2 * Copyright (c) 2014 Cisco Systems, Inc. and others. All rights reserved.
3 * Copyright (c) 2015 Brocade Communications Systems, Inc. and others. All rights reserved.
5 * This program and the accompanying materials are made available under the
6 * terms of the Eclipse Public License v1.0 which accompanies this distribution,
7 * and is available at http://www.eclipse.org/legal/epl-v10.html
10 package org.opendaylight.controller.cluster.raft;
12 import akka.actor.ActorRef;
13 import akka.actor.ActorSelection;
14 import akka.actor.PoisonPill;
15 import akka.japi.Procedure;
16 import com.google.common.annotations.VisibleForTesting;
17 import com.google.common.base.Objects;
18 import com.google.common.base.Optional;
19 import com.google.common.base.Supplier;
20 import com.google.common.collect.Lists;
21 import java.io.Serializable;
22 import java.util.Collection;
23 import java.util.HashMap;
24 import java.util.List;
26 import java.util.concurrent.TimeUnit;
27 import javax.annotation.Nonnull;
28 import org.apache.commons.lang3.time.DurationFormatUtils;
29 import org.opendaylight.controller.cluster.DataPersistenceProvider;
30 import org.opendaylight.controller.cluster.DelegatingPersistentDataProvider;
31 import org.opendaylight.controller.cluster.NonPersistentDataProvider;
32 import org.opendaylight.controller.cluster.PersistentDataProvider;
33 import org.opendaylight.controller.cluster.common.actor.AbstractUntypedPersistentActor;
34 import org.opendaylight.controller.cluster.notifications.LeaderStateChanged;
35 import org.opendaylight.controller.cluster.notifications.RoleChanged;
36 import org.opendaylight.controller.cluster.raft.base.messages.ApplyJournalEntries;
37 import org.opendaylight.controller.cluster.raft.base.messages.ApplyState;
38 import org.opendaylight.controller.cluster.raft.base.messages.InitiateCaptureSnapshot;
39 import org.opendaylight.controller.cluster.raft.base.messages.LeaderTransitioning;
40 import org.opendaylight.controller.cluster.raft.base.messages.Replicate;
41 import org.opendaylight.controller.cluster.raft.base.messages.SwitchBehavior;
42 import org.opendaylight.controller.cluster.raft.behaviors.AbstractLeader;
43 import org.opendaylight.controller.cluster.raft.behaviors.AbstractRaftActorBehavior;
44 import org.opendaylight.controller.cluster.raft.behaviors.Follower;
45 import org.opendaylight.controller.cluster.raft.behaviors.RaftActorBehavior;
46 import org.opendaylight.controller.cluster.raft.client.messages.FindLeader;
47 import org.opendaylight.controller.cluster.raft.client.messages.FindLeaderReply;
48 import org.opendaylight.controller.cluster.raft.client.messages.FollowerInfo;
49 import org.opendaylight.controller.cluster.raft.client.messages.GetOnDemandRaftState;
50 import org.opendaylight.controller.cluster.raft.client.messages.OnDemandRaftState;
51 import org.opendaylight.controller.cluster.raft.client.messages.Shutdown;
52 import org.opendaylight.controller.cluster.raft.protobuff.client.messages.Payload;
53 import org.slf4j.Logger;
54 import org.slf4j.LoggerFactory;
57 * RaftActor encapsulates a state machine that needs to be kept synchronized
58 * in a cluster. It implements the RAFT algorithm as described in the paper
59 * <a href='https://ramcloud.stanford.edu/wiki/download/attachments/11370504/raft.pdf'>
60 * In Search of an Understandable Consensus Algorithm</a>
62 * RaftActor has 3 states and each state has a certain behavior associated
63 * with it. A Raft actor can behave as,
66 * <li> A Follower (or) </li>
67 * <li> A Candidate </li>
71 * A RaftActor MUST be a Leader in order to accept requests from clients to
72 * change the state of it's encapsulated state machine. Once a RaftActor becomes
73 * a Leader it is also responsible for ensuring that all followers ultimately
74 * have the same log and therefore the same state machine as itself.
