opendaylight/md-sal/sal-akka-raft/src/main/java/org/opendaylight/controller/cluster/raft/behaviors/AbstractLeader.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.behaviors;
  10
  11 import akka.actor.ActorRef;
  12 import akka.actor.ActorSelection;
  13 import akka.actor.Cancellable;
  14 import com.google.common.annotations.VisibleForTesting;
  15 import com.google.common.base.Optional;
  16 import com.google.common.base.Preconditions;
  17 import com.google.common.collect.ImmutableMap;
  18 import com.google.common.collect.ImmutableMap.Builder;
  19 import com.google.protobuf.ByteString;
  20 import java.io.IOException;
  21 import java.util.Collection;
  22 import java.util.Collections;
  23 import java.util.HashMap;
  24 import java.util.Iterator;
  25 import java.util.LinkedList;
  26 import java.util.List;
  27 import java.util.Map;
  28 import java.util.Map.Entry;
  29 import java.util.concurrent.TimeUnit;
  30 import org.opendaylight.controller.cluster.raft.ClientRequestTracker;
  31 import org.opendaylight.controller.cluster.raft.ClientRequestTrackerImpl;
  32 import org.opendaylight.controller.cluster.raft.FollowerLogInformation;
  33 import org.opendaylight.controller.cluster.raft.FollowerLogInformationImpl;
  34 import org.opendaylight.controller.cluster.raft.RaftActorContext;
  35 import org.opendaylight.controller.cluster.raft.RaftState;
  36 import org.opendaylight.controller.cluster.raft.ReplicatedLogEntry;
  37 import org.opendaylight.controller.cluster.raft.base.messages.CaptureSnapshot;
  38 import org.opendaylight.controller.cluster.raft.base.messages.InitiateInstallSnapshot;
  39 import org.opendaylight.controller.cluster.raft.base.messages.Replicate;
  40 import org.opendaylight.controller.cluster.raft.base.messages.SendHeartBeat;
  41 import org.opendaylight.controller.cluster.raft.base.messages.SendInstallSnapshot;
  42 import org.opendaylight.controller.cluster.raft.messages.AppendEntries;
  43 import org.opendaylight.controller.cluster.raft.messages.AppendEntriesReply;
  44 import org.opendaylight.controller.cluster.raft.messages.InstallSnapshot;
  45 import org.opendaylight.controller.cluster.raft.messages.InstallSnapshotReply;
  46 import org.opendaylight.controller.cluster.raft.messages.RaftRPC;
  47 import org.opendaylight.controller.cluster.raft.messages.RequestVoteReply;
  48 import scala.concurrent.duration.FiniteDuration;
  49
  50 /**
  51  * The behavior of a RaftActor when it is in the Leader state
  52  * <p/>
  53  * Leaders:
  54  * <ul>
  55  * <li> Upon election: send initial empty AppendEntries RPCs
  56  * (heartbeat) to each server; repeat during idle periods to
  57  * prevent election timeouts (§5.2)
  58  * <li> If command received from client: append entry to local log,
  59  * respond after entry applied to state machine (§5.3)
  60  * <li> If last log index ≥ nextIndex for a follower: send
  61  * AppendEntries RPC with log entries starting at nextIndex
  62  * <ul>
  63  * <li> If successful: update nextIndex and matchIndex for
  64  * follower (§5.3)
  65  * <li> If AppendEntries fails because of log inconsistency:
  66  * decrement nextIndex and retry (§5.3)
  67  * </ul>
  68  * <li> If there exists an N such that N > commitIndex, a majority
  69  * of matchIndex[i] ≥ N, and log[N].term == currentTerm:
  70  * set commitIndex = N (§5.3, §5.4).
