2 * Copyright (c) 2015 Cisco Systems, Inc. and others. All rights reserved.
4 * This program and the accompanying materials are made available under the
5 * terms of the Eclipse Public License v1.0 which accompanies this distribution,
6 * and is available at http://www.eclipse.org/legal/epl-v10.html
8 package org.opendaylight.controller.cluster.datastore;
10 import akka.actor.ActorRef;
11 import akka.util.Timeout;
12 import com.google.common.annotations.VisibleForTesting;
13 import com.google.common.base.Optional;
14 import com.google.common.base.Preconditions;
15 import com.google.common.base.Stopwatch;
16 import com.google.common.base.Verify;
17 import com.google.common.collect.ImmutableList;
18 import com.google.common.collect.ImmutableMap;
19 import com.google.common.collect.ImmutableMap.Builder;
20 import com.google.common.primitives.UnsignedLong;
22 import java.io.IOException;
23 import java.util.AbstractMap.SimpleEntry;
24 import java.util.ArrayDeque;
25 import java.util.ArrayList;
26 import java.util.Collection;
27 import java.util.HashMap;
28 import java.util.Iterator;
30 import java.util.Map.Entry;
31 import java.util.Queue;
32 import java.util.concurrent.ExecutionException;
33 import java.util.concurrent.TimeUnit;
34 import java.util.concurrent.TimeoutException;
35 import java.util.function.UnaryOperator;
36 import javax.annotation.Nonnull;
37 import javax.annotation.concurrent.NotThreadSafe;
38 import org.opendaylight.controller.cluster.access.concepts.LocalHistoryIdentifier;
39 import org.opendaylight.controller.cluster.access.concepts.TransactionIdentifier;
40 import org.opendaylight.controller.cluster.datastore.DataTreeCohortActorRegistry.CohortRegistryCommand;
41 import org.opendaylight.controller.cluster.datastore.ShardDataTreeCohort.State;
42 import org.opendaylight.controller.cluster.datastore.persisted.CommitTransactionPayload;
43 import org.opendaylight.controller.cluster.datastore.persisted.MetadataShardDataTreeSnapshot;
44 import org.opendaylight.controller.cluster.datastore.persisted.ShardDataTreeSnapshot;
45 import org.opendaylight.controller.cluster.datastore.persisted.ShardDataTreeSnapshotMetadata;
46 import org.opendaylight.controller.cluster.datastore.utils.DataTreeModificationOutput;
47 import org.opendaylight.controller.cluster.datastore.utils.PruningDataTreeModification;
48 import org.opendaylight.controller.cluster.raft.protobuff.client.messages.Payload;
49 import org.opendaylight.controller.md.sal.common.api.data.AsyncDataBroker.DataChangeScope;
50 import org.opendaylight.controller.md.sal.common.api.data.AsyncDataChangeListener;
51 import org.opendaylight.controller.md.sal.common.api.data.OptimisticLockFailedException;
52 import org.opendaylight.controller.md.sal.common.api.data.TransactionCommitFailedException;
53 import org.opendaylight.controller.md.sal.dom.api.DOMDataTreeChangeListener;
54 import org.opendaylight.controller.md.sal.dom.store.impl.DataChangeListenerRegistration;
55 import org.opendaylight.yangtools.concepts.Identifier;
56 import org.opendaylight.yangtools.concepts.ListenerRegistration;
57 import org.opendaylight.yangtools.yang.data.api.YangInstanceIdentifier;
58 import org.opendaylight.yangtools.yang.data.api.schema.NormalizedNode;
59 import org.opendaylight.yangtools.yang.data.api.schema.tree.ConflictingModificationAppliedException;
60 import org.opendaylight.yangtools.yang.data.api.schema.tree.DataTreeCandidate;
61 import org.opendaylight.yangtools.yang.data.api.schema.tree.DataTreeCandidateTip;
62 import org.opendaylight.yangtools.yang.data.api.schema.tree.DataTreeCandidates;
63 import org.opendaylight.yangtools.yang.data.api.schema.tree.DataTreeModification;
64 import org.opendaylight.yangtools.yang.data.api.schema.tree.DataTreeSnapshot;
65 import org.opendaylight.yangtools.yang.data.api.schema.tree.DataValidationFailedException;
66 import org.opendaylight.yangtools.yang.data.api.schema.tree.ModificationType;
67 import org.opendaylight.yangtools.yang.data.api.schema.tree.TipProducingDataTree;
68 import org.opendaylight.yangtools.yang.data.api.schema.tree.TreeType;
69 import org.opendaylight.yangtools.yang.data.impl.schema.tree.InMemoryDataTreeFactory;
70 import org.opendaylight.yangtools.yang.model.api.SchemaContext;
71 import org.slf4j.Logger;
72 import org.slf4j.LoggerFactory;
73 import scala.concurrent.duration.Duration;
76 * Internal shard state, similar to a DOMStore, but optimized for use in the actor system,
77 * e.g. it does not expose public interfaces and assumes it is only ever called from a
80 * This class is not part of the API contract and is subject to change at any time.
