opendaylight/md-sal/sal-akka-raft/src/main/java/org/opendaylight/controller/cluster/raft/RaftActor.java

   1 /*
   2  * Copyright (c) 2014 Cisco Systems, Inc. and others.  All rights reserved.
   3  *
   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
   7  */
   8
   9 package org.opendaylight.controller.cluster.raft;
  10
  11 import akka.actor.ActorRef;
  12 import akka.actor.ActorSelection;
  13 import akka.event.Logging;
  14 import akka.event.LoggingAdapter;
  15 import akka.japi.Procedure;
  16 import akka.persistence.RecoveryCompleted;
  17 import akka.persistence.SaveSnapshotFailure;
  18 import akka.persistence.SaveSnapshotSuccess;
  19 import akka.persistence.SnapshotOffer;
  20 import akka.persistence.SnapshotSelectionCriteria;
  21 import com.google.common.annotations.VisibleForTesting;
  22 import com.google.common.base.Optional;
  23 import com.google.common.base.Stopwatch;
  24 import com.google.protobuf.ByteString;
  25 import java.io.Serializable;
  26 import java.util.Map;
  27 import org.opendaylight.controller.cluster.DataPersistenceProvider;
  28 import org.opendaylight.controller.cluster.common.actor.AbstractUntypedPersistentActor;
  29 import org.opendaylight.controller.cluster.notifications.RoleChanged;
  30 import org.opendaylight.controller.cluster.raft.base.messages.ApplyLogEntries;
  31 import org.opendaylight.controller.cluster.raft.base.messages.ApplySnapshot;
  32 import org.opendaylight.controller.cluster.raft.base.messages.ApplyState;
  33 import org.opendaylight.controller.cluster.raft.base.messages.CaptureSnapshot;
  34 import org.opendaylight.controller.cluster.raft.base.messages.CaptureSnapshotReply;
  35 import org.opendaylight.controller.cluster.raft.base.messages.Replicate;
  36 import org.opendaylight.controller.cluster.raft.base.messages.SendHeartBeat;
  37 import org.opendaylight.controller.cluster.raft.base.messages.SendInstallSnapshot;
  38 import org.opendaylight.controller.cluster.raft.behaviors.AbstractRaftActorBehavior;
  39 import org.opendaylight.controller.cluster.raft.behaviors.Follower;
  40 import org.opendaylight.controller.cluster.raft.behaviors.RaftActorBehavior;
  41 import org.opendaylight.controller.cluster.raft.client.messages.FindLeader;
  42 import org.opendaylight.controller.cluster.raft.client.messages.FindLeaderReply;
  43 import org.opendaylight.controller.cluster.raft.messages.AppendEntriesReply;
  44 import org.opendaylight.controller.cluster.raft.protobuff.client.messages.Payload;
  45 import org.opendaylight.controller.protobuff.messages.cluster.raft.AppendEntriesMessages;
  46
  47 /**
  48  * RaftActor encapsulates a state machine that needs to be kept synchronized
  49  * in a cluster. It implements the RAFT algorithm as described in the paper
  50  * <a href='https://ramcloud.stanford.edu/wiki/download/attachments/11370504/raft.pdf'>
  51  * In Search of an Understandable Consensus Algorithm</a>
  52  * <p/>
  53  * RaftActor has 3 states and each state has a certain behavior associated
  54  * with it. A Raft actor can behave as,
  55  * <ul>
  56  * <li> A Leader </li>
  57  * <li> A Follower (or) </li>
  58  * <li> A Candidate </li>
  59  * </ul>
  60  * <p/>
  61  * <p/>
  62  * A RaftActor MUST be a Leader in order to accept requests from clients to
  63  * change the state of it's encapsulated state machine. Once a RaftActor becomes
  64  * a Leader it is also responsible for ensuring that all followers ultimately
  65  * have the same log and therefore the same state machine as itself.
  66  * <p/>
  67  * <p/>
  68  * The current behavior of a RaftActor determines how election for leadership
  69  * is initiated and how peer RaftActors react to request for votes.
  70  * <p/>
  71  * <p/>
  72  * Each RaftActor also needs to know the current election term. It uses this
  73  * information for a couple of things. One is to simply figure out who it
  74  * voted for in the last election. Another is to figure out if the message
  75  * it received to update it's state is stale.
  76  * <p/>
  77  * <p/>
  78  * The RaftActor uses akka-persistence to store it's replicated log.