  71  */
  72 public abstract class AbstractLeader extends AbstractRaftActorBehavior {
  73
  74     // The index of the first chunk that is sent when installing a snapshot
  75     public static final int FIRST_CHUNK_INDEX = 1;
  76
  77     // The index that the follower should respond with if it needs the install snapshot to be reset
  78     public static final int INVALID_CHUNK_INDEX = -1;
  79
  80     // This would be passed as the hash code of the last chunk when sending the first chunk
  81     public static final int INITIAL_LAST_CHUNK_HASH_CODE = -1;
  82
  83     private final Map<String, FollowerLogInformation> followerToLog;
  84     private final Map<String, FollowerToSnapshot> mapFollowerToSnapshot = new HashMap<>();
  85
  86     private Cancellable heartbeatSchedule = null;
  87
  88     private final Collection<ClientRequestTracker> trackerList = new LinkedList<>();
  89
  90     protected final int minReplicationCount;
  91
  92     protected final int minIsolatedLeaderPeerCount;
  93
  94     private Optional<ByteString> snapshot;
  95
  96     private long replicatedToAllIndex = -1;
  97
  98     public AbstractLeader(RaftActorContext context) {
  99         super(context);
 100
 101         final Builder<String, FollowerLogInformation> ftlBuilder = ImmutableMap.builder();
 102         for (String followerId : context.getPeerAddresses().keySet()) {
 103             FollowerLogInformation followerLogInformation =
 104                 new FollowerLogInformationImpl(followerId,
 105                     context.getCommitIndex(), -1,
 106                     context.getConfigParams().getElectionTimeOutInterval());
 107
 108             ftlBuilder.put(followerId, followerLogInformation);
 109         }
 110         followerToLog = ftlBuilder.build();
 111
 112         leaderId = context.getId();
 113
 114         LOG.debug("{}: Election: Leader has following peers: {}", context.getId(), getFollowerIds());
 115
 116         minReplicationCount = getMajorityVoteCount(getFollowerIds().size());
 117
 118         // the isolated Leader peer count will be 1 less than the majority vote count.
 119         // this is because the vote count has the self vote counted in it
 120         // for e.g
 121         // 0 peers = 1 votesRequired , minIsolatedLeaderPeerCount = 0
 122         // 2 peers = 2 votesRequired , minIsolatedLeaderPeerCount = 1
 123         // 4 peers = 3 votesRequired, minIsolatedLeaderPeerCount = 2
 124         minIsolatedLeaderPeerCount = minReplicationCount > 0 ? (minReplicationCount - 1) : 0;
 125
 126         snapshot = Optional.absent();
 127
 128         // Immediately schedule a heartbeat
 129         // Upon election: send initial empty AppendEntries RPCs
 130         // (heartbeat) to each server; repeat during idle periods to
 131         // prevent election timeouts (§5.2)
 132         scheduleHeartBeat(new FiniteDuration(0, TimeUnit.SECONDS));
 133     }
 134
 135     /**
 136      * Return an immutable collection of follower identifiers.
 137      *
 138      * @return Collection of follower IDs
 139      */
 140     protected final Collection<String> getFollowerIds() {
 141         return followerToLog.keySet();
 142     }
 143
 144     private Optional<ByteString> getSnapshot() {
 145         return snapshot;
 146     }
 147
 148     @VisibleForTesting
 149     void setSnapshot(Optional<ByteString> snapshot) {
 150         this.snapshot = snapshot;
 151     }
 152
 153     @Override
 154     protected RaftActorBehavior handleAppendEntries(ActorRef sender,
 155         AppendEntries appendEntries) {
 156
 157         if(LOG.isDebugEnabled()) {
 158             LOG.debug("{}: handleAppendEntries: {}", context.getId(), appendEntries);
 159         }
 160
 161         return this;
 162     }
 163
 164     @Override
 165     protected RaftActorBehavior handleAppendEntriesReply(ActorRef sender,
 166         AppendEntriesReply appendEntriesReply) {
 167
 168         if(! appendEntriesReply.isSuccess()) {
 169             if(LOG.isDebugEnabled()) {
 170                 LOG.debug("{}: handleAppendEntriesReply: {}", context.getId(), appendEntriesReply);
 171             }
 172         }
 173
 174         // Update the FollowerLogInformation
 175         String followerId = appendEntriesReply.getFollowerId();
 176         FollowerLogInformation followerLogInformation =
 177             followerToLog.get(followerId);
 178
 179         if(followerLogInformation == null){
 180             LOG.error("{}: handleAppendEntriesReply - unknown follower {}", context.getId(), followerId);
 181             return this;
 182         }
 183
 184         followerLogInformation.markFollowerActive();
 185
 186         if (appendEntriesReply.isSuccess()) {
 187             followerLogInformation
 188                 .setMatchIndex(appendEntriesReply.getLogLastIndex());
 189             followerLogInformation
 190                 .setNextIndex(appendEntriesReply.getLogLastIndex() + 1);
 191         } else {
 192
 193             // TODO: When we find that the follower is out of sync with the
 194             // Leader we simply decrement that followers next index by 1.