83 public class ShardDataTree extends ShardDataTreeTransactionParent {
84 private static final class CommitEntry {
85 final SimpleShardDataTreeCohort cohort;
88 CommitEntry(final SimpleShardDataTreeCohort cohort, final long now) {
89 this.cohort = Preconditions.checkNotNull(cohort);
94 private static final Timeout COMMIT_STEP_TIMEOUT = new Timeout(Duration.create(5, TimeUnit.SECONDS));
95 private static final Logger LOG = LoggerFactory.getLogger(ShardDataTree.class);
97 private final Map<LocalHistoryIdentifier, ShardDataTreeTransactionChain> transactionChains = new HashMap<>();
98 private final DataTreeCohortActorRegistry cohortRegistry = new DataTreeCohortActorRegistry();
99 private final Queue<CommitEntry> pendingTransactions = new ArrayDeque<>();
100 private final ShardDataTreeChangeListenerPublisher treeChangeListenerPublisher;
101 private final ShardDataChangeListenerPublisher dataChangeListenerPublisher;
102 private final Collection<ShardDataTreeMetadata<?>> metadata;
103 private final TipProducingDataTree dataTree;
104 private final String logContext;
105 private final Shard shard;
106 private Runnable runOnPendingTransactionsComplete;
108 private SchemaContext schemaContext;
110 public ShardDataTree(final Shard shard, final SchemaContext schemaContext, final TipProducingDataTree dataTree,
111 final ShardDataTreeChangeListenerPublisher treeChangeListenerPublisher,
112 final ShardDataChangeListenerPublisher dataChangeListenerPublisher, final String logContext,
113 final ShardDataTreeMetadata<?>... metadata) {
114 this.dataTree = Preconditions.checkNotNull(dataTree);
115 updateSchemaContext(schemaContext);
117 this.shard = Preconditions.checkNotNull(shard);
118 this.treeChangeListenerPublisher = Preconditions.checkNotNull(treeChangeListenerPublisher);
119 this.dataChangeListenerPublisher = Preconditions.checkNotNull(dataChangeListenerPublisher);
120 this.logContext = Preconditions.checkNotNull(logContext);
121 this.metadata = ImmutableList.copyOf(metadata);
124 public ShardDataTree(final Shard shard, final SchemaContext schemaContext, final TreeType treeType,
125 final ShardDataTreeChangeListenerPublisher treeChangeListenerPublisher,
126 final ShardDataChangeListenerPublisher dataChangeListenerPublisher, final String logContext) {
127 this(shard, schemaContext, InMemoryDataTreeFactory.getInstance().create(treeType),
128 treeChangeListenerPublisher, dataChangeListenerPublisher, logContext);
132 public ShardDataTree(final Shard shard, final SchemaContext schemaContext, final TreeType treeType) {
133 this(shard, schemaContext, treeType, new DefaultShardDataTreeChangeListenerPublisher(),
134 new DefaultShardDataChangeListenerPublisher(), "");
137 String logContext() {
141 public TipProducingDataTree getDataTree() {
145 SchemaContext getSchemaContext() {
146 return schemaContext;
149 void updateSchemaContext(final SchemaContext schemaContext) {
150 dataTree.setSchemaContext(schemaContext);
151 this.schemaContext = Preconditions.checkNotNull(schemaContext);
155 * Take a snapshot of current state for later recovery.
