
return true;
} else {
LOG.trace("UC block {} insufficiently-replicated since numLive "
+ "({}) < minR ({})", block, numLive,
blockManager.minReplication);
}
} else {
// Can decom a non-UC as long as the default replication is met
if (numLive >= blockManager.defaultReplication) {
return true;
}
}
}
return false;
}
private static void logBlockReplicationInfo(Block block, BlockCollection bc,
DatanodeDescriptor srcNode, NumberReplicas num,
Iterable storages) {
int curReplicas = num.liveReplicas();
int curExpectedReplicas = bc.getBlockReplication();
StringBuilder nodeList = new StringBuilder();
for (DatanodeStorageInfo storage : storages) {
final DatanodeDescriptor node = storage.getDatanodeDescriptor();
nodeList.append(node);
nodeList.append(" ");
}
LOG.info("Block: " + block + ", Expected Replicas: "
+ curExpectedReplicas + ", live replicas: " + curReplicas
+ ", corrupt replicas: " + num.corruptReplicas()
+ ", decommissioned replicas: " + num.decommissioned()
+ ", decommissioning replicas: " + num.decommissioning()
+ ", excess replicas: " + num.excessReplicas()
+ ", Is Open File: " + bc.isUnderConstruction()
+ ", Datanodes having this block: " + nodeList + ", Current Datanode: "
+ srcNode + ", Is current datanode decommissioning: "
+ srcNode.isDecommissionInProgress());
}
@VisibleForTesting
public int getNumPendingNodes() {
return pendingNodes.size();
}
@VisibleForTesting
public int getNumTrackedNodes() {
return decomNodeBlocks.size();
}
@VisibleForTesting
public int getNumNodesChecked() {
return monitor.numNodesChecked;
}
/**
* Checks to see if DNs have finished decommissioning.
*
* Since this is done while holding the namesystem lock,
* the amount of work per monitor tick is limited.
*/
private class Monitor implements Runnable {
/**
* The maximum number of blocks to check per tick.
*/
private final int numBlocksPerCheck;
/**
* The maximum number of nodes to check per tick.
*/
private final int numNodesPerCheck;
/**
* The maximum number of nodes to track in decomNodeBlocks. A value of 0
* means no limit.
*/
private final int maxConcurrentTrackedNodes;
/**
* The number of blocks that have been checked on this tick.
*/
private int numBlocksChecked = 0;
/**
* The number of blocks checked after (re)holding lock.
*/
private int numBlocksCheckedPerLock = 0;
/**
* The number of nodes that have been checked on this tick. Used for
* testing.
*/
private int numNodesChecked = 0;
/**
* The last datanode in decomNodeBlocks that we've processed
*/
private DatanodeDescriptor iterkey = new DatanodeDescriptor(new
DatanodeID("", "", "", 0, 0, 0, 0));
Monitor(int numBlocksPerCheck, int numNodesPerCheck, int
maxConcurrentTrackedNodes) {
this.numBlocksPerCheck = numBlocksPerCheck;
this.numNodesPerCheck = numNodesPerCheck;
this.maxConcurrentTrackedNodes = maxConcurrentTrackedNodes;
}
private boolean exceededNumBlocksPerCheck() {
LOG.trace("Processed {} blocks so far this tick", numBlocksChecked);
return numBlocksChecked >= numBlocksPerCheck;
}
@Deprecated
private boolean exceededNumNodesPerCheck() {
LOG.trace("Processed {} nodes so far this tick", numNodesChecked);
return numNodesChecked >= numNodesPerCheck;
}
@Override
public void run() {
if (!namesystem.isRunning()) {
LOG.info("Namesystem is not running, skipping decommissioning checks"
+ ".");
return;
}
// Reset the checked count at beginning of each iteration
numBlocksChecked = 0;
numBlocksCheckedPerLock = 0;
numNodesChecked = 0;
// Check decom progress
namesystem.writeLock();
try {
processPendingNodes();
check();
} finally {
namesystem.writeUnlock();
}
if (numBlocksChecked + numNodesChecked > 0) {
LOG.info("Checked {} blocks and {} nodes this tick", numBlocksChecked,
numNodesChecked);
}
}
/**
* Pop datanodes off the pending list and into decomNodeBlocks,
* subject to the maxConcurrentTrackedNodes limit.