77 * The current behavior of a RaftActor determines how election for leadership
78 * is initiated and how peer RaftActors react to request for votes.
81 * Each RaftActor also needs to know the current election term. It uses this
82 * information for a couple of things. One is to simply figure out who it
83 * voted for in the last election. Another is to figure out if the message
84 * it received to update it's state is stale.
87 * The RaftActor uses akka-persistence to store it's replicated log.
88 * Furthermore through it's behaviors a Raft Actor determines
91 * <li> when a log entry should be persisted </li>
92 * <li> when a log entry should be applied to the state machine (and) </li>
93 * <li> when a snapshot should be saved </li>
96 public abstract class RaftActor extends AbstractUntypedPersistentActor {
98 private static final long APPLY_STATE_DELAY_THRESHOLD_IN_NANOS = TimeUnit.MILLISECONDS.toNanos(50L); // 50 millis
100 protected final Logger LOG = LoggerFactory.getLogger(getClass());
103 * This context should NOT be passed directly to any other actor it is
104 * only to be consumed by the RaftActorBehaviors
106 private final RaftActorContextImpl context;
108 private final DelegatingPersistentDataProvider delegatingPersistenceProvider;
110 private final PersistentDataProvider persistentProvider;
112 private RaftActorRecoverySupport raftRecovery;
114 private RaftActorSnapshotMessageSupport snapshotSupport;
116 private final BehaviorStateHolder reusableBehaviorStateHolder = new BehaviorStateHolder();
118 private final SwitchBehaviorSupplier reusableSwitchBehaviorSupplier = new SwitchBehaviorSupplier();
120 private RaftActorServerConfigurationSupport serverConfigurationSupport;
122 private RaftActorLeadershipTransferCohort leadershipTransferInProgress;
124 private boolean shuttingDown;
126 public RaftActor(String id, Map<String, String> peerAddresses,
127 Optional<ConfigParams> configParams, short payloadVersion) {
129 persistentProvider = new PersistentDataProvider(this);
130 delegatingPersistenceProvider = new RaftActorDelegatingPersistentDataProvider(null, persistentProvider);
132 context = new RaftActorContextImpl(this.getSelf(),
133 this.getContext(), id, new ElectionTermImpl(persistentProvider, id, LOG),
134 -1, -1, peerAddresses,
135 (configParams.isPresent() ? configParams.get(): new DefaultConfigParamsImpl()),
136 delegatingPersistenceProvider, LOG);
138 context.setPayloadVersion(payloadVersion);
139 context.setReplicatedLog(ReplicatedLogImpl.newInstance(context));
143 public void preStart() throws Exception {
144 LOG.info("Starting recovery for {} with journal batch size {}", persistenceId(),
145 context.getConfigParams().getJournalRecoveryLogBatchSize());
149 snapshotSupport = newRaftActorSnapshotMessageSupport();
150 serverConfigurationSupport = new RaftActorServerConfigurationSupport(this);
154 public void postStop() {
160 protected void handleRecover(Object message) {
161 if(raftRecovery == null) {
162 raftRecovery = newRaftActorRecoverySupport();
165 boolean recoveryComplete = raftRecovery.handleRecoveryMessage(message, persistentProvider);
166 if(recoveryComplete) {
167 onRecoveryComplete();
169 initializeBehavior();
173 if (context.getReplicatedLog().size() > 0) {
174 self().tell(new InitiateCaptureSnapshot(), self());