  79  * Furthermore through it's behaviors a Raft Actor determines
  80  * <p/>
  81  * <ul>
  82  * <li> when a log entry should be persisted </li>
  83  * <li> when a log entry should be applied to the state machine (and) </li>
  84  * <li> when a snapshot should be saved </li>
  85  * </ul>
  86  */
  87 public abstract class RaftActor extends AbstractUntypedPersistentActor {
  88     protected final LoggingAdapter LOG =
  89         Logging.getLogger(getContext().system(), this);
  90
  91     /**
  92      * The current state determines the current behavior of a RaftActor
  93      * A Raft Actor always starts off in the Follower State
  94      */
  95     private RaftActorBehavior currentBehavior;
  96
  97     /**
  98      * This context should NOT be passed directly to any other actor it is
  99      * only to be consumed by the RaftActorBehaviors
 100      */
 101     private final RaftActorContext context;
 102
 103     /**
 104      * The in-memory journal
 105      */
 106     private ReplicatedLogImpl replicatedLog = new ReplicatedLogImpl();
 107
 108     private CaptureSnapshot captureSnapshot = null;
 109
 110     private volatile boolean hasSnapshotCaptureInitiated = false;
 111
 112     private Stopwatch recoveryTimer;
 113
 114     private int currentRecoveryBatchCount;
 115
 116
 117
 118     public RaftActor(String id, Map<String, String> peerAddresses) {
 119         this(id, peerAddresses, Optional.<ConfigParams>absent());
 120     }
 121
 122     public RaftActor(String id, Map<String, String> peerAddresses,
 123          Optional<ConfigParams> configParams) {
 124
 125         context = new RaftActorContextImpl(this.getSelf(),
 126             this.getContext(), id, new ElectionTermImpl(),
 127             -1, -1, replicatedLog, peerAddresses,
 128             (configParams.isPresent() ? configParams.get(): new DefaultConfigParamsImpl()),
 129             LOG);
 130     }
 131
 132     private void initRecoveryTimer() {
 133         if(recoveryTimer == null) {
 134             recoveryTimer = new Stopwatch();
 135             recoveryTimer.start();
 136         }
 137     }
 138
 139     @Override
 140     public void preStart() throws Exception {
 141         LOG.info("Starting recovery for {} with journal batch size {}", persistenceId(),
 142                 context.getConfigParams().getJournalRecoveryLogBatchSize());
 143
 144         super.preStart();
 145     }
 146
 147     @Override
 148     public void handleRecover(Object message) {
 149         if(persistence().isRecoveryApplicable()) {
 150             if (message instanceof SnapshotOffer) {
 151                 onRecoveredSnapshot((SnapshotOffer) message);
 152             } else if (message instanceof ReplicatedLogEntry) {
 153                 onRecoveredJournalLogEntry((ReplicatedLogEntry) message);
 154             } else if (message instanceof ApplyLogEntries) {
 155                 onRecoveredApplyLogEntries((ApplyLogEntries) message);
 156             } else if (message instanceof DeleteEntries) {
 157                 replicatedLog.removeFrom(((DeleteEntries) message).getFromIndex());
 158             } else if (message instanceof UpdateElectionTerm) {
 159                 context.getTermInformation().update(((UpdateElectionTerm) message).getCurrentTerm(),
 160                         ((UpdateElectionTerm) message).getVotedFor());
 161             } else if (message instanceof RecoveryCompleted) {
 162                 onRecoveryCompletedMessage();
 163             }
 164         } else {
 165             if (message instanceof RecoveryCompleted) {
 166                 // Delete all the messages from the akka journal so that we do not end up with consistency issues
 167                 // Note I am not using the dataPersistenceProvider and directly using the akka api here
 168                 deleteMessages(lastSequenceNr());
 169
 170                 // Delete all the akka snapshots as they will not be needed
 171                 deleteSnapshots(new SnapshotSelectionCriteria(scala.Long.MaxValue(), scala.Long.MaxValue()));
 172
 173                 onRecoveryComplete();
 174
 175                 initializeBehavior();
 176             }
 177         }
 178     }
 179
 180     private void onRecoveredSnapshot(SnapshotOffer offer) {
 181         if(LOG.isDebugEnabled()) {
 182             LOG.debug("{}: SnapshotOffer called..", persistenceId());
 183         }
 184
 185         initRecoveryTimer();
 186
 187         Snapshot snapshot = (Snapshot) offer.snapshot();
 188
 189         // Create a replicated log with the snapshot information
 190         // The replicated log can be used later on to retrieve this snapshot
 191         // when we need to install it on a peer