157 * @return A state snapshot
159 @Nonnull ShardDataTreeSnapshot takeStateSnapshot() {
160 final NormalizedNode<?, ?> rootNode = dataTree.takeSnapshot().readNode(YangInstanceIdentifier.EMPTY).get();
161 final Builder<Class<? extends ShardDataTreeSnapshotMetadata<?>>, ShardDataTreeSnapshotMetadata<?>> metaBuilder =
162 ImmutableMap.builder();
164 for (ShardDataTreeMetadata<?> m : metadata) {
165 final ShardDataTreeSnapshotMetadata<?> meta = m.toSnapshot();
167 metaBuilder.put(meta.getType(), meta);
171 return new MetadataShardDataTreeSnapshot(rootNode, metaBuilder.build());
174 private void applySnapshot(final @Nonnull ShardDataTreeSnapshot snapshot,
175 final UnaryOperator<DataTreeModification> wrapper) throws DataValidationFailedException {
176 final Stopwatch elapsed = Stopwatch.createStarted();
178 if (!pendingTransactions.isEmpty()) {
179 LOG.warn("{}: applying state snapshot with pending transactions", logContext);
182 final Map<Class<? extends ShardDataTreeSnapshotMetadata<?>>, ShardDataTreeSnapshotMetadata<?>> snapshotMeta;
183 if (snapshot instanceof MetadataShardDataTreeSnapshot) {
184 snapshotMeta = ((MetadataShardDataTreeSnapshot) snapshot).getMetadata();
186 snapshotMeta = ImmutableMap.of();
189 for (ShardDataTreeMetadata<?> m : metadata) {
190 final ShardDataTreeSnapshotMetadata<?> s = snapshotMeta.get(m.getSupportedType());
198 final DataTreeModification mod = wrapper.apply(dataTree.takeSnapshot().newModification());
199 // delete everything first
200 mod.delete(YangInstanceIdentifier.EMPTY);
202 final java.util.Optional<NormalizedNode<?, ?>> maybeNode = snapshot.getRootNode();
203 if (maybeNode.isPresent()) {
204 // Add everything from the remote node back
205 mod.write(YangInstanceIdentifier.EMPTY, maybeNode.get());
209 final DataTreeModification unwrapped = unwrap(mod);
210 dataTree.validate(unwrapped);
211 DataTreeCandidateTip candidate = dataTree.prepare(unwrapped);
212 dataTree.commit(candidate);
213 notifyListeners(candidate);
215 LOG.debug("{}: state snapshot applied in %s", logContext, elapsed);
218 private PruningDataTreeModification wrapWithPruning(final DataTreeModification delegate) {
219 return new PruningDataTreeModification(delegate, dataTree, schemaContext);
222 private static DataTreeModification unwrap(final DataTreeModification modification) {
223 if (modification instanceof PruningDataTreeModification) {
224 return ((PruningDataTreeModification)modification).delegate();
230 * Apply a snapshot coming from recovery. This method does not assume the SchemaContexts match and performs data
231 * pruning in an attempt to adjust the state to our current SchemaContext.
233 * @param snapshot Snapshot that needs to be applied
234 * @throws DataValidationFailedException when the snapshot fails to apply
236 void applyRecoverySnapshot(final @Nonnull ShardDataTreeSnapshot snapshot) throws DataValidationFailedException {
237 applySnapshot(snapshot, this::wrapWithPruning);
242 * Apply a snapshot coming from the leader. This method assumes the leader and follower SchemaContexts match and
243 * does not perform any pruning.