 195             // Would it be possible to do better than this? The RAFT spec
 196             // does not explicitly deal with it but may be something for us to
 197             // think about
 198
 199             followerLogInformation.decrNextIndex();
 200         }
 201
 202         // Now figure out if this reply warrants a change in the commitIndex
 203         // If there exists an N such that N > commitIndex, a majority
 204         // of matchIndex[i] ≥ N, and log[N].term == currentTerm:
 205         // set commitIndex = N (§5.3, §5.4).
 206         for (long N = context.getCommitIndex() + 1; ; N++) {
 207             int replicatedCount = 1;
 208
 209             for (FollowerLogInformation info : followerToLog.values()) {
 210                 if (info.getMatchIndex() >= N) {
 211                     replicatedCount++;
 212                 }
 213             }
 214
 215             if (replicatedCount >= minReplicationCount) {
 216                 ReplicatedLogEntry replicatedLogEntry = context.getReplicatedLog().get(N);
 217                 if (replicatedLogEntry != null &&
 218                     replicatedLogEntry.getTerm() == currentTerm()) {
 219                     context.setCommitIndex(N);
 220                 }
 221             } else {
 222                 break;
 223             }
 224         }
 225
 226         // Apply the change to the state machine
 227         if (context.getCommitIndex() > context.getLastApplied()) {
 228             applyLogToStateMachine(context.getCommitIndex());
 229         }
 230
 231         if (!context.isSnapshotCaptureInitiated()) {
 232             purgeInMemoryLog();
 233         }
 234
 235         return this;
 236     }
 237
 238     private void purgeInMemoryLog() {
 239         //find the lowest index across followers which has been replicated to all. -1 if there are no followers.
 240         // we would delete the in-mem log from that index on, in-order to minimize mem usage
 241         // we would also share this info thru AE with the followers so that they can delete their log entries as well.
 242         long minReplicatedToAllIndex = followerToLog.isEmpty() ? -1 : Long.MAX_VALUE;
 243         for (FollowerLogInformation info : followerToLog.values()) {
 244             minReplicatedToAllIndex = Math.min(minReplicatedToAllIndex, info.getMatchIndex());
 245         }
 246
 247         replicatedToAllIndex = fakeSnapshot(minReplicatedToAllIndex, replicatedToAllIndex);
 248     }
 249
 250     @Override
 251     protected ClientRequestTracker removeClientRequestTracker(long logIndex) {
 252         final Iterator<ClientRequestTracker> it = trackerList.iterator();
 253         while (it.hasNext()) {
 254             final ClientRequestTracker t = it.next();
 255             if (t.getIndex() == logIndex) {
 256                 it.remove();
 257                 return t;
 258             }
 259         }
 260
 261         return null;
 262     }
 263
 264     @Override
 265     protected ClientRequestTracker findClientRequestTracker(long logIndex) {
 266         for (ClientRequestTracker tracker : trackerList) {
 267             if (tracker.getIndex() == logIndex) {
 268                 return tracker;
 269             }
 270         }
 271         return null;
 272     }
 273
 274     @Override
 275     protected RaftActorBehavior handleRequestVoteReply(ActorRef sender,
 276         RequestVoteReply requestVoteReply) {
 277         return this;
 278     }
 279
 280     @Override
 281     public RaftState state() {
 282         return RaftState.Leader;
 283     }
 284
 285     @Override
 286     public RaftActorBehavior handleMessage(ActorRef sender, Object originalMessage) {
 287         Preconditions.checkNotNull(sender, "sender should not be null");
 288
 289         Object message = fromSerializableMessage(originalMessage);
 290
 291         if (message instanceof RaftRPC) {
 292             RaftRPC rpc = (RaftRPC) message;
 293             // If RPC request or response contains term T > currentTerm:
 294             // set currentTerm = T, convert to follower (§5.1)
 295             // This applies to all RPC messages and responses
 296             if (rpc.getTerm() > context.getTermInformation().getCurrentTerm()) {
 297                 context.getTermInformation().updateAndPersist(rpc.getTerm(), null);
 298
 299                 return switchBehavior(new Follower(context));
 300             }
 301         }
 302
 303         try {
 304             if (message instanceof SendHeartBeat) {
 305                 sendHeartBeat();
 306                 return this;
 307
 308             } else if(message instanceof InitiateInstallSnapshot) {
 309                 installSnapshotIfNeeded();
 310
 311             } else if(message instanceof SendInstallSnapshot) {
 312                 // received from RaftActor