 192         replicatedLog = new ReplicatedLogImpl(snapshot);
 193
 194         context.setReplicatedLog(replicatedLog);
 195         context.setLastApplied(snapshot.getLastAppliedIndex());
 196         context.setCommitIndex(snapshot.getLastAppliedIndex());
 197
 198         Stopwatch timer = new Stopwatch();
 199         timer.start();
 200
 201         // Apply the snapshot to the actors state
 202         applyRecoverySnapshot(snapshot.getState());
 203
 204         timer.stop();
 205         LOG.info("Recovery snapshot applied for {} in {}: snapshotIndex={}, snapshotTerm={}, journal-size=" +
 206                 replicatedLog.size(), persistenceId(), timer.toString(),
 207                 replicatedLog.getSnapshotIndex(), replicatedLog.getSnapshotTerm());
 208     }
 209
 210     private void onRecoveredJournalLogEntry(ReplicatedLogEntry logEntry) {
 211         if(LOG.isDebugEnabled()) {
 212             LOG.debug("{}: Received ReplicatedLogEntry for recovery: {}", persistenceId(), logEntry.getIndex());
 213         }
 214
 215         replicatedLog.append(logEntry);
 216     }
 217
 218     private void onRecoveredApplyLogEntries(ApplyLogEntries ale) {
 219         if(LOG.isDebugEnabled()) {
 220             LOG.debug("{}: Received ApplyLogEntries for recovery, applying to state: {} to {}",
 221                     persistenceId(), context.getLastApplied() + 1, ale.getToIndex());
 222         }
 223
 224         for (long i = context.getLastApplied() + 1; i <= ale.getToIndex(); i++) {
 225             batchRecoveredLogEntry(replicatedLog.get(i));
 226         }
 227
 228         context.setLastApplied(ale.getToIndex());
 229         context.setCommitIndex(ale.getToIndex());
 230     }
 231
 232     private void batchRecoveredLogEntry(ReplicatedLogEntry logEntry) {
 233         initRecoveryTimer();
 234
 235         int batchSize = context.getConfigParams().getJournalRecoveryLogBatchSize();
 236         if(currentRecoveryBatchCount == 0) {
 237             startLogRecoveryBatch(batchSize);
 238         }
 239
 240         appendRecoveredLogEntry(logEntry.getData());
 241
 242         if(++currentRecoveryBatchCount >= batchSize) {
 243             endCurrentLogRecoveryBatch();
 244         }
 245     }
 246
 247     private void endCurrentLogRecoveryBatch() {
 248         applyCurrentLogRecoveryBatch();
 249         currentRecoveryBatchCount = 0;
 250     }
 251
 252     private void onRecoveryCompletedMessage() {
 253         if(currentRecoveryBatchCount > 0) {
 254             endCurrentLogRecoveryBatch();
 255         }
 256
 257         onRecoveryComplete();
 258
 259         String recoveryTime = "";
 260         if(recoveryTimer != null) {
 261             recoveryTimer.stop();
 262             recoveryTime = " in " + recoveryTimer.toString();
 263             recoveryTimer = null;
 264         }
 265
 266         LOG.info(
 267             "Recovery completed" + recoveryTime + " - Switching actor to Follower - " +
 268                 "Persistence Id =  " + persistenceId() +
 269                 " Last index in log={}, snapshotIndex={}, snapshotTerm={}, " +
 270                 "journal-size={}",
 271             replicatedLog.lastIndex(), replicatedLog.getSnapshotIndex(),
 272             replicatedLog.getSnapshotTerm(), replicatedLog.size());
 273
 274         initializeBehavior();
 275     }
 276
 277     protected void initializeBehavior(){
 278         changeCurrentBehavior(new Follower(context));
 279     }
 280
 281     protected void changeCurrentBehavior(RaftActorBehavior newBehavior){
 282         RaftActorBehavior oldBehavior = currentBehavior;
 283         currentBehavior = newBehavior;
 284         handleBehaviorChange(oldBehavior, currentBehavior);
 285     }
 286
 287     @Override public void handleCommand(Object message) {
 288         if (message instanceof ApplyState){
 289             ApplyState applyState = (ApplyState) message;
 290
 291             if(LOG.isDebugEnabled()) {
 292                 LOG.debug("{}: Applying state for log index {} data {}",
 293                     persistenceId(), applyState.getReplicatedLogEntry().getIndex(),
 294                     applyState.getReplicatedLogEntry().getData());
 295             }
 296
 297             applyState(applyState.getClientActor(), applyState.getIdentifier(),
 298                 applyState.getReplicatedLogEntry().getData());
 299
 300         } else if (message instanceof ApplyLogEntries){
 301             ApplyLogEntries ale = (ApplyLogEntries) message;
 302             if(LOG.isDebugEnabled()) {
 303                 LOG.debug("{}: Persisting ApplyLogEntries with index={}", persistenceId(), ale.getToIndex());
 304             }
