use of org.opensearch.common.unit.TimeValue in project OpenSearch by opensearch-project.
the class RecoverySourceHandler method recoverToTarget.
/**
* performs the recovery from the local engine to the target
*/
public void recoverToTarget(ActionListener<RecoveryResponse> listener) {
addListener(listener);
final Closeable releaseResources = () -> IOUtils.close(resources);
try {
cancellableThreads.setOnCancel((reason, beforeCancelEx) -> {
final RuntimeException e;
if (shard.state() == IndexShardState.CLOSED) {
// check if the shard got closed on us
e = new IndexShardClosedException(shard.shardId(), "shard is closed and recovery was canceled reason [" + reason + "]");
} else {
e = new CancellableThreads.ExecutionCancelledException("recovery was canceled reason [" + reason + "]");
}
if (beforeCancelEx != null) {
e.addSuppressed(beforeCancelEx);
}
IOUtils.closeWhileHandlingException(releaseResources, () -> future.onFailure(e));
throw e;
});
final Consumer<Exception> onFailure = e -> {
assert Transports.assertNotTransportThread(RecoverySourceHandler.this + "[onFailure]");
IOUtils.closeWhileHandlingException(releaseResources, () -> future.onFailure(e));
};
final SetOnce<RetentionLease> retentionLeaseRef = new SetOnce<>();
runUnderPrimaryPermit(() -> {
final IndexShardRoutingTable routingTable = shard.getReplicationGroup().getRoutingTable();
ShardRouting targetShardRouting = routingTable.getByAllocationId(request.targetAllocationId());
if (targetShardRouting == null) {
logger.debug("delaying recovery of {} as it is not listed as assigned to target node {}", request.shardId(), request.targetNode());
throw new DelayRecoveryException("source node does not have the shard listed in its state as allocated on the node");
}
assert targetShardRouting.initializing() : "expected recovery target to be initializing but was " + targetShardRouting;
retentionLeaseRef.set(shard.getRetentionLeases().get(ReplicationTracker.getPeerRecoveryRetentionLeaseId(targetShardRouting)));
}, shardId + " validating recovery target [" + request.targetAllocationId() + "] registered ", shard, cancellableThreads, logger);
final Closeable retentionLock = shard.acquireHistoryRetentionLock();
resources.add(retentionLock);
final long startingSeqNo;
final boolean isSequenceNumberBasedRecovery = request.startingSeqNo() != SequenceNumbers.UNASSIGNED_SEQ_NO && isTargetSameHistory() && shard.hasCompleteHistoryOperations(PEER_RECOVERY_NAME, request.startingSeqNo()) && ((retentionLeaseRef.get() == null && shard.useRetentionLeasesInPeerRecovery() == false) || (retentionLeaseRef.get() != null && retentionLeaseRef.get().retainingSequenceNumber() <= request.startingSeqNo()));
if (isSequenceNumberBasedRecovery && retentionLeaseRef.get() != null) {
// all the history we need is retained by an existing retention lease, so we do not need a separate retention lock
retentionLock.close();
logger.trace("history is retained by {}", retentionLeaseRef.get());
} else {
// all the history we need is retained by the retention lock, obtained before calling shard.hasCompleteHistoryOperations()
// and before acquiring the safe commit we'll be using, so we can be certain that all operations after the safe commit's
// local checkpoint will be retained for the duration of this recovery.