175 LOG.info("{}: Snapshot capture initiated after recovery", persistenceId());
177 LOG.info("{}: Snapshot capture NOT initiated after recovery, journal empty", persistenceId());
182 protected RaftActorRecoverySupport newRaftActorRecoverySupport() {
183 return new RaftActorRecoverySupport(context, getRaftActorRecoveryCohort());
187 void initializeBehavior(){
188 changeCurrentBehavior(new Follower(context));
192 protected void changeCurrentBehavior(RaftActorBehavior newBehavior){
193 if(getCurrentBehavior() != null) {
195 getCurrentBehavior().close();
196 } catch(Exception e) {
197 LOG.warn("{}: Error closing behavior {}", persistence(), getCurrentBehavior(), e);
201 reusableBehaviorStateHolder.init(getCurrentBehavior());
202 setCurrentBehavior(newBehavior);
203 handleBehaviorChange(reusableBehaviorStateHolder, getCurrentBehavior());
207 protected void handleCommand(final Object message) {
208 if (serverConfigurationSupport.handleMessage(message, getSender())) {
211 if (snapshotSupport.handleSnapshotMessage(message, getSender())) {
215 if (message instanceof ApplyState) {
216 ApplyState applyState = (ApplyState) message;
218 long elapsedTime = (System.nanoTime() - applyState.getStartTime());
219 if(elapsedTime >= APPLY_STATE_DELAY_THRESHOLD_IN_NANOS){
220 LOG.warn("ApplyState took more time than expected. Elapsed Time = {} ms ApplyState = {}",
221 TimeUnit.NANOSECONDS.toMillis(elapsedTime), applyState);
224 if(LOG.isDebugEnabled()) {
225 LOG.debug("{}: Applying state for log index {} data {}",
226 persistenceId(), applyState.getReplicatedLogEntry().getIndex(),
227 applyState.getReplicatedLogEntry().getData());
230 applyState(applyState.getClientActor(), applyState.getIdentifier(),
231 applyState.getReplicatedLogEntry().getData());
233 if (!hasFollowers()) {
234 // for single node, the capture should happen after the apply state
235 // as we delete messages from the persistent journal which have made it to the snapshot
236 // capturing the snapshot before applying makes the persistent journal and snapshot out of sync
237 // and recovery shows data missing
238 context.getReplicatedLog().captureSnapshotIfReady(applyState.getReplicatedLogEntry());
240 context.getSnapshotManager().trimLog(context.getLastApplied());
243 } else if (message instanceof ApplyJournalEntries) {
244 ApplyJournalEntries applyEntries = (ApplyJournalEntries) message;
245 if(LOG.isDebugEnabled()) {
246 LOG.debug("{}: Persisting ApplyJournalEntries with index={}", persistenceId(), applyEntries.getToIndex());
249 persistence().persist(applyEntries, NoopProcedure.instance());
251 } else if (message instanceof FindLeader) {
253 new FindLeaderReply(getLeaderAddress()),
256 } else if(message instanceof GetOnDemandRaftState) {
257 onGetOnDemandRaftStats();
258 } else if(message instanceof InitiateCaptureSnapshot) {
260 } else if(message instanceof SwitchBehavior) {
261 switchBehavior(((SwitchBehavior) message));
262 } else if(message instanceof LeaderTransitioning) {
263 onLeaderTransitioning();
264 } else if(message instanceof Shutdown) {
266 } else if(message instanceof Runnable) {
267 ((Runnable)message).run();
269 switchBehavior(reusableSwitchBehaviorSupplier.handleMessage(getSender(), message));
273 void initiateLeadershipTransfer(final RaftActorLeadershipTransferCohort.OnComplete onComplete) {