 305             persistence().persist(new ApplyLogEntries(ale.getToIndex()), new Procedure<ApplyLogEntries>() {
 306                 @Override
 307                 public void apply(ApplyLogEntries param) throws Exception {
 308                 }
 309             });
 310
 311         } else if(message instanceof ApplySnapshot ) {
 312             Snapshot snapshot = ((ApplySnapshot) message).getSnapshot();
 313
 314             if(LOG.isDebugEnabled()) {
 315                 LOG.debug("{}: ApplySnapshot called on Follower Actor " +
 316                         "snapshotIndex:{}, snapshotTerm:{}", persistenceId(), snapshot.getLastAppliedIndex(),
 317                     snapshot.getLastAppliedTerm()
 318                 );
 319             }
 320
 321             applySnapshot(snapshot.getState());
 322
 323             //clears the followers log, sets the snapshot index to ensure adjusted-index works
 324             replicatedLog = new ReplicatedLogImpl(snapshot);
 325             context.setReplicatedLog(replicatedLog);
 326             context.setLastApplied(snapshot.getLastAppliedIndex());
 327
 328         } else if (message instanceof FindLeader) {
 329             getSender().tell(
 330                 new FindLeaderReply(getLeaderAddress()),
 331                 getSelf()
 332             );
 333
 334         } else if (message instanceof SaveSnapshotSuccess) {
 335             SaveSnapshotSuccess success = (SaveSnapshotSuccess) message;
 336             LOG.info("{}: SaveSnapshotSuccess received for snapshot", persistenceId());
 337
 338             long sequenceNumber = success.metadata().sequenceNr();
 339
 340             commitSnapshot(sequenceNumber);
 341
 342         } else if (message instanceof SaveSnapshotFailure) {
 343             SaveSnapshotFailure saveSnapshotFailure = (SaveSnapshotFailure) message;
 344
 345             LOG.error(saveSnapshotFailure.cause(), "{}: SaveSnapshotFailure received for snapshot Cause:",
 346                     persistenceId());
 347
 348             context.getReplicatedLog().snapshotRollback();
 349
 350             LOG.info("{}: Replicated Log rollbacked. Snapshot will be attempted in the next cycle." +
 351                 "snapshotIndex:{}, snapshotTerm:{}, log-size:{}", persistenceId(),
 352                 context.getReplicatedLog().getSnapshotIndex(),
 353                 context.getReplicatedLog().getSnapshotTerm(),
 354                 context.getReplicatedLog().size());
 355
 356         } else if (message instanceof CaptureSnapshot) {
 357             LOG.info("{}: CaptureSnapshot received by actor", persistenceId());
 358
 359             if(captureSnapshot == null) {
 360                 captureSnapshot = (CaptureSnapshot)message;
 361                 createSnapshot();
 362             }
 363
 364         } else if (message instanceof CaptureSnapshotReply){
 365             handleCaptureSnapshotReply(((CaptureSnapshotReply) message).getSnapshot());
 366
 367         } else {
 368             if (!(message instanceof AppendEntriesMessages.AppendEntries)
 369                 && !(message instanceof AppendEntriesReply) && !(message instanceof SendHeartBeat)) {
 370                 if(LOG.isDebugEnabled()) {
 371                     LOG.debug("{}: onReceiveCommand: message: {}", persistenceId(), message.getClass());
 372                 }
 373             }
 374
 375             RaftActorBehavior oldBehavior = currentBehavior;
 376             currentBehavior = currentBehavior.handleMessage(getSender(), message);
 377
 378             handleBehaviorChange(oldBehavior, currentBehavior);
 379         }
 380     }
 381
 382     private void handleBehaviorChange(RaftActorBehavior oldBehavior, RaftActorBehavior currentBehavior) {
 383         if (oldBehavior != currentBehavior){
 384             onStateChanged();
 385         }
 386
 387         String oldBehaviorLeaderId = oldBehavior == null? null : oldBehavior.getLeaderId();
 388         String oldBehaviorState = oldBehavior == null? null : oldBehavior.state().name();
 389
 390         // it can happen that the state has not changed but the leader has changed.
 391         onLeaderChanged(oldBehaviorLeaderId, currentBehavior.getLeaderId());
 392
 393         if (getRoleChangeNotifier().isPresent() &&
 394                 (oldBehavior == null || (oldBehavior.state() != currentBehavior.state()))) {
 395             getRoleChangeNotifier().get().tell(
 396                     new RoleChanged(getId(), oldBehaviorState , currentBehavior.state().name()),
 397                     getSelf());
 398         }
 399     }
 400
 401     /**
 402      * When a derived RaftActor needs to persist something it must call
 403      * persistData.