logger.trace("history is retained by retention lock");
}
final StepListener<SendFileResult> sendFileStep = new StepListener<>();
final StepListener<TimeValue> prepareEngineStep = new StepListener<>();
final StepListener<SendSnapshotResult> sendSnapshotStep = new StepListener<>();
final StepListener<Void> finalizeStep = new StepListener<>();
if (isSequenceNumberBasedRecovery) {
logger.trace("performing sequence numbers based recovery. starting at [{}]", request.startingSeqNo());
startingSeqNo = request.startingSeqNo();
if (retentionLeaseRef.get() == null) {
createRetentionLease(startingSeqNo, ActionListener.map(sendFileStep, ignored -> SendFileResult.EMPTY));
} else {
sendFileStep.onResponse(SendFileResult.EMPTY);
}
} else {
final Engine.IndexCommitRef safeCommitRef;
try {
safeCommitRef = acquireSafeCommit(shard);
resources.add(safeCommitRef);
} catch (final Exception e) {
throw new RecoveryEngineException(shard.shardId(), 1, "snapshot failed", e);
}
// Try and copy enough operations to the recovering peer so that if it is promoted to primary then it has a chance of being
// able to recover other replicas using operations-based recoveries. If we are not using retention leases then we
// conservatively copy all available operations. If we are using retention leases then "enough operations" is just the
// operations from the local checkpoint of the safe commit onwards, because when using soft deletes the safe commit retains
// at least as much history as anything else. The safe commit will often contain all the history retained by the current set
// of retention leases, but this is not guaranteed: an earlier peer recovery from a different primary might have created a
// retention lease for some history that this primary already discarded, since we discard history when the global checkpoint
// advances and not when creating a new safe commit. In any case this is a best-effort thing since future recoveries can
// always fall back to file-based ones, and only really presents a problem if this primary fails before things have settled
// down.
startingSeqNo = Long.parseLong(safeCommitRef.getIndexCommit().getUserData().get(SequenceNumbers.LOCAL_CHECKPOINT_KEY)) + 1L;
logger.trace("performing file-based recovery followed by history replay starting at [{}]", startingSeqNo);
try {
final int estimateNumOps = estimateNumberOfHistoryOperations(startingSeqNo);
final Releasable releaseStore = acquireStore(shard.store());
resources.add(releaseStore);
sendFileStep.whenComplete(r -> IOUtils.close(safeCommitRef, releaseStore), e -> {
try {
IOUtils.close(safeCommitRef, releaseStore);
} catch (final IOException ex) {
logger.warn("releasing snapshot caused exception", ex);
}
});
final StepListener<ReplicationResponse> deleteRetentionLeaseStep = new StepListener<>();
runUnderPrimaryPermit(() -> {
try {
// If the target previously had a copy of this shard then a file-based recovery might move its global
// checkpoint backwards. We must therefore remove any existing retention lease so that we can create a
// new one later on in the recovery.
shard.removePeerRecoveryRetentionLease(request.targetNode().getId(), new ThreadedActionListener<>(logger, shard.getThreadPool(), ThreadPool.Names.GENERIC, deleteRetentionLeaseStep, false));
} catch (RetentionLeaseNotFoundException e) {
logger.debug("no peer-recovery retention lease for " + request.targetAllocationId());
deleteRetentionLeaseStep.onResponse(null);
}
}, shardId + " removing retention lease for [" + request.targetAllocationId() + "]", shard, cancellableThreads, logger);
deleteRetentionLeaseStep.whenComplete(ignored -> {
assert Transports.assertNotTransportThread(RecoverySourceHandler.this + "[phase1]");
phase1(safeCommitRef.getIndexCommit(), startingSeqNo, () -> estimateNumOps, sendFileStep);
}, onFailure);
} catch (final Exception e) {
throw new RecoveryEngineException(shard.shardId(), 1, "sendFileStep failed", e);
}
}
assert startingSeqNo >= 0 : "startingSeqNo must be non negative. got: " + startingSeqNo;
sendFileStep.whenComplete(r -> {
assert Transports.assertNotTransportThread(RecoverySourceHandler.this + "[prepareTargetForTranslog]");
// For a sequence based recovery, the target can keep its local translog
prepareTargetForTranslog(estimateNumberOfHistoryOperations(startingSeqNo), prepareEngineStep);
}, onFailure);
prepareEngineStep.whenComplete(prepareEngineTime -> {
assert Transports.assertNotTransportThread(RecoverySourceHandler.this + "[phase2]");
/*
* add shard to replication group (shard will receive replication requests from this point on) now that engine is open.