274 LOG.debug("{}: Initiating leader transfer", persistenceId());
276 if(leadershipTransferInProgress == null) {
277 leadershipTransferInProgress = new RaftActorLeadershipTransferCohort(this, getSender());
278 leadershipTransferInProgress.addOnComplete(new RaftActorLeadershipTransferCohort.OnComplete() {
280 public void onSuccess(ActorRef raftActorRef, ActorRef replyTo) {
281 leadershipTransferInProgress = null;
285 public void onFailure(ActorRef raftActorRef, ActorRef replyTo) {
286 leadershipTransferInProgress = null;
290 leadershipTransferInProgress.addOnComplete(onComplete);
291 leadershipTransferInProgress.init();
293 LOG.debug("{}: prior leader transfer in progress - adding callback", persistenceId());
294 leadershipTransferInProgress.addOnComplete(onComplete);
298 private void onShutDown() {
299 LOG.debug("{}: onShutDown", persistenceId());
307 final RaftActorBehavior currentBehavior = context.getCurrentBehavior();
308 if(currentBehavior.state() == RaftState.Leader && context.hasFollowers()) {
309 initiateLeadershipTransfer(new RaftActorLeadershipTransferCohort.OnComplete() {
311 public void onSuccess(ActorRef raftActorRef, ActorRef replyTo) {
312 LOG.debug("{}: leader transfer succeeded - sending PoisonPill", persistenceId());
313 raftActorRef.tell(PoisonPill.getInstance(), raftActorRef);
317 public void onFailure(ActorRef raftActorRef, ActorRef replyTo) {
318 LOG.debug("{}: leader transfer failed - sending PoisonPill", persistenceId());
319 raftActorRef.tell(PoisonPill.getInstance(), raftActorRef);
322 } else if(currentBehavior.state() == RaftState.Leader) {
323 pauseLeader(new TimedRunnable(context.getConfigParams().getElectionTimeOutInterval(), this) {
325 protected void doRun() {
326 self().tell(PoisonPill.getInstance(), self());
330 protected void doCancel() {
331 self().tell(PoisonPill.getInstance(), self());
335 self().tell(PoisonPill.getInstance(), self());
339 private void onLeaderTransitioning() {
340 LOG.debug("{}: onLeaderTransitioning", persistenceId());
341 Optional<ActorRef> roleChangeNotifier = getRoleChangeNotifier();
342 if(getRaftState() == RaftState.Follower && roleChangeNotifier.isPresent()) {
343 roleChangeNotifier.get().tell(newLeaderStateChanged(getId(), null,
344 getCurrentBehavior().getLeaderPayloadVersion()), getSelf());
348 private void switchBehavior(SwitchBehavior message) {
349 if(!getRaftActorContext().getRaftPolicy().automaticElectionsEnabled()) {
350 RaftState newState = message.getNewState();
351 if( newState == RaftState.Leader || newState == RaftState.Follower) {
352 switchBehavior(reusableSwitchBehaviorSupplier.handleMessage(getSender(), message));
353 getRaftActorContext().getTermInformation().updateAndPersist(message.getNewTerm(), "");
355 LOG.warn("Switching to behavior : {} - not supported", newState);
360 private void switchBehavior(Supplier<RaftActorBehavior> supplier){
361 reusableBehaviorStateHolder.init(getCurrentBehavior());
363 setCurrentBehavior(supplier.get());
365 handleBehaviorChange(reusableBehaviorStateHolder, getCurrentBehavior());
369 RaftActorSnapshotMessageSupport newRaftActorSnapshotMessageSupport() {
370 return new RaftActorSnapshotMessageSupport(context, getRaftActorSnapshotCohort());
373 private void onGetOnDemandRaftStats() {
374 // Debugging message to retrieve raft stats.