 313                 setSnapshot(Optional.of(((SendInstallSnapshot) message).getSnapshot()));
 314                 sendInstallSnapshot();
 315
 316             } else if (message instanceof Replicate) {
 317                 replicate((Replicate) message);
 318
 319             } else if (message instanceof InstallSnapshotReply){
 320                 handleInstallSnapshotReply((InstallSnapshotReply) message);
 321
 322             }
 323         } finally {
 324             scheduleHeartBeat(context.getConfigParams().getHeartBeatInterval());
 325         }
 326
 327         return super.handleMessage(sender, message);
 328     }
 329
 330     private void handleInstallSnapshotReply(InstallSnapshotReply reply) {
 331         String followerId = reply.getFollowerId();
 332         FollowerToSnapshot followerToSnapshot = mapFollowerToSnapshot.get(followerId);
 333         FollowerLogInformation followerLogInformation = followerToLog.get(followerId);
 334         followerLogInformation.markFollowerActive();
 335
 336         if (followerToSnapshot != null &&
 337             followerToSnapshot.getChunkIndex() == reply.getChunkIndex()) {
 338
 339             if (reply.isSuccess()) {
 340                 if(followerToSnapshot.isLastChunk(reply.getChunkIndex())) {
 341                     //this was the last chunk reply
 342                     if(LOG.isDebugEnabled()) {
 343                         LOG.debug("{}: InstallSnapshotReply received, " +
 344                                 "last chunk received, Chunk:{}. Follower:{} Setting nextIndex:{}",
 345                                 context.getId(), reply.getChunkIndex(), followerId,
 346                             context.getReplicatedLog().getSnapshotIndex() + 1
 347                         );
 348                     }
 349
 350                     followerLogInformation.setMatchIndex(
 351                         context.getReplicatedLog().getSnapshotIndex());
 352                     followerLogInformation.setNextIndex(
 353                         context.getReplicatedLog().getSnapshotIndex() + 1);
 354                     mapFollowerToSnapshot.remove(followerId);
 355
 356                     if(LOG.isDebugEnabled()) {
 357                         LOG.debug("{}: followerToLog.get(followerId).getNextIndex()=" +
 358                                 context.getId(), followerToLog.get(followerId).getNextIndex());
 359                     }
 360
 361                     if (mapFollowerToSnapshot.isEmpty()) {
 362                         // once there are no pending followers receiving snapshots
 363                         // we can remove snapshot from the memory
 364                         setSnapshot(Optional.<ByteString>absent());
 365                     }
 366
 367                 } else {
 368                     followerToSnapshot.markSendStatus(true);
 369                 }
 370             } else {
 371                 LOG.info("{}: InstallSnapshotReply received sending snapshot chunk failed, Will retry, Chunk: {}",
 372                         context.getId(), reply.getChunkIndex());
 373
 374                 followerToSnapshot.markSendStatus(false);
 375             }
 376
 377         } else {
 378             LOG.error("{}: FollowerId in InstallSnapshotReply not known to Leader" +
 379                     " or Chunk Index in InstallSnapshotReply not matching {} != {}",
 380                     context.getId(), followerToSnapshot.getChunkIndex(), reply.getChunkIndex()
 381             );
 382
 383             if(reply.getChunkIndex() == INVALID_CHUNK_INDEX){
 384                 // Since the Follower did not find this index to be valid we should reset the follower snapshot
 385                 // so that Installing the snapshot can resume from the beginning
 386                 followerToSnapshot.reset();
 387             }
 388         }
 389     }
 390
 391     private void replicate(Replicate replicate) {
 392         long logIndex = replicate.getReplicatedLogEntry().getIndex();
 393
 394         if(LOG.isDebugEnabled()) {
 395             LOG.debug("{}: Replicate message {}", context.getId(), logIndex);
 396         }
 397
 398         // Create a tracker entry we will use this later to notify the
 399         // client actor
 400         trackerList.add(
 401             new ClientRequestTrackerImpl(replicate.getClientActor(),
 402                 replicate.getIdentifier(),
 403                 logIndex)
 404         );
 405
 406         if (followerToLog.isEmpty()) {
 407             context.setCommitIndex(logIndex);
 408             applyLogToStateMachine(logIndex);
 409         } else {
 410             sendAppendEntries();
 411         }
 412     }
 413
 414     private void sendAppendEntries() {
 415         // Send an AppendEntries to all followers
 416         for (Entry<String, FollowerLogInformation> e : followerToLog.entrySet()) {