* This means that any document indexed into the primary after this will be replicated to this replica as well
* make sure to do this before sampling the max sequence number in the next step, to ensure that we send
* all documents up to maxSeqNo in phase2.
*/
runUnderPrimaryPermit(() -> shard.initiateTracking(request.targetAllocationId()), shardId + " initiating tracking of " + request.targetAllocationId(), shard, cancellableThreads, logger);
final long endingSeqNo = shard.seqNoStats().getMaxSeqNo();
if (logger.isTraceEnabled()) {
logger.trace("snapshot translog for recovery; current size is [{}]", estimateNumberOfHistoryOperations(startingSeqNo));
}
final Translog.Snapshot phase2Snapshot = shard.newChangesSnapshot(PEER_RECOVERY_NAME, startingSeqNo, Long.MAX_VALUE, false);
resources.add(phase2Snapshot);
retentionLock.close();
// we have to capture the max_seen_auto_id_timestamp and the max_seq_no_of_updates to make sure that these values
// are at least as high as the corresponding values on the primary when any of these operations were executed on it.
final long maxSeenAutoIdTimestamp = shard.getMaxSeenAutoIdTimestamp();
final long maxSeqNoOfUpdatesOrDeletes = shard.getMaxSeqNoOfUpdatesOrDeletes();
final RetentionLeases retentionLeases = shard.getRetentionLeases();
final long mappingVersionOnPrimary = shard.indexSettings().getIndexMetadata().getMappingVersion();
phase2(startingSeqNo, endingSeqNo, phase2Snapshot, maxSeenAutoIdTimestamp, maxSeqNoOfUpdatesOrDeletes, retentionLeases, mappingVersionOnPrimary, sendSnapshotStep);
}, onFailure);
// Recovery target can trim all operations >= startingSeqNo as we have sent all these operations in the phase 2
final long trimAboveSeqNo = startingSeqNo - 1;
sendSnapshotStep.whenComplete(r -> finalizeRecovery(r.targetLocalCheckpoint, trimAboveSeqNo, finalizeStep), onFailure);
finalizeStep.whenComplete(r -> {
// TODO: return the actual throttle time
final long phase1ThrottlingWaitTime = 0L;
final SendSnapshotResult sendSnapshotResult = sendSnapshotStep.result();
final SendFileResult sendFileResult = sendFileStep.result();
final RecoveryResponse response = new RecoveryResponse(sendFileResult.phase1FileNames, sendFileResult.phase1FileSizes, sendFileResult.phase1ExistingFileNames, sendFileResult.phase1ExistingFileSizes, sendFileResult.totalSize, sendFileResult.existingTotalSize, sendFileResult.took.millis(), phase1ThrottlingWaitTime, prepareEngineStep.result().millis(), sendSnapshotResult.sentOperations, sendSnapshotResult.tookTime.millis());
try {
future.onResponse(response);
} finally {
IOUtils.close(resources);
}
}, onFailure);
} catch (Exception e) {
IOUtils.closeWhileHandlingException(releaseResources, () -> future.onFailure(e));
}
}
use of org.opensearch.common.unit.TimeValue in project OpenSearch by opensearch-project.
the class JvmStats method toXContent.