 404      *
 405      * @param clientActor
 406      * @param identifier
 407      * @param data
 408      */
 409     protected void persistData(final ActorRef clientActor, final String identifier,
 410         final Payload data) {
 411
 412         ReplicatedLogEntry replicatedLogEntry = new ReplicatedLogImplEntry(
 413             context.getReplicatedLog().lastIndex() + 1,
 414             context.getTermInformation().getCurrentTerm(), data);
 415
 416         if(LOG.isDebugEnabled()) {
 417             LOG.debug("{}: Persist data {}", persistenceId(), replicatedLogEntry);
 418         }
 419
 420         final RaftActorContext raftContext = getRaftActorContext();
 421
 422         replicatedLog
 423                 .appendAndPersist(replicatedLogEntry, new Procedure<ReplicatedLogEntry>() {
 424                     @Override
 425                     public void apply(ReplicatedLogEntry replicatedLogEntry) throws Exception {
 426                         if(!hasFollowers()){
 427                             // Increment the Commit Index and the Last Applied values
 428                             raftContext.setCommitIndex(replicatedLogEntry.getIndex());
 429                             raftContext.setLastApplied(replicatedLogEntry.getIndex());
 430
 431                             // Apply the state immediately
 432                             applyState(clientActor, identifier, data);
 433
 434                             // Send a ApplyLogEntries message so that we write the fact that we applied
 435                             // the state to durable storage
 436                             self().tell(new ApplyLogEntries((int) replicatedLogEntry.getIndex()), self());
 437
 438                             // Check if the "real" snapshot capture has been initiated. If no then do the fake snapshot
 439                             if(!hasSnapshotCaptureInitiated){
 440                                 raftContext.getReplicatedLog().snapshotPreCommit(raftContext.getLastApplied(),
 441                                         raftContext.getTermInformation().getCurrentTerm());
 442                                 raftContext.getReplicatedLog().snapshotCommit();
 443                             } else {
 444                                 LOG.debug("{}: Skipping fake snapshotting for {} because real snapshotting is in progress",
 445                                         persistenceId(), getId());
 446                             }
 447                         } else if (clientActor != null) {
 448                             // Send message for replication
 449                             currentBehavior.handleMessage(getSelf(),
 450                                     new Replicate(clientActor, identifier,
 451                                             replicatedLogEntry)
 452                             );
 453                         }
 454
 455                     }
 456                 });    }
 457
 458     protected String getId() {
 459         return context.getId();
 460     }
 461
 462     /**
 463      * Derived actors can call the isLeader method to check if the current
 464      * RaftActor is the Leader or not
 465      *
 466      * @return true it this RaftActor is a Leader false otherwise
 467      */
 468     protected boolean isLeader() {
 469         return context.getId().equals(currentBehavior.getLeaderId());
 470     }
 471
 472     /**
 473      * Derived actor can call getLeader if they need a reference to the Leader.
 474      * This would be useful for example in forwarding a request to an actor
 475      * which is the leader
 476      *
 477      * @return A reference to the leader if known, null otherwise
 478      */
 479     protected ActorSelection getLeader(){
 480         String leaderAddress = getLeaderAddress();
 481
 482         if(leaderAddress == null){
 483             return null;
 484         }
 485
 486         return context.actorSelection(leaderAddress);
 487     }
 488
 489     /**
 490      *
 491      * @return the current leader's id
 492      */
 493     protected String getLeaderId(){
 494         return currentBehavior.getLeaderId();
 495     }
 496
 497     protected RaftState getRaftState() {
 498         return currentBehavior.state();
 499     }
 500
 501     protected ReplicatedLogEntry getLastLogEntry() {
 502         return replicatedLog.last();
 503     }
 504
 505     protected Long getCurrentTerm(){
 506         return context.getTermInformation().getCurrentTerm();
 507     }
 508
 509     protected Long getCommitIndex(){
 510         return context.getCommitIndex();
 511     }
 512
 513     protected Long getLastApplied(){
 514         return context.getLastApplied();
 515     }
 516
 517     protected RaftActorContext getRaftActorContext() {
 518         return context;
 519     }
 520
 521     /**
 522      * setPeerAddress sets the address of a known peer at a later time.
 523      * <p>
 524      * This is to account for situations where a we know that a peer
 525      * exists but we do not know an address up-front. This may also be used in
 526      * situations where a known peer starts off in a different location and we
 527      * need to change it's address
 528      * <p>
 529      * Note that if the peerId does not match the list of peers passed to
 530      * this actor during construction an IllegalStateException will be thrown.
 531      *
 532      * @param peerId
 533      * @param peerAddress
 534      */
 535     protected void setPeerAddress(String peerId, String peerAddress){
 536         context.setPeerAddress(peerId, peerAddress);
 537     }
 538
 539     protected void commitSnapshot(long sequenceNumber) {
 540         context.getReplicatedLog().snapshotCommit();
 541
 542         // TODO: Not sure if we want to be this aggressive with trimming stuff
 543         trimPersistentData(sequenceNumber);
 544     }
 545
 546     /**
 547      * The applyState method will be called by the RaftActor when some data
 548      * needs to be applied to the actor's state
 549      *
 550      * @param clientActor A reference to the client who sent this message. This
 551      *                    is the same reference that was passed to persistData
 552      *                    by the derived actor. clientActor may be null when
 553      *                    the RaftActor is behaving as a follower or during
 554      *                    recovery.
 555      * @param identifier  The identifier of the persisted data. This is also
 556      *                    the same identifier that was passed to persistData by
 557      *                    the derived actor. identifier may be null when
 558      *                    the RaftActor is behaving as a follower or during
 559      *                    recovery
 560      * @param data        A piece of data that was persisted by the persistData call.
 561      *                    This should NEVER be null.
 562      */
 563     protected abstract void applyState(ActorRef clientActor, String identifier,
 564         Object data);
 565
 566     /**
 567      * This method is called during recovery at the start of a batch of state entries. Derived
 568      * classes should perform any initialization needed to start a batch.
 569      */
 570     protected abstract void startLogRecoveryBatch(int maxBatchSize);
 571
 572     /**
 573      * This method is called during recovery to append state data to the current batch. This method
 574      * is called 1 or more times after {@link #startLogRecoveryBatch}.
 575      *
 576      * @param data the state data
 577      */
 578     protected abstract void appendRecoveredLogEntry(Payload data);
 579
 580     /**
 581      * This method is called during recovery to reconstruct the state of the actor.