 417             final String followerId = e.getKey();
 418             ActorSelection followerActor = context.getPeerActorSelection(followerId);
 419
 420             if (followerActor != null) {
 421                 FollowerLogInformation followerLogInformation = followerToLog.get(followerId);
 422                 long followerNextIndex = followerLogInformation.getNextIndex();
 423                 boolean isFollowerActive = followerLogInformation.isFollowerActive();
 424
 425                 if (mapFollowerToSnapshot.get(followerId) != null) {
 426                     // if install snapshot is in process , then sent next chunk if possible
 427                     if (isFollowerActive && mapFollowerToSnapshot.get(followerId).canSendNextChunk()) {
 428                         sendSnapshotChunk(followerActor, followerId);
 429                     } else {
 430                         // we send a heartbeat even if we have not received a reply for the last chunk
 431                         sendAppendEntriesToFollower(followerActor, followerNextIndex,
 432                             Collections.<ReplicatedLogEntry>emptyList());
 433                     }
 434
 435                 } else {
 436                     long leaderLastIndex = context.getReplicatedLog().lastIndex();
 437                     long leaderSnapShotIndex = context.getReplicatedLog().getSnapshotIndex();
 438                     final List<ReplicatedLogEntry> entries;
 439
 440                     if (isFollowerActive &&
 441                         context.getReplicatedLog().isPresent(followerNextIndex)) {
 442                         // FIXME : Sending one entry at a time
 443                         entries = context.getReplicatedLog().getFrom(followerNextIndex, 1);
 444
 445                     } else if (isFollowerActive && followerNextIndex >= 0 &&
 446                         leaderLastIndex >= followerNextIndex ) {
 447                         // if the followers next index is not present in the leaders log, and
 448                         // if the follower is just not starting and if leader's index is more than followers index
 449                         // then snapshot should be sent
 450
 451                         if(LOG.isDebugEnabled()) {
 452                             LOG.debug(String.format("%s: InitiateInstallSnapshot to follower: %s," +
 453                                     "follower-nextIndex: %s, leader-snapshot-index: %s,  " +
 454                                     "leader-last-index: %s", context.getId(), followerId,
 455                                 followerNextIndex, leaderSnapShotIndex, leaderLastIndex));
 456                         }
 457                         actor().tell(new InitiateInstallSnapshot(), actor());
 458
 459                         // we would want to sent AE as the capture snapshot might take time
 460                         entries =  Collections.<ReplicatedLogEntry>emptyList();
 461
 462                     } else {
 463                         //we send an AppendEntries, even if the follower is inactive
 464                         // in-order to update the followers timestamp, in case it becomes active again
 465                         entries =  Collections.<ReplicatedLogEntry>emptyList();
 466                     }
 467
 468                     sendAppendEntriesToFollower(followerActor, followerNextIndex, entries);
 469
 470                 }
 471             }
 472         }
 473     }
 474
 475     private void sendAppendEntriesToFollower(ActorSelection followerActor, long followerNextIndex,
 476         List<ReplicatedLogEntry> entries) {
 477         followerActor.tell(
 478             new AppendEntries(currentTerm(), context.getId(),
 479                 prevLogIndex(followerNextIndex),
 480                 prevLogTerm(followerNextIndex), entries,
 481                 context.getCommitIndex(),
 482                 replicatedToAllIndex).toSerializable(),
 483             actor()
 484         );
 485     }
 486
 487     /**
 488      * An installSnapshot is scheduled at a interval that is a multiple of
 489      * a HEARTBEAT_INTERVAL. This is to avoid the need to check for installing
 490      * snapshots at every heartbeat.
 491      *
 492      * Install Snapshot works as follows
 493      * 1. Leader sends a InitiateInstallSnapshot message to self
 494      * 2. Leader then initiates the capture snapshot by sending a CaptureSnapshot message to actor
 495      * 3. RaftActor on receipt of the CaptureSnapshotReply (from Shard), stores the received snapshot in the replicated log
 496      * and makes a call to Leader's handleMessage , with SendInstallSnapshot message.
 497      * 4. Leader , picks the snapshot from im-mem ReplicatedLog and sends it in chunks to the Follower
 498      * 5. On complete, Follower sends back a InstallSnapshotReply.