245 * @param snapshot Snapshot that needs to be applied
246 * @throws DataValidationFailedException when the snapshot fails to apply
248 void applySnapshot(final @Nonnull ShardDataTreeSnapshot snapshot) throws DataValidationFailedException {
249 applySnapshot(snapshot, UnaryOperator.identity());
252 private void applyRecoveryCandidate(final DataTreeCandidate candidate) throws DataValidationFailedException {
253 final PruningDataTreeModification mod = wrapWithPruning(dataTree.takeSnapshot().newModification());
254 DataTreeCandidates.applyToModification(mod, candidate);
257 final DataTreeModification unwrapped = mod.delegate();
258 LOG.trace("{}: Applying recovery modification {}", logContext, unwrapped);
261 dataTree.validate(unwrapped);
262 dataTree.commit(dataTree.prepare(unwrapped));
263 } catch (Exception e) {
264 File file = new File(System.getProperty("karaf.data", "."),
265 "failed-recovery-payload-" + logContext + ".out");
266 DataTreeModificationOutput.toFile(file, unwrapped);
267 throw new IllegalStateException(String.format(
268 "%s: Failed to apply recovery payload. Modification data was written to file %s",
269 logContext, file), e);
274 * Apply a payload coming from recovery. This method does not assume the SchemaContexts match and performs data
275 * pruning in an attempt to adjust the state to our current SchemaContext.
277 * @param payload Payload
278 * @throws IOException when the snapshot fails to deserialize
279 * @throws DataValidationFailedException when the snapshot fails to apply
281 void applyRecoveryPayload(final @Nonnull Payload payload) throws IOException, DataValidationFailedException {
282 if (payload instanceof CommitTransactionPayload) {
283 final Entry<TransactionIdentifier, DataTreeCandidate> e = ((CommitTransactionPayload) payload).getCandidate();
284 applyRecoveryCandidate(e.getValue());
285 allMetadataCommittedTransaction(e.getKey());
286 } else if (payload instanceof DataTreeCandidatePayload) {
287 applyRecoveryCandidate(((DataTreeCandidatePayload) payload).getCandidate());
289 LOG.debug("{}: ignoring unhandled payload {}", logContext, payload);
293 private void applyReplicatedCandidate(final Identifier identifier, final DataTreeCandidate foreign)
294 throws DataValidationFailedException {
295 LOG.debug("{}: Applying foreign transaction {}", logContext, identifier);
297 final DataTreeModification mod = dataTree.takeSnapshot().newModification();
298 DataTreeCandidates.applyToModification(mod, foreign);
301 LOG.trace("{}: Applying foreign modification {}", logContext, mod);
302 dataTree.validate(mod);
303 final DataTreeCandidate candidate = dataTree.prepare(mod);
304 dataTree.commit(candidate);
306 notifyListeners(candidate);
310 * Apply a payload coming from the leader, which could actually be us. This method assumes the leader and follower
311 * SchemaContexts match and does not perform any pruning.
313 * @param identifier Payload identifier as returned from RaftActor
314 * @param payload Payload
315 * @throws IOException when the snapshot fails to deserialize
316 * @throws DataValidationFailedException when the snapshot fails to apply
318 void applyReplicatedPayload(final Identifier identifier, final Payload payload) throws IOException,
319 DataValidationFailedException {
321 * This is a bit more involved than it needs to be due to to the fact we do not want to be touching the payload
322 * if we are the leader and it has originated with us.
324 * The identifier will only ever be non-null when we were the leader which achieved consensus. Unfortunately,
325 * though, this may not be the case anymore, as we are being called some time afterwards and we may not be
326 * acting in that capacity anymore.
328 * In any case, we know that this is an entry coming from replication, hence we can be sure we will not observe
329 * pre-Boron state -- which limits the number of options here.