376 Map<String, String> peerAddresses = new HashMap<>();
377 for(String peerId: context.getPeerIds()) {
378 peerAddresses.put(peerId, context.getPeerAddress(peerId));
381 final RaftActorBehavior currentBehavior = context.getCurrentBehavior();
382 OnDemandRaftState.Builder builder = OnDemandRaftState.builder()
383 .commitIndex(context.getCommitIndex())
384 .currentTerm(context.getTermInformation().getCurrentTerm())
385 .inMemoryJournalDataSize(replicatedLog().dataSize())
386 .inMemoryJournalLogSize(replicatedLog().size())
387 .isSnapshotCaptureInitiated(context.getSnapshotManager().isCapturing())
388 .lastApplied(context.getLastApplied())
389 .lastIndex(replicatedLog().lastIndex())
390 .lastTerm(replicatedLog().lastTerm())
391 .leader(getLeaderId())
392 .raftState(currentBehavior.state().toString())
393 .replicatedToAllIndex(currentBehavior.getReplicatedToAllIndex())
394 .snapshotIndex(replicatedLog().getSnapshotIndex())
395 .snapshotTerm(replicatedLog().getSnapshotTerm())
396 .votedFor(context.getTermInformation().getVotedFor())
397 .peerAddresses(peerAddresses)
398 .customRaftPolicyClassName(context.getConfigParams().getCustomRaftPolicyImplementationClass());
400 ReplicatedLogEntry lastLogEntry = replicatedLog().last();
401 if (lastLogEntry != null) {
402 builder.lastLogIndex(lastLogEntry.getIndex());
403 builder.lastLogTerm(lastLogEntry.getTerm());
406 if(getCurrentBehavior() instanceof AbstractLeader) {
407 AbstractLeader leader = (AbstractLeader)getCurrentBehavior();
408 Collection<String> followerIds = leader.getFollowerIds();
409 List<FollowerInfo> followerInfoList = Lists.newArrayListWithCapacity(followerIds.size());
410 for(String id: followerIds) {
411 final FollowerLogInformation info = leader.getFollower(id);
412 followerInfoList.add(new FollowerInfo(id, info.getNextIndex(), info.getMatchIndex(),
413 info.isFollowerActive(), DurationFormatUtils.formatDurationHMS(info.timeSinceLastActivity())));
416 builder.followerInfoList(followerInfoList);
419 sender().tell(builder.build(), self());
423 private void handleBehaviorChange(BehaviorStateHolder oldBehaviorState, RaftActorBehavior currentBehavior) {
424 RaftActorBehavior oldBehavior = oldBehaviorState.getBehavior();
426 if (oldBehavior != currentBehavior){
430 String lastValidLeaderId = oldBehavior == null ? null : oldBehaviorState.getLastValidLeaderId();
431 String oldBehaviorStateName = oldBehavior == null ? null : oldBehavior.state().name();
433 // it can happen that the state has not changed but the leader has changed.
434 Optional<ActorRef> roleChangeNotifier = getRoleChangeNotifier();
435 if(!Objects.equal(lastValidLeaderId, currentBehavior.getLeaderId()) ||
436 oldBehaviorState.getLeaderPayloadVersion() != currentBehavior.getLeaderPayloadVersion()) {
437 if(roleChangeNotifier.isPresent()) {
438 roleChangeNotifier.get().tell(newLeaderStateChanged(getId(), currentBehavior.getLeaderId(),
439 currentBehavior.getLeaderPayloadVersion()), getSelf());
442 onLeaderChanged(lastValidLeaderId, currentBehavior.getLeaderId());
444 if(leadershipTransferInProgress != null) {
445 leadershipTransferInProgress.onNewLeader(currentBehavior.getLeaderId());
448 serverConfigurationSupport.onNewLeader(currentBehavior.getLeaderId());
451 if (roleChangeNotifier.isPresent() &&
452 (oldBehavior == null || (oldBehavior.state() != currentBehavior.state()))) {
453 roleChangeNotifier.get().tell(new RoleChanged(getId(), oldBehaviorStateName ,
454 currentBehavior.state().name()), getSelf());
458 protected LeaderStateChanged newLeaderStateChanged(String memberId, String leaderId, short leaderPayloadVersion) {
459 return new LeaderStateChanged(memberId, leaderId, leaderPayloadVersion);
463 public long snapshotSequenceNr() {
464 // When we do a snapshot capture, we also capture and save the sequence-number of the persistent journal,
465 // so that we can delete the persistent journal based on the saved sequence-number
466 // However , when akka replays the journal during recovery, it replays it from the sequence number when the snapshot
467 // was saved and not the number we saved.
468 // We would want to override it , by asking akka to use the last-sequence number known to us.