 499      * 6. On receipt of the InstallSnapshotReply for the last chunk, Leader marks the install complete for that follower
 500      * and replenishes the memory by deleting the snapshot in Replicated log.
 501      *
 502      */
 503     private void installSnapshotIfNeeded() {
 504         for (Entry<String, FollowerLogInformation> e : followerToLog.entrySet()) {
 505             final ActorSelection followerActor = context.getPeerActorSelection(e.getKey());
 506
 507             if (followerActor != null) {
 508                 long nextIndex = e.getValue().getNextIndex();
 509
 510                 if (!context.getReplicatedLog().isPresent(nextIndex) &&
 511                     context.getReplicatedLog().isInSnapshot(nextIndex)) {
 512                     LOG.info("{}: {} follower needs a snapshot install", context.getId(), e.getKey());
 513                     if (snapshot.isPresent()) {
 514                         // if a snapshot is present in the memory, most likely another install is in progress
 515                         // no need to capture snapshot
 516                         sendSnapshotChunk(followerActor, e.getKey());
 517
 518                     } else {
 519                         initiateCaptureSnapshot();
 520                         //we just need 1 follower who would need snapshot to be installed.
 521                         // when we have the snapshot captured, we would again check (in SendInstallSnapshot)
 522                         // who needs an install and send to all who need
 523                         break;
 524                     }
 525
 526                 }
 527             }
 528         }
 529     }
 530
 531     // on every install snapshot, we try to capture the snapshot.
 532     // Once a capture is going on, another one issued will get ignored by RaftActor.
 533     private void initiateCaptureSnapshot() {
 534         LOG.info("{}: Initiating Snapshot Capture to Install Snapshot, Leader:{}", context.getId(), getLeaderId());
 535         ReplicatedLogEntry lastAppliedEntry = context.getReplicatedLog().get(context.getLastApplied());
 536         long lastAppliedIndex = -1;
 537         long lastAppliedTerm = -1;
 538
 539         if (lastAppliedEntry != null) {
 540             lastAppliedIndex = lastAppliedEntry.getIndex();
 541             lastAppliedTerm = lastAppliedEntry.getTerm();
 542         } else if (context.getReplicatedLog().getSnapshotIndex() > -1)  {
 543             lastAppliedIndex = context.getReplicatedLog().getSnapshotIndex();
 544             lastAppliedTerm = context.getReplicatedLog().getSnapshotTerm();
 545         }
 546
 547         boolean isInstallSnapshotInitiated = true;
 548         actor().tell(new CaptureSnapshot(lastIndex(), lastTerm(),
 549                 lastAppliedIndex, lastAppliedTerm, isInstallSnapshotInitiated),
 550             actor());
 551     }
 552
 553
 554     private void sendInstallSnapshot() {
 555         for (Entry<String, FollowerLogInformation> e : followerToLog.entrySet()) {
 556             ActorSelection followerActor = context.getPeerActorSelection(e.getKey());
 557
 558             if (followerActor != null) {
 559                 long nextIndex = e.getValue().getNextIndex();
 560
 561                 if (!context.getReplicatedLog().isPresent(nextIndex) &&
 562                     context.getReplicatedLog().isInSnapshot(nextIndex)) {
 563                     sendSnapshotChunk(followerActor, e.getKey());
 564                 }
 565             }
 566         }
 567     }
 568
 569     /**
 570      *  Sends a snapshot chunk to a given follower
 571      *  InstallSnapshot should qualify as a heartbeat too.