@Override
public XContentBuilder toXContent(XContentBuilder builder, Params params) throws IOException {
builder.startObject(Fields.JVM);
builder.field(Fields.TIMESTAMP, timestamp);
builder.humanReadableField(Fields.UPTIME_IN_MILLIS, Fields.UPTIME, new TimeValue(uptime));
builder.startObject(Fields.MEM);
builder.humanReadableField(Fields.HEAP_USED_IN_BYTES, Fields.HEAP_USED, new ByteSizeValue(mem.heapUsed));
if (mem.getHeapUsedPercent() >= 0) {
builder.field(Fields.HEAP_USED_PERCENT, mem.getHeapUsedPercent());
}
builder.humanReadableField(Fields.HEAP_COMMITTED_IN_BYTES, Fields.HEAP_COMMITTED, new ByteSizeValue(mem.heapCommitted));
builder.humanReadableField(Fields.HEAP_MAX_IN_BYTES, Fields.HEAP_MAX, new ByteSizeValue(mem.heapMax));
builder.humanReadableField(Fields.NON_HEAP_USED_IN_BYTES, Fields.NON_HEAP_USED, new ByteSizeValue(mem.nonHeapUsed));
builder.humanReadableField(Fields.NON_HEAP_COMMITTED_IN_BYTES, Fields.NON_HEAP_COMMITTED, new ByteSizeValue(mem.nonHeapCommitted));
builder.startObject(Fields.POOLS);
for (MemoryPool pool : mem) {
builder.startObject(pool.getName());
builder.humanReadableField(Fields.USED_IN_BYTES, Fields.USED, new ByteSizeValue(pool.used));
builder.humanReadableField(Fields.MAX_IN_BYTES, Fields.MAX, new ByteSizeValue(pool.max));
builder.humanReadableField(Fields.PEAK_USED_IN_BYTES, Fields.PEAK_USED, new ByteSizeValue(pool.peakUsed));
builder.humanReadableField(Fields.PEAK_MAX_IN_BYTES, Fields.PEAK_MAX, new ByteSizeValue(pool.peakMax));
builder.startObject(Fields.LAST_GC_STATS);
builder.humanReadableField(Fields.USED_IN_BYTES, Fields.USED, new ByteSizeValue(pool.getLastGcStats().used));
builder.humanReadableField(Fields.MAX_IN_BYTES, Fields.MAX, new ByteSizeValue(pool.getLastGcStats().max));
builder.field(Fields.USAGE_PERCENT, pool.getLastGcStats().getUsagePercent());
builder.endObject();
builder.endObject();
}
builder.endObject();
builder.endObject();
builder.startObject(Fields.THREADS);
builder.field(Fields.COUNT, threads.getCount());
builder.field(Fields.PEAK_COUNT, threads.getPeakCount());
builder.endObject();
builder.startObject(Fields.GC);
builder.startObject(Fields.COLLECTORS);
for (GarbageCollector collector : gc) {
builder.startObject(collector.getName());
builder.field(Fields.COLLECTION_COUNT, collector.getCollectionCount());
builder.humanReadableField(Fields.COLLECTION_TIME_IN_MILLIS, Fields.COLLECTION_TIME, new TimeValue(collector.collectionTime));
builder.endObject();
}
builder.endObject();
builder.endObject();
if (bufferPools != null) {
builder.startObject(Fields.BUFFER_POOLS);
for (BufferPool bufferPool : bufferPools) {
builder.startObject(bufferPool.getName());
builder.field(Fields.COUNT, bufferPool.getCount());
builder.humanReadableField(Fields.USED_IN_BYTES, Fields.USED, new ByteSizeValue(bufferPool.used));
builder.humanReadableField(Fields.TOTAL_CAPACITY_IN_BYTES, Fields.TOTAL_CAPACITY, new ByteSizeValue(bufferPool.totalCapacity));
builder.endObject();
}
builder.endObject();
}
builder.startObject(Fields.CLASSES);
builder.field(Fields.CURRENT_LOADED_COUNT, classes.getLoadedClassCount());
builder.field(Fields.TOTAL_LOADED_COUNT, classes.getTotalLoadedClassCount());
builder.field(Fields.TOTAL_UNLOADED_COUNT, classes.getUnloadedClassCount());
builder.endObject();
builder.endObject();
return builder;
}
use of org.opensearch.common.unit.TimeValue in project OpenSearch by opensearch-project.
the class PersistentTasksService method waitForPersistentTaskCondition.
/**
* Waits for a given persistent task to comply with a given predicate, then call back the listener accordingly.