469 return context.getSnapshotManager().getLastSequenceNumber();
473 * When a derived RaftActor needs to persist something it must call
480 protected void persistData(final ActorRef clientActor, final String identifier,
481 final Payload data) {
483 ReplicatedLogEntry replicatedLogEntry = new ReplicatedLogImplEntry(
484 context.getReplicatedLog().lastIndex() + 1,
485 context.getTermInformation().getCurrentTerm(), data);
487 if(LOG.isDebugEnabled()) {
488 LOG.debug("{}: Persist data {}", persistenceId(), replicatedLogEntry);
491 final RaftActorContext raftContext = getRaftActorContext();
493 replicatedLog().appendAndPersist(replicatedLogEntry, new Procedure<ReplicatedLogEntry>() {
495 public void apply(ReplicatedLogEntry replicatedLogEntry) {
496 if (!hasFollowers()){
497 // Increment the Commit Index and the Last Applied values
498 raftContext.setCommitIndex(replicatedLogEntry.getIndex());
499 raftContext.setLastApplied(replicatedLogEntry.getIndex());
501 // Apply the state immediately.
502 self().tell(new ApplyState(clientActor, identifier, replicatedLogEntry), self());
504 // Send a ApplyJournalEntries message so that we write the fact that we applied
505 // the state to durable storage
506 self().tell(new ApplyJournalEntries(replicatedLogEntry.getIndex()), self());
508 } else if (clientActor != null) {
509 context.getReplicatedLog().captureSnapshotIfReady(replicatedLogEntry);
511 // Send message for replication
512 getCurrentBehavior().handleMessage(getSelf(),
513 new Replicate(clientActor, identifier, replicatedLogEntry));
519 private ReplicatedLog replicatedLog() {
520 return context.getReplicatedLog();
523 protected String getId() {
524 return context.getId();
528 void setCurrentBehavior(RaftActorBehavior behavior) {
529 context.setCurrentBehavior(behavior);
532 protected RaftActorBehavior getCurrentBehavior() {
533 return context.getCurrentBehavior();
537 * Derived actors can call the isLeader method to check if the current
538 * RaftActor is the Leader or not
540 * @return true it this RaftActor is a Leader false otherwise
542 protected boolean isLeader() {
543 return context.getId().equals(getCurrentBehavior().getLeaderId());
546 protected final boolean isLeaderActive() {
547 return getRaftState() != RaftState.IsolatedLeader && !shuttingDown &&
548 !isLeadershipTransferInProgress();
551 private boolean isLeadershipTransferInProgress() {
552 return leadershipTransferInProgress != null && leadershipTransferInProgress.isTransferring();
556 * Derived actor can call getLeader if they need a reference to the Leader.
557 * This would be useful for example in forwarding a request to an actor
558 * which is the leader
560 * @return A reference to the leader if known, null otherwise
562 protected ActorSelection getLeader(){
563 String leaderAddress = getLeaderAddress();
565 if(leaderAddress == null){
569 return context.actorSelection(leaderAddress);
574 * @return the current leader's id
576 protected final String getLeaderId(){
577 return getCurrentBehavior().getLeaderId();
581 protected final RaftState getRaftState() {
582 return getCurrentBehavior().state();
585 protected Long getCurrentTerm(){
586 return context.getTermInformation().getCurrentTerm();
589 protected RaftActorContext getRaftActorContext() {
593 protected void updateConfigParams(ConfigParams configParams) {
595 // obtain the RaftPolicy for oldConfigParams and the updated one.
596 String oldRaftPolicy = context.getConfigParams().
597 getCustomRaftPolicyImplementationClass();
598 String newRaftPolicy = configParams.
599 getCustomRaftPolicyImplementationClass();
601 LOG.debug("{}: RaftPolicy used with prev.config {}, RaftPolicy used with newConfig {}", persistenceId(),
602 oldRaftPolicy, newRaftPolicy);
603 context.setConfigParams(configParams);
604 if (!Objects.equal(oldRaftPolicy, newRaftPolicy)) {
605 // The RaftPolicy was modified. If the current behavior is Follower then re-initialize to Follower
606 // but transfer the previous leaderId so it doesn't immediately try to schedule an election. This
607 // avoids potential disruption. Otherwise, switch to Follower normally.