 572      */
 573     private void sendSnapshotChunk(ActorSelection followerActor, String followerId) {
 574         try {
 575             if (snapshot.isPresent()) {
 576                 followerActor.tell(
 577                     new InstallSnapshot(currentTerm(), context.getId(),
 578                         context.getReplicatedLog().getSnapshotIndex(),
 579                         context.getReplicatedLog().getSnapshotTerm(),
 580                         getNextSnapshotChunk(followerId,snapshot.get()),
 581                         mapFollowerToSnapshot.get(followerId).incrementChunkIndex(),
 582                         mapFollowerToSnapshot.get(followerId).getTotalChunks(),
 583                         Optional.of(mapFollowerToSnapshot.get(followerId).getLastChunkHashCode())
 584                     ).toSerializable(),
 585                     actor()
 586                 );
 587                 LOG.info("{}: InstallSnapshot sent to follower {}, Chunk: {}/{}",
 588                         context.getId(), followerActor.path(),
 589                         mapFollowerToSnapshot.get(followerId).getChunkIndex(),
 590                         mapFollowerToSnapshot.get(followerId).getTotalChunks());
 591             }
 592         } catch (IOException e) {
 593             LOG.error(e, "{}: InstallSnapshot failed for Leader.", context.getId());
 594         }
 595     }
 596
 597     /**
 598      * Acccepts snaphot as ByteString, enters into map for future chunks
 599      * creates and return a ByteString chunk
 600      */
 601     private ByteString getNextSnapshotChunk(String followerId, ByteString snapshotBytes) throws IOException {
 602         FollowerToSnapshot followerToSnapshot = mapFollowerToSnapshot.get(followerId);
 603         if (followerToSnapshot == null) {
 604             followerToSnapshot = new FollowerToSnapshot(snapshotBytes);
 605             mapFollowerToSnapshot.put(followerId, followerToSnapshot);
 606         }
 607         ByteString nextChunk = followerToSnapshot.getNextChunk();
 608         if (LOG.isDebugEnabled()) {
 609             LOG.debug("{}: Leader's snapshot nextChunk size:{}", context.getId(), nextChunk.size());
 610         }
 611         return nextChunk;
 612     }
 613
 614     private void sendHeartBeat() {
 615         if (!followerToLog.isEmpty()) {
 616             sendAppendEntries();
 617         }
 618     }
 619
 620     private void stopHeartBeat() {
 621         if (heartbeatSchedule != null && !heartbeatSchedule.isCancelled()) {
 622             heartbeatSchedule.cancel();
 623         }
 624     }
 625
 626     private void scheduleHeartBeat(FiniteDuration interval) {
 627         if (followerToLog.isEmpty()) {
 628             // Optimization - do not bother scheduling a heartbeat as there are
 629             // no followers
 630             return;
 631         }
 632
 633         stopHeartBeat();
 634
 635         // Schedule a heartbeat. When the scheduler triggers a SendHeartbeat
 636         // message is sent to itself.
 637         // Scheduling the heartbeat only once here because heartbeats do not
 638         // need to be sent if there are other messages being sent to the remote
 639         // actor.
 640         heartbeatSchedule = context.getActorSystem().scheduler().scheduleOnce(
 641             interval, context.getActor(), new SendHeartBeat(),
 642             context.getActorSystem().dispatcher(), context.getActor());
 643     }
 644
 645     @Override
 646     public void close() throws Exception {
 647         stopHeartBeat();
 648     }
 649
 650     @Override
 651     public String getLeaderId() {
 652         return context.getId();
 653     }
 654
 655     protected boolean isLeaderIsolated() {
 656         int minPresent = minIsolatedLeaderPeerCount;
 657         for (FollowerLogInformation followerLogInformation : followerToLog.values()) {
 658             if (followerLogInformation.isFollowerActive()) {
 659                 --minPresent;
 660                 if (minPresent == 0) {
 661                     break;
 662                 }
 663             }
 664         }
 665         return (minPresent != 0);
 666     }
 667
 668     /**
 669      * Encapsulates the snapshot bytestring and handles the logic of sending
 670      * snapshot chunks
 671      */
 672     protected class FollowerToSnapshot {
 673         private final ByteString snapshotBytes;
 674         private int offset = 0;
 675         // the next snapshot chunk is sent only if the replyReceivedForOffset matches offset
 676         private int replyReceivedForOffset;
 677         // if replyStatus is false, the previous chunk is attempted
 678         private boolean replyStatus = false;
 679         private int chunkIndex;
 680         private final int totalChunks;
 681         private int lastChunkHashCode = AbstractLeader.INITIAL_LAST_CHUNK_HASH_CODE;
 682         private int nextChunkHashCode = AbstractLeader.INITIAL_LAST_CHUNK_HASH_CODE;
 683
 684         public FollowerToSnapshot(ByteString snapshotBytes) {
 685             this.snapshotBytes = snapshotBytes;
 686             int size = snapshotBytes.size();