*
* @param taskId the persistent task id
* @param predicate the persistent task predicate to evaluate
* @param timeout a timeout for waiting
* @param listener the callback listener
*/
public void waitForPersistentTaskCondition(final String taskId, final Predicate<PersistentTask<?>> predicate, @Nullable final TimeValue timeout, final WaitForPersistentTaskListener<?> listener) {
final Predicate<ClusterState> clusterStatePredicate = clusterState -> predicate.test(PersistentTasksCustomMetadata.getTaskWithId(clusterState, taskId));
final ClusterStateObserver observer = new ClusterStateObserver(clusterService, timeout, logger, threadPool.getThreadContext());
final ClusterState clusterState = observer.setAndGetObservedState();
if (clusterStatePredicate.test(clusterState)) {
listener.onResponse(PersistentTasksCustomMetadata.getTaskWithId(clusterState, taskId));
} else {
observer.waitForNextChange(new ClusterStateObserver.Listener() {
@Override
public void onNewClusterState(ClusterState state) {
listener.onResponse(PersistentTasksCustomMetadata.getTaskWithId(state, taskId));
}
@Override
public void onClusterServiceClose() {
listener.onFailure(new NodeClosedException(clusterService.localNode()));
}
@Override
public void onTimeout(TimeValue timeout) {
listener.onTimeout(timeout);
}
}, clusterStatePredicate);
}
}
use of org.opensearch.common.unit.TimeValue in project OpenSearch by opensearch-project.
the class Node method start.
/**
* Start the node. If the node is already started, this method is no-op.
*/
public Node start() throws NodeValidationException {
if (!lifecycle.moveToStarted()) {
return this;
}
logger.info("starting ...");
pluginLifecycleComponents.forEach(LifecycleComponent::start);
injector.getInstance(MappingUpdatedAction.class).setClient(client);
injector.getInstance(IndicesService.class).start();
injector.getInstance(IndicesClusterStateService.class).start();
injector.getInstance(SnapshotsService.class).start();
injector.getInstance(SnapshotShardsService.class).start();
injector.getInstance(RepositoriesService.class).start();
injector.getInstance(SearchService.class).start();
injector.getInstance(FsHealthService.class).start();
nodeService.getMonitorService().start();
final ClusterService clusterService = injector.getInstance(ClusterService.class);
final NodeConnectionsService nodeConnectionsService = injector.getInstance(NodeConnectionsService.class);
nodeConnectionsService.start();
clusterService.setNodeConnectionsService(nodeConnectionsService);
injector.getInstance(GatewayService.class).start();
Discovery discovery = injector.getInstance(Discovery.class);
clusterService.getMasterService().setClusterStatePublisher(discovery::publish);
// Start the transport service now so the publish address will be added to the local disco node in ClusterService
TransportService transportService = injector.getInstance(TransportService.class);
transportService.getTaskManager().setTaskResultsService(injector.getInstance(TaskResultsService.class));
transportService.getTaskManager().setTaskCancellationService(new TaskCancellationService(transportService));
transportService.start();
assert localNodeFactory.getNode() != null;
assert transportService.getLocalNode().equals(localNodeFactory.getNode()) : "transportService has a different local node than the factory provided";
injector.getInstance(PeerRecoverySourceService.class).start();
// Load (and maybe upgrade) the metadata stored on disk
final GatewayMetaState gatewayMetaState = injector.getInstance(GatewayMetaState.class);
gatewayMetaState.start(settings(), transportService, clusterService, injector.getInstance(MetaStateService.class), injector.getInstance(MetadataIndexUpgradeService.class), injector.getInstance(MetadataUpgrader.class), injector.getInstance(PersistedClusterStateService.class));
if (Assertions.ENABLED) {
try {
assert injector.getInstance(MetaStateService.class).loadFullState().v1().isEmpty();
final NodeMetadata nodeMetadata = NodeMetadata.FORMAT.loadLatestState(logger, NamedXContentRegistry.EMPTY, nodeEnvironment.nodeDataPaths());
assert nodeMetadata != null;
assert nodeMetadata.nodeVersion().equals(Version.CURRENT);
assert nodeMetadata.nodeId().equals(localNodeFactory.getNode().getId());
} catch (IOException e) {
assert false : e;
}
}
// we load the global state here (the persistent part of the cluster state stored on disk) to
// pass it to the bootstrap checks to allow plugins to enforce certain preconditions based on the recovered state.