608 RaftActorBehavior behavior = getCurrentBehavior();
609 if(behavior instanceof Follower) {
610 String previousLeaderId = ((Follower)behavior).getLeaderId();
611 short previousLeaderPayloadVersion = behavior.getLeaderPayloadVersion();
613 LOG.debug("{}: Re-initializing to Follower with previous leaderId {}", persistenceId(), previousLeaderId);
615 changeCurrentBehavior(new Follower(context, previousLeaderId, previousLeaderPayloadVersion));
617 initializeBehavior();
622 public final DataPersistenceProvider persistence() {
623 return delegatingPersistenceProvider.getDelegate();
626 public void setPersistence(DataPersistenceProvider provider) {
627 delegatingPersistenceProvider.setDelegate(provider);
630 protected void setPersistence(boolean persistent) {
632 setPersistence(new PersistentDataProvider(this));
634 setPersistence(new NonPersistentDataProvider() {
636 * The way snapshotting works is,
638 * <li> RaftActor calls createSnapshot on the Shard
639 * <li> Shard sends a CaptureSnapshotReply and RaftActor then calls saveSnapshot
640 * <li> When saveSnapshot is invoked on the akka-persistence API it uses the SnapshotStore to save
641 * the snapshot. The SnapshotStore sends SaveSnapshotSuccess or SaveSnapshotFailure. When the
642 * RaftActor gets SaveSnapshot success it commits the snapshot to the in-memory journal. This
643 * commitSnapshot is mimicking what is done in SaveSnapshotSuccess.
647 public void saveSnapshot(Object o) {
648 // Make saving Snapshot successful
649 // Committing the snapshot here would end up calling commit in the creating state which would
650 // be a state violation. That's why now we send a message to commit the snapshot.
651 self().tell(RaftActorSnapshotMessageSupport.COMMIT_SNAPSHOT, self());
658 * setPeerAddress sets the address of a known peer at a later time.
660 * This is to account for situations where a we know that a peer
661 * exists but we do not know an address up-front. This may also be used in
662 * situations where a known peer starts off in a different location and we
663 * need to change it's address
665 * Note that if the peerId does not match the list of peers passed to
666 * this actor during construction an IllegalStateException will be thrown.
671 protected void setPeerAddress(String peerId, String peerAddress){
672 context.setPeerAddress(peerId, peerAddress);
676 * The applyState method will be called by the RaftActor when some data
677 * needs to be applied to the actor's state
679 * @param clientActor A reference to the client who sent this message. This
680 * is the same reference that was passed to persistData
681 * by the derived actor. clientActor may be null when
682 * the RaftActor is behaving as a follower or during
684 * @param identifier The identifier of the persisted data. This is also
685 * the same identifier that was passed to persistData by
686 * the derived actor. identifier may be null when
687 * the RaftActor is behaving as a follower or during
689 * @param data A piece of data that was persisted by the persistData call.
690 * This should NEVER be null.
692 protected abstract void applyState(ActorRef clientActor, String identifier,
696 * Returns the RaftActorRecoveryCohort to participate in persistence recovery.
699 protected abstract RaftActorRecoveryCohort getRaftActorRecoveryCohort();
702 * This method is called when recovery is complete.
704 protected abstract void onRecoveryComplete();
707 * Returns the RaftActorSnapshotCohort to participate in persistence recovery.
710 protected abstract RaftActorSnapshotCohort getRaftActorSnapshotCohort();
713 * This method will be called by the RaftActor when the state of the
714 * RaftActor changes. The derived actor can then use methods like
715 * isLeader or getLeader to do something useful
717 protected abstract void onStateChanged();
720 * Notifier Actor for this RaftActor to notify when a role change happens
721 * @return ActorRef - ActorRef of the notifier or Optional.absent if none.