 582      *
 583      * @param snapshot A snapshot of the state of the actor
 584      */
 585     protected abstract void applyRecoverySnapshot(byte[] snapshotBytes);
 586
 587     /**
 588      * This method is called during recovery at the end of a batch to apply the current batched
 589      * log entries. This method is called after {@link #appendRecoveredLogEntry}.
 590      */
 591     protected abstract void applyCurrentLogRecoveryBatch();
 592
 593     /**
 594      * This method is called when recovery is complete.
 595      */
 596     protected abstract void onRecoveryComplete();
 597
 598     /**
 599      * This method will be called by the RaftActor when a snapshot needs to be
 600      * created. The derived actor should respond with its current state.
 601      * <p/>
 602      * During recovery the state that is returned by the derived actor will
 603      * be passed back to it by calling the applySnapshot  method
 604      *
 605      * @return The current state of the actor
 606      */
 607     protected abstract void createSnapshot();
 608
 609     /**
 610      * This method can be called at any other point during normal
 611      * operations when the derived actor is out of sync with it's peers
 612      * and the only way to bring it in sync is by applying a snapshot
 613      *
 614      * @param snapshotBytes A snapshot of the state of the actor
 615      */
 616     protected abstract void applySnapshot(byte[] snapshotBytes);
 617
 618     /**
 619      * This method will be called by the RaftActor when the state of the
 620      * RaftActor changes. The derived actor can then use methods like
 621      * isLeader or getLeader to do something useful
 622      */
 623     protected abstract void onStateChanged();
 624
 625     protected abstract DataPersistenceProvider persistence();
 626
 627     /**
 628      * Notifier Actor for this RaftActor to notify when a role change happens
 629      * @return ActorRef - ActorRef of the notifier or Optional.absent if none.
 630      */
 631     protected abstract Optional<ActorRef> getRoleChangeNotifier();
 632
 633     protected void onLeaderChanged(String oldLeader, String newLeader){};
 634
 635     private void trimPersistentData(long sequenceNumber) {
 636         // Trim akka snapshots
 637         // FIXME : Not sure how exactly the SnapshotSelectionCriteria is applied
 638         // For now guessing that it is ANDed.
 639         persistence().deleteSnapshots(new SnapshotSelectionCriteria(
 640             sequenceNumber - context.getConfigParams().getSnapshotBatchCount(), 43200000));
 641
 642         // Trim akka journal
 643         persistence().deleteMessages(sequenceNumber);
 644     }
 645
 646     private String getLeaderAddress(){
 647         if(isLeader()){
 648             return getSelf().path().toString();
 649         }
 650         String leaderId = currentBehavior.getLeaderId();
 651         if (leaderId == null) {
 652             return null;
 653         }
 654         String peerAddress = context.getPeerAddress(leaderId);
 655         if(LOG.isDebugEnabled()) {
 656             LOG.debug("{}: getLeaderAddress leaderId = {} peerAddress = {}",
 657                     persistenceId(), leaderId, peerAddress);
 658         }
 659
 660         return peerAddress;
 661     }
 662
 663     private void handleCaptureSnapshotReply(byte[] snapshotBytes) {
 664         LOG.info("{}: CaptureSnapshotReply received by actor: snapshot size {}", persistenceId(), snapshotBytes.length);
 665
 666         // create a snapshot object from the state provided and save it
 667         // when snapshot is saved async, SaveSnapshotSuccess is raised.
 668
 669         Snapshot sn = Snapshot.create(snapshotBytes,
 670             context.getReplicatedLog().getFrom(captureSnapshot.getLastAppliedIndex() + 1),
 671             captureSnapshot.getLastIndex(), captureSnapshot.getLastTerm(),
 672             captureSnapshot.getLastAppliedIndex(), captureSnapshot.getLastAppliedTerm());
 673
 674         persistence().saveSnapshot(sn);
 675
 676         LOG.info("{}: Persisting of snapshot done:{}", persistenceId(), sn.getLogMessage());
 677
 678         //be greedy and remove entries from in-mem journal which are in the snapshot
 679         // and update snapshotIndex and snapshotTerm without waiting for the success,
 680
 681         context.getReplicatedLog().snapshotPreCommit(
 682             captureSnapshot.getLastAppliedIndex(),
 683             captureSnapshot.getLastAppliedTerm());
 684
 685         LOG.info("{}: Removed in-memory snapshotted entries, adjusted snaphsotIndex:{} " +
 686             "and term:{}", persistenceId(), captureSnapshot.getLastAppliedIndex(),
 687             captureSnapshot.getLastAppliedTerm());
 688
 689         if (isLeader() && captureSnapshot.isInstallSnapshotInitiated()) {
 690             // this would be call straight to the leader and won't initiate in serialization
 691             currentBehavior.handleMessage(getSelf(), new SendInstallSnapshot(
 692                     ByteString.copyFrom(snapshotBytes)));
 693         }
 694
 695         captureSnapshot = null;
 696         hasSnapshotCaptureInitiated = false;
 697     }
 698
 699     protected boolean hasFollowers(){
 700         return getRaftActorContext().getPeerAddresses().keySet().size() > 0;
 701     }
 702
 703     private class ReplicatedLogImpl extends AbstractReplicatedLogImpl {
 704
 705         private static final int DATA_SIZE_DIVIDER = 5;
 706         private long dataSizeSinceLastSnapshot = 0;
 707
 708         public ReplicatedLogImpl(Snapshot snapshot) {
 709             super(snapshot.getLastAppliedIndex(), snapshot.getLastAppliedTerm(),
 710                 snapshot.getUnAppliedEntries());
 711         }
 712
 713         public ReplicatedLogImpl() {
 714             super();
 715         }
 716
 717         @Override public void removeFromAndPersist(long logEntryIndex) {
 718             int adjustedIndex = adjustedIndex(logEntryIndex);
 719
 720             if (adjustedIndex < 0) {
 721                 return;