final Metadata onDiskMetadata = gatewayMetaState.getPersistedState().getLastAcceptedState().metadata();
// this is never null
assert onDiskMetadata != null : "metadata is null but shouldn't";
validateNodeBeforeAcceptingRequests(new BootstrapContext(environment, onDiskMetadata), transportService.boundAddress(), pluginsService.filterPlugins(Plugin.class).stream().flatMap(p -> p.getBootstrapChecks().stream()).collect(Collectors.toList()));
clusterService.addStateApplier(transportService.getTaskManager());
// start after transport service so the local disco is known
// start before cluster service so that it can set initial state on ClusterApplierService
discovery.start();
clusterService.start();
assert clusterService.localNode().equals(localNodeFactory.getNode()) : "clusterService has a different local node than the factory provided";
transportService.acceptIncomingRequests();
discovery.startInitialJoin();
final TimeValue initialStateTimeout = DiscoverySettings.INITIAL_STATE_TIMEOUT_SETTING.get(settings());
configureNodeAndClusterIdStateListener(clusterService);
if (initialStateTimeout.millis() > 0) {
final ThreadPool thread = injector.getInstance(ThreadPool.class);
ClusterState clusterState = clusterService.state();
ClusterStateObserver observer = new ClusterStateObserver(clusterState, clusterService, null, logger, thread.getThreadContext());
if (clusterState.nodes().getMasterNodeId() == null) {
logger.debug("waiting to join the cluster. timeout [{}]", initialStateTimeout);
final CountDownLatch latch = new CountDownLatch(1);
observer.waitForNextChange(new ClusterStateObserver.Listener() {
@Override
public void onNewClusterState(ClusterState state) {
latch.countDown();
}
@Override
public void onClusterServiceClose() {
latch.countDown();
}
@Override
public void onTimeout(TimeValue timeout) {
logger.warn("timed out while waiting for initial discovery state - timeout: {}", initialStateTimeout);
latch.countDown();
}
}, state -> state.nodes().getMasterNodeId() != null, initialStateTimeout);
try {
latch.await();
} catch (InterruptedException e) {
throw new OpenSearchTimeoutException("Interrupted while waiting for initial discovery state");
}
}
}
injector.getInstance(HttpServerTransport.class).start();
if (WRITE_PORTS_FILE_SETTING.get(settings())) {
TransportService transport = injector.getInstance(TransportService.class);
writePortsFile("transport", transport.boundAddress());
HttpServerTransport http = injector.getInstance(HttpServerTransport.class);
writePortsFile("http", http.boundAddress());
}
logger.info("started");
pluginsService.filterPlugins(ClusterPlugin.class).forEach(ClusterPlugin::onNodeStarted);
return this;
}
use of org.opensearch.common.unit.TimeValue in project OpenSearch by opensearch-project.
the class OsInfo method toXContent.
@Override
public XContentBuilder toXContent(XContentBuilder builder, Params params) throws IOException {
builder.startObject(Fields.OS);
builder.humanReadableField(Fields.REFRESH_INTERVAL_IN_MILLIS, Fields.REFRESH_INTERVAL, new TimeValue(refreshInterval));
if (name != null) {
builder.field(Fields.NAME, name);
}
if (prettyName != null) {
builder.field(Fields.PRETTY_NAME, prettyName);
}
if (arch != null) {
builder.field(Fields.ARCH, arch);
}
if (version != null) {
builder.field(Fields.VERSION, version);
}
builder.field(Fields.AVAILABLE_PROCESSORS, availableProcessors);
builder.field(Fields.ALLOCATED_PROCESSORS, allocatedProcessors);
builder.endObject();
return builder;
}
Aggregations