331 if (payload instanceof CommitTransactionPayload) {
332 if (identifier == null) {
333 final Entry<TransactionIdentifier, DataTreeCandidate> e = ((CommitTransactionPayload) payload).getCandidate();
334 applyReplicatedCandidate(e.getKey(), e.getValue());
335 allMetadataCommittedTransaction(e.getKey());
337 Verify.verify(identifier instanceof TransactionIdentifier);
338 payloadReplicationComplete((TransactionIdentifier) identifier);
341 LOG.warn("{}: ignoring unhandled identifier {} payload {}", logContext, identifier, payload);
345 private void payloadReplicationComplete(final TransactionIdentifier txId) {
346 final CommitEntry current = pendingTransactions.peek();
347 if (current == null) {
348 LOG.warn("{}: No outstanding transactions, ignoring consensus on transaction {}", logContext, txId);
352 if (!current.cohort.getIdentifier().equals(txId)) {
353 LOG.warn("{}: Head of queue is {}, ignoring consensus on transaction {}", logContext,
354 current.cohort.getIdentifier(), txId);
358 finishCommit(current.cohort);
361 private void allMetadataCommittedTransaction(final TransactionIdentifier txId) {
362 for (ShardDataTreeMetadata<?> m : metadata) {
363 m.onTransactionCommitted(txId);
367 private ShardDataTreeTransactionChain ensureTransactionChain(final LocalHistoryIdentifier localHistoryIdentifier) {
368 ShardDataTreeTransactionChain chain = transactionChains.get(localHistoryIdentifier);
370 chain = new ShardDataTreeTransactionChain(localHistoryIdentifier, this);
371 transactionChains.put(localHistoryIdentifier, chain);
377 ReadOnlyShardDataTreeTransaction newReadOnlyTransaction(final TransactionIdentifier txId) {
378 if (txId.getHistoryId().getHistoryId() == 0) {
379 return new ReadOnlyShardDataTreeTransaction(txId, dataTree.takeSnapshot());
382 return ensureTransactionChain(txId.getHistoryId()).newReadOnlyTransaction(txId);
385 ReadWriteShardDataTreeTransaction newReadWriteTransaction(final TransactionIdentifier txId) {
386 if (txId.getHistoryId().getHistoryId() == 0) {
387 return new ReadWriteShardDataTreeTransaction(ShardDataTree.this, txId, dataTree.takeSnapshot()
391 return ensureTransactionChain(txId.getHistoryId()).newReadWriteTransaction(txId);
395 public void notifyListeners(final DataTreeCandidate candidate) {
396 treeChangeListenerPublisher.publishChanges(candidate, logContext);
397 dataChangeListenerPublisher.publishChanges(candidate, logContext);
400 void notifyOfInitialData(final DataChangeListenerRegistration<AsyncDataChangeListener<YangInstanceIdentifier,
401 NormalizedNode<?, ?>>> listenerReg, final Optional<DataTreeCandidate> currentState) {
402 if (currentState.isPresent()) {
403 ShardDataChangeListenerPublisher localPublisher = dataChangeListenerPublisher.newInstance();
404 localPublisher.registerDataChangeListener(listenerReg.getPath(), listenerReg.getInstance(),
405 listenerReg.getScope());
406 localPublisher.publishChanges(currentState.get(), logContext);
410 void notifyOfInitialData(final YangInstanceIdentifier path, final DOMDataTreeChangeListener listener,
411 final Optional<DataTreeCandidate> currentState) {
412 if (currentState.isPresent()) {
413 ShardDataTreeChangeListenerPublisher localPublisher = treeChangeListenerPublisher.newInstance();
414 localPublisher.registerTreeChangeListener(path, listener);
415 localPublisher.publishChanges(currentState.get(), logContext);
419 void closeAllTransactionChains() {
420 for (ShardDataTreeTransactionChain chain : transactionChains.values()) {
424 transactionChains.clear();
427 void closeTransactionChain(final LocalHistoryIdentifier transactionChainId) {
428 final ShardDataTreeTransactionChain chain = transactionChains.remove(transactionChainId);
432 LOG.debug("{}: Closing non-existent transaction chain {}", logContext, transactionChainId);