 722             }
 723
 724             // FIXME: Maybe this should be done after the command is saved
 725             journal.subList(adjustedIndex , journal.size()).clear();
 726
 727             persistence().persist(new DeleteEntries(adjustedIndex), new Procedure<DeleteEntries>(){
 728
 729                 @Override public void apply(DeleteEntries param)
 730                     throws Exception {
 731                     //FIXME : Doing nothing for now
 732                     dataSize = 0;
 733                     for(ReplicatedLogEntry entry : journal){
 734                         dataSize += entry.size();
 735                     }
 736                 }
 737             });
 738         }
 739
 740         @Override public void appendAndPersist(
 741             final ReplicatedLogEntry replicatedLogEntry) {
 742             appendAndPersist(replicatedLogEntry, null);
 743         }
 744
 745         @Override
 746         public int dataSize() {
 747             return dataSize;
 748         }
 749
 750         public void appendAndPersist(
 751             final ReplicatedLogEntry replicatedLogEntry,
 752             final Procedure<ReplicatedLogEntry> callback)  {
 753
 754             if(LOG.isDebugEnabled()) {
 755                 LOG.debug("{}: Append log entry and persist {} ", persistenceId(), replicatedLogEntry);
 756             }
 757
 758             // FIXME : By adding the replicated log entry to the in-memory journal we are not truly ensuring durability of the logs
 759             journal.add(replicatedLogEntry);
 760
 761             // When persisting events with persist it is guaranteed that the
 762             // persistent actor will not receive further commands between the
 763             // persist call and the execution(s) of the associated event
 764             // handler. This also holds for multiple persist calls in context
 765             // of a single command.
 766             persistence().persist(replicatedLogEntry,
 767                 new Procedure<ReplicatedLogEntry>() {
 768                     @Override
 769                     public void apply(ReplicatedLogEntry evt) throws Exception {
 770                         int logEntrySize = replicatedLogEntry.size();
 771
 772                         dataSize += logEntrySize;
 773                         long dataSizeForCheck = dataSize;
 774
 775                         dataSizeSinceLastSnapshot += logEntrySize;
 776                         long journalSize = lastIndex()+1;
 777
 778                         if(!hasFollowers()) {
 779                             // When we do not have followers we do not maintain an in-memory log
 780                             // due to this the journalSize will never become anything close to the
 781                             // snapshot batch count. In fact will mostly be 1.
 782                             // Similarly since the journal's dataSize depends on the entries in the
 783                             // journal the journal's dataSize will never reach a value close to the
 784                             // memory threshold.
 785                             // By maintaining the dataSize outside the journal we are tracking essentially
 786                             // what we have written to the disk however since we no longer are in
 787                             // need of doing a snapshot just for the sake of freeing up memory we adjust
 788                             // the real size of data by the DATA_SIZE_DIVIDER so that we do not snapshot as often
 789                             // as if we were maintaining a real snapshot
 790                             dataSizeForCheck = dataSizeSinceLastSnapshot / DATA_SIZE_DIVIDER;
 791                         }
 792
 793                         long dataThreshold = Runtime.getRuntime().totalMemory() *
 794                                 getRaftActorContext().getConfigParams().getSnapshotDataThresholdPercentage() / 100;
 795
 796                         // when a snaphsot is being taken, captureSnapshot != null
 797                         if (hasSnapshotCaptureInitiated == false &&
 798                                 ( journalSize % context.getConfigParams().getSnapshotBatchCount() == 0 ||
 799                                         dataSizeForCheck > dataThreshold)) {
 800
 801                             dataSizeSinceLastSnapshot = 0;
 802
 803                             LOG.info("{}: Initiating Snapshot Capture..", persistenceId());
 804                             long lastAppliedIndex = -1;
 805                             long lastAppliedTerm = -1;
 806
 807                             ReplicatedLogEntry lastAppliedEntry = get(context.getLastApplied());
 808                             if (!hasFollowers()) {
 809                                 lastAppliedIndex = replicatedLogEntry.getIndex();
 810                                 lastAppliedTerm = replicatedLogEntry.getTerm();
 811                             } else if (lastAppliedEntry != null) {
 812                                 lastAppliedIndex = lastAppliedEntry.getIndex();
 813                                 lastAppliedTerm = lastAppliedEntry.getTerm();
 814                             }
 815
 816                             if(LOG.isDebugEnabled()) {
 817                                 LOG.debug("{}: Snapshot Capture logSize: {}", persistenceId(), journal.size());
 818                                 LOG.debug("{}: Snapshot Capture lastApplied:{} ",
 819                                         persistenceId(), context.getLastApplied());
 820                                 LOG.debug("{}: Snapshot Capture lastAppliedIndex:{}", persistenceId(),
 821                                         lastAppliedIndex);
 822                                 LOG.debug("{}: Snapshot Capture lastAppliedTerm:{}", persistenceId(),
 823                                         lastAppliedTerm);
 824                             }
 825
 826                             // send a CaptureSnapshot to self to make the expensive operation async.