*/
private void processPendingNodes() {
while (!pendingNodes.isEmpty() &&
(maxConcurrentTrackedNodes == 0 ||
decomNodeBlocks.size() < maxConcurrentTrackedNodes)) {
decomNodeBlocks.put(pendingNodes.poll(), null);
}
}
private void check() {
final Iterator>>
it = new CyclicIteration<>(decomNodeBlocks, iterkey).iterator();
final LinkedList toRemove = new LinkedList<>();
while (it.hasNext() && !exceededNumBlocksPerCheck()
&& !exceededNumNodesPerCheck() && namesystem.isRunning()) {
numNodesChecked++;
final Map.Entry>
entry = it.next();
final DatanodeDescriptor dn = entry.getKey();
AbstractList blocks = entry.getValue();
boolean fullScan = false;
if (blocks == null) {
// This is a newly added datanode, run through its list to schedule
// under-replicated blocks for replication and collect the blocks
// that are insufficiently replicated for further tracking
LOG.debug("Newly-added node {}, doing full scan to find " +
"insufficiently-replicated blocks.", dn);
blocks = handleInsufficientlyReplicated(dn);
decomNodeBlocks.put(dn, blocks);
fullScan = true;
} else {
// This is a known datanode, check if its # of insufficiently
// replicated blocks has dropped to zero and if it can be decommed
LOG.debug("Processing decommission-in-progress node {}", dn);
pruneSufficientlyReplicated(dn, blocks);
}
if (blocks.size() == 0) {
if (!fullScan) {
// If we didn't just do a full scan, need to re-check with the
// full block map.
//
// We've replicated all the known insufficiently replicated
// blocks. Re-check with the full block map before finally
// marking the datanode as decommissioned
LOG.debug("Node {} has finished replicating current set of "
+ "blocks, checking with the full block map.", dn);
blocks = handleInsufficientlyReplicated(dn);
decomNodeBlocks.put(dn, blocks);
}
// If the full scan is clean AND the node liveness is okay,
// we can finally mark as decommissioned.
final boolean isHealthy =
blockManager.isNodeHealthyForDecommission(dn);
if (blocks.size() == 0 && isHealthy) {
setDecommissioned(dn);
toRemove.add(dn);
LOG.debug("Node {} is sufficiently replicated and healthy, "
+ "marked as decommissioned.", dn);
} else {
if (LOG.isDebugEnabled()) {
StringBuilder b = new StringBuilder("Node {} ");
if (isHealthy) {
b.append("is ");
} else {
b.append("isn't ");
}
b.append("healthy and still needs to replicate {} more blocks," +
" decommissioning is still in progress.");
LOG.debug(b.toString(), dn, blocks.size());
}
}
} else {
LOG.debug("Node {} still has {} blocks to replicate "
+ "before it is a candidate to finish decommissioning.",
dn, blocks.size());
}
iterkey = dn;
}
// Remove the datanodes that are decommissioned
for (DatanodeDescriptor dn : toRemove) {
Preconditions.checkState(dn.isDecommissioned(),
"Removing a node that is not yet decommissioned!");
decomNodeBlocks.remove(dn);
}
}
/**
* Removes sufficiently replicated blocks from the block list of a
* datanode.
*/
private void pruneSufficientlyReplicated(final DatanodeDescriptor datanode,
AbstractList blocks) {
processBlocksForDecomInternal(datanode, blocks.iterator(), null, true);
}
/**
* Returns a list of blocks on a datanode that are insufficiently
* replicated, i.e. are under-replicated enough to prevent decommission.
*
* As part of this, it also schedules replication work for
* any under-replicated blocks.
*
* @param datanode
* @return List of insufficiently replicated blocks
*/
private AbstractList handleInsufficientlyReplicated(
final DatanodeDescriptor datanode) {
AbstractList insufficient = new ChunkedArrayList<>();
processBlocksForDecomInternal(datanode, datanode.getBlockIterator(),
insufficient, false);
return insufficient;
}
/**
* Used while checking if decommission-in-progress datanodes can be marked
* as decommissioned. Combines shared logic of
* pruneSufficientlyReplicated and handleInsufficientlyReplicated.