723 protected abstract Optional<ActorRef> getRoleChangeNotifier();
726 * This method is called prior to operations such as leadership transfer and actor shutdown when the leader
727 * must pause or stop its duties. This method allows derived classes to gracefully pause or finish current
728 * work prior to performing the operation. On completion of any work, the run method must be called on the
729 * given Runnable to proceed with the given operation. <b>Important:</b> the run method must be called on
730 * this actor's thread dispatcher as as it modifies internal state.
732 * The default implementation immediately runs the operation.
734 * @param operation the operation to run
736 protected void pauseLeader(Runnable operation) {
740 protected void onLeaderChanged(String oldLeader, String newLeader) {
744 private String getLeaderAddress(){
746 return getSelf().path().toString();
748 String leaderId = getLeaderId();
749 if (leaderId == null) {
752 String peerAddress = context.getPeerAddress(leaderId);
753 if(LOG.isDebugEnabled()) {
754 LOG.debug("{}: getLeaderAddress leaderId = {} peerAddress = {}",
755 persistenceId(), leaderId, peerAddress);
761 protected boolean hasFollowers(){
762 return getRaftActorContext().hasFollowers();
765 private void captureSnapshot() {
766 SnapshotManager snapshotManager = context.getSnapshotManager();
768 if (!snapshotManager.isCapturing()) {
769 final long idx = getCurrentBehavior().getReplicatedToAllIndex();
770 LOG.debug("Take a snapshot of current state. lastReplicatedLog is {} and replicatedToAllIndex is {}",
771 replicatedLog().last(), idx);
773 snapshotManager.capture(replicatedLog().last(), idx);
778 * @deprecated Deprecated in favor of {@link org.opendaylight.controller.cluster.raft.base.messages.DeleteEntries}
779 * whose type for fromIndex is long instead of int. This class was kept for backwards
780 * compatibility with Helium.
782 // Suppressing this warning as we can't set serialVersionUID to maintain backwards compatibility.
783 @SuppressWarnings("serial")
785 static class DeleteEntries implements Serializable {
786 private final int fromIndex;
788 public DeleteEntries(int fromIndex) {
789 this.fromIndex = fromIndex;
792 public int getFromIndex() {
798 * @deprecated Deprecated in favor of non-inner class {@link org.opendaylight.controller.cluster.raft.base.messages.UpdateElectionTerm}
799 * which has serialVersionUID set. This class was kept for backwards compatibility with Helium.
801 // Suppressing this warning as we can't set serialVersionUID to maintain backwards compatibility.
802 @SuppressWarnings("serial")
804 static class UpdateElectionTerm implements Serializable {
805 private final long currentTerm;
806 private final String votedFor;
808 public UpdateElectionTerm(long currentTerm, String votedFor) {
809 this.currentTerm = currentTerm;
810 this.votedFor = votedFor;
813 public long getCurrentTerm() {
817 public String getVotedFor() {
822 private static class BehaviorStateHolder {
823 private RaftActorBehavior behavior;
824 private String lastValidLeaderId;
825 private short leaderPayloadVersion;
827 void init(RaftActorBehavior behavior) {
828 this.behavior = behavior;
829 this.leaderPayloadVersion = behavior != null ? behavior.getLeaderPayloadVersion() : -1;
831 String behaviorLeaderId = behavior != null ? behavior.getLeaderId() : null;
832 if(behaviorLeaderId != null) {
833 this.lastValidLeaderId = behaviorLeaderId;
837 RaftActorBehavior getBehavior() {
841 String getLastValidLeaderId() {
842 return lastValidLeaderId;
845 short getLeaderPayloadVersion() {
846 return leaderPayloadVersion;
850 private class SwitchBehaviorSupplier implements Supplier<RaftActorBehavior> {
851 private Object message;
852 private ActorRef sender;
854 public SwitchBehaviorSupplier handleMessage(ActorRef sender, Object message){
855 this.sender = sender;
856 this.message = message;
861 public RaftActorBehavior get() {
862 if(this.message instanceof SwitchBehavior){
863 return AbstractRaftActorBehavior.createBehavior(context, ((SwitchBehavior) message).getNewState());
865 return getCurrentBehavior().handleMessage(sender, message);