 827                             getSelf().tell(new CaptureSnapshot(
 828                                 lastIndex(), lastTerm(), lastAppliedIndex, lastAppliedTerm),
 829                                 null);
 830                             hasSnapshotCaptureInitiated = true;
 831                         }
 832                         if(callback != null){
 833                             callback.apply(replicatedLogEntry);
 834                         }
 835                     }
 836                 }
 837             );
 838         }
 839
 840     }
 841
 842     static class DeleteEntries implements Serializable {
 843         private static final long serialVersionUID = 1L;
 844         private final int fromIndex;
 845
 846         public DeleteEntries(int fromIndex) {
 847             this.fromIndex = fromIndex;
 848         }
 849
 850         public int getFromIndex() {
 851             return fromIndex;
 852         }
 853     }
 854
 855
 856     private class ElectionTermImpl implements ElectionTerm {
 857         /**
 858          * Identifier of the actor whose election term information this is
 859          */
 860         private long currentTerm = 0;
 861         private String votedFor = null;
 862
 863         @Override
 864         public long getCurrentTerm() {
 865             return currentTerm;
 866         }
 867
 868         @Override
 869         public String getVotedFor() {
 870             return votedFor;
 871         }
 872
 873         @Override public void update(long currentTerm, String votedFor) {
 874             if(LOG.isDebugEnabled()) {
 875                 LOG.debug("{}: Set currentTerm={}, votedFor={}", persistenceId(), currentTerm, votedFor);
 876             }
 877             this.currentTerm = currentTerm;
 878             this.votedFor = votedFor;
 879         }
 880
 881         @Override
 882         public void updateAndPersist(long currentTerm, String votedFor){
 883             update(currentTerm, votedFor);
 884             // FIXME : Maybe first persist then update the state
 885             persistence().persist(new UpdateElectionTerm(this.currentTerm, this.votedFor), new Procedure<UpdateElectionTerm>(){
 886
 887                 @Override public void apply(UpdateElectionTerm param)
 888                     throws Exception {
 889
 890                 }
 891             });
 892         }
 893     }
 894
 895     static class UpdateElectionTerm implements Serializable {
 896         private static final long serialVersionUID = 1L;
 897         private final long currentTerm;
 898         private final String votedFor;
 899
 900         public UpdateElectionTerm(long currentTerm, String votedFor) {
 901             this.currentTerm = currentTerm;
 902             this.votedFor = votedFor;
 903         }
 904
 905         public long getCurrentTerm() {
 906             return currentTerm;
 907         }
 908
 909         public String getVotedFor() {
 910             return votedFor;
 911         }
 912     }
 913
 914     protected class NonPersistentRaftDataProvider extends NonPersistentDataProvider {
 915
 916         public NonPersistentRaftDataProvider(){
 917
 918         }
 919
 920         /**
 921          * The way snapshotting works is,
 922          * <ol>
 923          * <li> RaftActor calls createSnapshot on the Shard
 924          * <li> Shard sends a CaptureSnapshotReply and RaftActor then calls saveSnapshot
 925          * <li> When saveSnapshot is invoked on the akka-persistence API it uses the SnapshotStore to save the snapshot.
 926          * The SnapshotStore sends SaveSnapshotSuccess or SaveSnapshotFailure. When the RaftActor gets SaveSnapshot
 927          * success it commits the snapshot to the in-memory journal. This commitSnapshot is mimicking what is done
 928          * in SaveSnapshotSuccess.
 929          * </ol>
 930          * @param o
 931          */
 932         @Override
 933         public void saveSnapshot(Object o) {
 934             // Make saving Snapshot successful
 935             commitSnapshot(-1L);
 936         }
 937     }
 938
 939     @VisibleForTesting
 940     void setCurrentBehavior(AbstractRaftActorBehavior behavior) {
 941         currentBehavior = behavior;
 942     }
 943
 944     protected RaftActorBehavior getCurrentBehavior() {
 945         return currentBehavior;
 946     }
 947
 948 }