436 Entry<DataChangeListenerRegistration<AsyncDataChangeListener<YangInstanceIdentifier, NormalizedNode<?, ?>>>,
437 Optional<DataTreeCandidate>> registerChangeListener(final YangInstanceIdentifier path,
438 final AsyncDataChangeListener<YangInstanceIdentifier, NormalizedNode<?, ?>> listener,
439 final DataChangeScope scope) {
440 final DataChangeListenerRegistration<AsyncDataChangeListener<YangInstanceIdentifier, NormalizedNode<?, ?>>> reg =
441 dataChangeListenerPublisher.registerDataChangeListener(path, listener, scope);
443 return new SimpleEntry<>(reg, readCurrentData());
446 private Optional<DataTreeCandidate> readCurrentData() {
447 final Optional<NormalizedNode<?, ?>> currentState = dataTree.takeSnapshot().readNode(YangInstanceIdentifier.EMPTY);
448 return currentState.isPresent() ? Optional.of(DataTreeCandidates.fromNormalizedNode(
449 YangInstanceIdentifier.EMPTY, currentState.get())) : Optional.<DataTreeCandidate>absent();
452 public Entry<ListenerRegistration<DOMDataTreeChangeListener>, Optional<DataTreeCandidate>> registerTreeChangeListener(
453 final YangInstanceIdentifier path, final DOMDataTreeChangeListener listener) {
454 final ListenerRegistration<DOMDataTreeChangeListener> reg = treeChangeListenerPublisher.registerTreeChangeListener(
457 return new SimpleEntry<>(reg, readCurrentData());
461 return pendingTransactions.size();
465 void abortTransaction(final AbstractShardDataTreeTransaction<?> transaction) {
470 ShardDataTreeCohort finishTransaction(final ReadWriteShardDataTreeTransaction transaction) {
471 final DataTreeModification snapshot = transaction.getSnapshot();
474 return createReadyCohort(transaction.getId(), snapshot);
477 public Optional<NormalizedNode<?, ?>> readNode(final YangInstanceIdentifier path) {
478 return dataTree.takeSnapshot().readNode(path);
481 DataTreeSnapshot takeSnapshot() {
482 return dataTree.takeSnapshot();
486 public DataTreeModification newModification() {
487 return dataTree.takeSnapshot().newModification();
491 * @deprecated This method violates DataTree containment and will be removed.
495 public DataTreeCandidate commit(final DataTreeModification modification) throws DataValidationFailedException {
496 modification.ready();
497 dataTree.validate(modification);
498 DataTreeCandidate candidate = dataTree.prepare(modification);
499 dataTree.commit(candidate);
503 public Collection<ShardDataTreeCohort> getAndClearPendingTransactions() {
504 Collection<ShardDataTreeCohort> ret = new ArrayList<>(pendingTransactions.size());
505 for(CommitEntry entry: pendingTransactions) {
506 ret.add(entry.cohort);
509 pendingTransactions.clear();
513 private void processNextTransaction() {
514 while (!pendingTransactions.isEmpty()) {
515 final CommitEntry entry = pendingTransactions.peek();
516 final SimpleShardDataTreeCohort cohort = entry.cohort;
517 final DataTreeModification modification = cohort.getDataTreeModification();
519 if(cohort.getState() != State.CAN_COMMIT_PENDING) {
523 LOG.debug("{}: Validating transaction {}", logContext, cohort.getIdentifier());
526 dataTree.validate(modification);
527 LOG.debug("{}: Transaction {} validated", logContext, cohort.getIdentifier());
528 cohort.successfulCanCommit();
529 entry.lastAccess = shard.ticker().read();
531 } catch (ConflictingModificationAppliedException e) {
532 LOG.warn("{}: Store Tx {}: Conflicting modification for path {}.", logContext, cohort.getIdentifier(),
534 cause = new OptimisticLockFailedException("Optimistic lock failed.", e);
535 } catch (DataValidationFailedException e) {
536 LOG.warn("{}: Store Tx {}: Data validation failed for path {}.", logContext, cohort.getIdentifier(),
539 // For debugging purposes, allow dumping of the modification. Coupled with the above
540 // precondition log, it should allow us to understand what went on.