 687             totalChunks = ( size / context.getConfigParams().getSnapshotChunkSize()) +
 688                 ((size % context.getConfigParams().getSnapshotChunkSize()) > 0 ? 1 : 0);
 689             if(LOG.isDebugEnabled()) {
 690                 LOG.debug("{}: Snapshot {} bytes, total chunks to send:{}",
 691                         context.getId(), size, totalChunks);
 692             }
 693             replyReceivedForOffset = -1;
 694             chunkIndex = AbstractLeader.FIRST_CHUNK_INDEX;
 695         }
 696
 697         public ByteString getSnapshotBytes() {
 698             return snapshotBytes;
 699         }
 700
 701         public int incrementOffset() {
 702             if(replyStatus) {
 703                 // if prev chunk failed, we would want to sent the same chunk again
 704                 offset = offset + context.getConfigParams().getSnapshotChunkSize();
 705             }
 706             return offset;
 707         }
 708
 709         public int incrementChunkIndex() {
 710             if (replyStatus) {
 711                 // if prev chunk failed, we would want to sent the same chunk again
 712                 chunkIndex =  chunkIndex + 1;
 713             }
 714             return chunkIndex;
 715         }
 716
 717         public int getChunkIndex() {
 718             return chunkIndex;
 719         }
 720
 721         public int getTotalChunks() {
 722             return totalChunks;
 723         }
 724
 725         public boolean canSendNextChunk() {
 726             // we only send a false if a chunk is sent but we have not received a reply yet
 727             return replyReceivedForOffset == offset;
 728         }
 729
 730         public boolean isLastChunk(int chunkIndex) {
 731             return totalChunks == chunkIndex;
 732         }
 733
 734         public void markSendStatus(boolean success) {
 735             if (success) {
 736                 // if the chunk sent was successful
 737                 replyReceivedForOffset = offset;
 738                 replyStatus = true;
 739                 lastChunkHashCode = nextChunkHashCode;
 740             } else {
 741                 // if the chunk sent was failure
 742                 replyReceivedForOffset = offset;
 743                 replyStatus = false;
 744             }
 745         }
 746
 747         public ByteString getNextChunk() {
 748             int snapshotLength = getSnapshotBytes().size();
 749             int start = incrementOffset();
 750             int size = context.getConfigParams().getSnapshotChunkSize();
 751             if (context.getConfigParams().getSnapshotChunkSize() > snapshotLength) {
 752                 size = snapshotLength;
 753             } else {
 754                 if ((start + context.getConfigParams().getSnapshotChunkSize()) > snapshotLength) {
 755                     size = snapshotLength - start;
 756                 }
 757             }
 758
 759             if(LOG.isDebugEnabled()) {
 760                 LOG.debug("{}: Next chunk: length={}, offset={},size={}", context.getId(),
 761                     snapshotLength, start, size);
 762             }
 763             ByteString substring = getSnapshotBytes().substring(start, start + size);
 764             nextChunkHashCode = substring.hashCode();
 765             return substring;
 766         }
 767
 768         /**
 769          * reset should be called when the Follower needs to be sent the snapshot from the beginning
 770          */
 771         public void reset(){
 772             offset = 0;
 773             replyStatus = false;
 774             replyReceivedForOffset = offset;
 775             chunkIndex = AbstractLeader.FIRST_CHUNK_INDEX;
 776             lastChunkHashCode = AbstractLeader.INITIAL_LAST_CHUNK_HASH_CODE;
 777         }
 778
 779         public int getLastChunkHashCode() {
 780             return lastChunkHashCode;
 781         }
 782     }
 783
 784     // called from example-actor for printing the follower-states
 785     public String printFollowerStates() {
 786         final StringBuilder sb = new StringBuilder();
 787
 788         sb.append('[');
 789         for (FollowerLogInformation followerLogInformation : followerToLog.values()) {
 790             sb.append('{');
 791             sb.append(followerLogInformation.getId());
 792             sb.append(" state:");
 793             sb.append(followerLogInformation.isFollowerActive());
 794             sb.append("},");
 795         }
 796         sb.append(']');
 797
 798         return sb.toString();
 799     }
 800
 801     @VisibleForTesting
 802     public FollowerLogInformation getFollower(String followerId) {
 803         return followerToLog.get(followerId);
 804     }
 805
 806     @VisibleForTesting
 807     protected void setFollowerSnapshot(String followerId, FollowerToSnapshot snapshot) {
 808         mapFollowerToSnapshot.put(followerId, snapshot);
 809     }
 810
 811     @VisibleForTesting
 812     public int followerSnapshotSize() {
 813         return mapFollowerToSnapshot.size();
 814     }
 815
 816     @VisibleForTesting
 817     public int followerLogSize() {
 818         return followerToLog.size();
 819     }
 820 }