*
* @param datanode Datanode
* @param it Iterator over the blocks on the
* datanode
* @param insufficientlyReplicated Return parameter. If it's not null,
* will contain the insufficiently
* replicated-blocks from the list.
* @param pruneSufficientlyReplicated whether to remove sufficiently
* replicated blocks from the iterator
* @return true if there are under-replicated blocks in the provided block
* iterator, else false.
*/
private void processBlocksForDecomInternal(
final DatanodeDescriptor datanode,
final Iterator it,
final List insufficientlyReplicated,
boolean pruneSufficientlyReplicated) {
boolean firstReplicationLog = true;
int underReplicatedBlocks = 0;
int decommissionOnlyReplicas = 0;
int underReplicatedInOpenFiles = 0;
while (it.hasNext()) {
if (insufficientlyReplicated == null
&& numBlocksCheckedPerLock >= numBlocksPerCheck) {
// During fullscan insufficientlyReplicated will NOT be null, iterator
// will be DN's iterator. So should not yield lock, otherwise
// ConcurrentModificationException could occur.
// Once the fullscan done, iterator will be a copy. So can yield the
// lock.
// Yielding is required in case of block number is greater than the
// configured per-iteration-limit.
namesystem.writeUnlock();
try {
LOG.debug("Yielded lock during decommission check");
Thread.sleep(0, 500);
} catch (InterruptedException ignored) {
return;
}
//reset
numBlocksCheckedPerLock = 0;
namesystem.writeLock();
}
numBlocksChecked++;
numBlocksCheckedPerLock++;
final BlockInfoContiguous block = it.next();
// Remove the block from the list if it's no longer in the block map,
// e.g. the containing file has been deleted
if (blockManager.blocksMap.getStoredBlock(block) == null) {
LOG.trace("Removing unknown block {}", block);
it.remove();
continue;
}
BlockCollection bc = blockManager.blocksMap.getBlockCollection(block);
if (bc == null) {
// Orphan block, will be invalidated eventually. Skip.
continue;
}
final NumberReplicas num = blockManager.countNodes(block);
final int liveReplicas = num.liveReplicas();
final int curReplicas = liveReplicas;
// Schedule under-replicated blocks for replication if not already
// pending
if (blockManager.isNeededReplication(block, bc.getBlockReplication(),
liveReplicas)) {
if (!blockManager.neededReplications.contains(block) &&
blockManager.pendingReplications.getNumReplicas(block) == 0 &&
namesystem.isPopulatingReplQueues()) {
// Process these blocks only when active NN is out of safe mode.
blockManager.neededReplications.add(block,
curReplicas,
num.decommissionedAndDecommissioning(),
bc.getBlockReplication());
}
}
// Even if the block is under-replicated,
// it doesn't block decommission if it's sufficiently replicated
if (isSufficientlyReplicated(block, bc, num)) {
if (pruneSufficientlyReplicated) {
it.remove();
}
continue;
}
// We've found an insufficiently replicated block.
if (insufficientlyReplicated != null) {
insufficientlyReplicated.add(block);
}
// Log if this is our first time through
if (firstReplicationLog) {
logBlockReplicationInfo(block, bc, datanode, num,
blockManager.blocksMap.getStorages(block));
firstReplicationLog = false;
}
// Update various counts
underReplicatedBlocks++;
if (bc.isUnderConstruction()) {
underReplicatedInOpenFiles++;
}
if ((curReplicas == 0) && (num.decommissionedAndDecommissioning() > 0)) {
decommissionOnlyReplicas++;
}
}
datanode.decommissioningStatus.set(underReplicatedBlocks,
decommissionOnlyReplicas,
underReplicatedInOpenFiles);
}
}
@VisibleForTesting
void runMonitor() throws ExecutionException, InterruptedException {
Future f = executor.submit(monitor);
f.get();
}
}