541 LOG.debug("{}: Store Tx {}: modifications: {} tree: {}", cohort.getIdentifier(), modification, dataTree);
542 cause = new TransactionCommitFailedException("Data did not pass validation.", e);
543 } catch (Exception e) {
544 LOG.warn("{}: Unexpected failure in validation phase", logContext, e);
548 // Failure path: propagate the failure, remove the transaction from the queue and loop to the next one
549 pendingTransactions.poll().cohort.failedCanCommit(cause);
552 maybeRunOperationOnPendingTransactionsComplete();
555 void startCanCommit(final SimpleShardDataTreeCohort cohort) {
556 final SimpleShardDataTreeCohort current = pendingTransactions.peek().cohort;
557 if (!cohort.equals(current)) {
558 LOG.debug("{}: Transaction {} scheduled for canCommit step", logContext, cohort.getIdentifier());
562 processNextTransaction();
565 private void failPreCommit(final Exception cause) {
566 shard.getShardMBean().incrementFailedTransactionsCount();
567 pendingTransactions.poll().cohort.failedPreCommit(cause);
568 processNextTransaction();
571 void startPreCommit(final SimpleShardDataTreeCohort cohort) {
572 final CommitEntry entry = pendingTransactions.peek();
573 Preconditions.checkState(entry != null, "Attempted to pre-commit of %s when no transactions pending", cohort);
575 final SimpleShardDataTreeCohort current = entry.cohort;
576 Verify.verify(cohort.equals(current), "Attempted to pre-commit %s while %s is pending", cohort, current);
577 final DataTreeCandidateTip candidate;
579 candidate = dataTree.prepare(cohort.getDataTreeModification());
580 } catch (Exception e) {
586 cohort.userPreCommit(candidate);
587 } catch (ExecutionException | TimeoutException e) {
592 entry.lastAccess = shard.ticker().read();
593 cohort.successfulPreCommit(candidate);
596 private void failCommit(final Exception cause) {
597 shard.getShardMBean().incrementFailedTransactionsCount();
598 pendingTransactions.poll().cohort.failedCommit(cause);
599 processNextTransaction();
602 private void finishCommit(final SimpleShardDataTreeCohort cohort) {
603 final TransactionIdentifier txId = cohort.getIdentifier();
604 final DataTreeCandidate candidate = cohort.getCandidate();
606 LOG.debug("{}: Resuming commit of transaction {}", logContext, txId);
609 dataTree.commit(candidate);
610 } catch (Exception e) {
611 LOG.error("{}: Failed to commit transaction {}", logContext, txId, e);
616 shard.getShardMBean().incrementCommittedTransactionCount();
617 shard.getShardMBean().setLastCommittedTransactionTime(System.currentTimeMillis());
619 // FIXME: propagate journal index
620 pendingTransactions.poll().cohort.successfulCommit(UnsignedLong.ZERO);
622 LOG.trace("{}: Transaction {} committed, proceeding to notify", logContext, txId);
623 notifyListeners(candidate);
625 processNextTransaction();
628 void startCommit(final SimpleShardDataTreeCohort cohort, final DataTreeCandidate candidate) {
629 final CommitEntry entry = pendingTransactions.peek();
630 Preconditions.checkState(entry != null, "Attempted to start commit of %s when no transactions pending", cohort);
632 final SimpleShardDataTreeCohort current = entry.cohort;
633 Verify.verify(cohort.equals(current), "Attempted to commit %s while %s is pending", cohort, current);
635 if (shard.canSkipPayload() || candidate.getRootNode().getModificationType() == ModificationType.UNMODIFIED) {
636 LOG.debug("{}: No replication required, proceeding to finish commit", logContext);
637 finishCommit(cohort);
641 final TransactionIdentifier txId = cohort.getIdentifier();
642 final Payload payload;
644 payload = CommitTransactionPayload.create(txId, candidate);
645 } catch (IOException e) {
646 LOG.error("{}: Failed to encode transaction {} candidate {}", logContext, txId, candidate, e);
647 pendingTransactions.poll().cohort.failedCommit(e);
651 // Once completed, we will continue via payloadReplicationComplete
652 entry.lastAccess = shard.ticker().read();
653 shard.persistPayload(txId, payload);
654 LOG.debug("{}: Transaction {} submitted to persistence", logContext, txId);
657 void processCohortRegistryCommand(final ActorRef sender, final CohortRegistryCommand message) {
658 cohortRegistry.process(sender, message);
661 ShardDataTreeCohort createReadyCohort(final TransactionIdentifier txId,
662 final DataTreeModification modification) {
663 SimpleShardDataTreeCohort cohort = new SimpleShardDataTreeCohort(this, modification, txId,
664 cohortRegistry.createCohort(schemaContext, txId, COMMIT_STEP_TIMEOUT));
665 pendingTransactions.add(new CommitEntry(cohort, shard.ticker().read()));
669 void checkForExpiredTransactions(final long transactionCommitTimeoutMillis) {
670 final long timeout = TimeUnit.MILLISECONDS.toNanos(transactionCommitTimeoutMillis);
671 final long now = shard.ticker().read();
672 final CommitEntry currentTx = pendingTransactions.peek();
673 if (currentTx != null && currentTx.lastAccess + timeout < now) {
674 LOG.warn("{}: Current transaction {} has timed out after {} ms in state {}", logContext,
675 currentTx.cohort.getIdentifier(), transactionCommitTimeoutMillis, currentTx.cohort.getState());
676 boolean processNext = true;
677 switch (currentTx.cohort.getState()) {
678 case CAN_COMMIT_PENDING:
679 pendingTransactions.poll().cohort.failedCanCommit(new TimeoutException());
681 case CAN_COMMIT_COMPLETE:
682 pendingTransactions.poll().cohort.reportFailure(new TimeoutException());
684 case PRE_COMMIT_PENDING:
685 pendingTransactions.poll().cohort.failedPreCommit(new TimeoutException());
687 case PRE_COMMIT_COMPLETE:
688 // FIXME: this is a legacy behavior problem. Three-phase commit protocol specifies that after we
689 // are ready we should commit the transaction, not abort it. Our current software stack does
690 // not allow us to do that consistently, because we persist at the time of commit, hence
691 // we can end up in a state where we have pre-committed a transaction, then a leader failover
692 // occurred ... the new leader does not see the pre-committed transaction and does not have
693 // a running timer. To fix this we really need two persistence events.
695 // The first one, done at pre-commit time will hold the transaction payload. When consensus
696 // is reached, we exit the pre-commit phase and start the pre-commit timer. Followers do not
697 // apply the state in this event.
699 // The second one, done at commit (or abort) time holds only the transaction identifier and
700 // signals to followers that the state should (or should not) be applied.
702 // In order to make the pre-commit timer working across failovers, though, we need
703 // a per-shard cluster-wide monotonic time, so a follower becoming the leader can accurately
704 // restart the timer.
705 pendingTransactions.poll().cohort.reportFailure(new TimeoutException());
708 LOG.warn("{}: Transaction {} is still committing, cannot abort", logContext,
709 currentTx.cohort.getIdentifier());
710 currentTx.lastAccess = now;
718 pendingTransactions.poll();
722 processNextTransaction();
727 void startAbort(final SimpleShardDataTreeCohort cohort) {
728 final Iterator<CommitEntry> it = pendingTransactions.iterator();
730 LOG.debug("{}: no open transaction while attempting to abort {}", logContext, cohort.getIdentifier());
734 // First entry is special, as it may already be committing
735 final CommitEntry first = it.next();
736 if (cohort.equals(first.cohort)) {
737 if (cohort.getState() != State.COMMIT_PENDING) {
738 LOG.debug("{}: aborted head of queue {} in state {}", logContext, cohort.getIdentifier(),
739 cohort.getIdentifier());
740 pendingTransactions.poll();
741 processNextTransaction();
743 LOG.warn("{}: transaction {} is committing, skipping abort", logContext, cohort.getIdentifier());
749 while (it.hasNext()) {
750 final CommitEntry e = it.next();
751 if (cohort.equals(e.cohort)) {
752 LOG.debug("{}: aborting queued transaction {}", logContext, cohort.getIdentifier());
758 LOG.debug("{}: aborted transaction {} not found in the queue", logContext, cohort.getIdentifier());
761 void setRunOnPendingTransactionsComplete(final Runnable operation) {
762 runOnPendingTransactionsComplete = operation;
763 maybeRunOperationOnPendingTransactionsComplete();
766 private void maybeRunOperationOnPendingTransactionsComplete() {
767 if (runOnPendingTransactionsComplete != null && pendingTransactions.isEmpty()) {
768 LOG.debug("{}: Pending transactions complete - running operation {}", logContext,
769 runOnPendingTransactionsComplete);
771 runOnPendingTransactionsComplete.run();
772 runOnPendingTransactionsComplete = null;