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 org.opendaylight.controller.cluster.raft.ClientRequestTracker;
  30 import org.opendaylight.controller.cluster.raft.ClientRequestTrackerImpl;
  31 import org.opendaylight.controller.cluster.raft.FollowerLogInformation;
  32 import org.opendaylight.controller.cluster.raft.FollowerLogInformationImpl;
  33 import org.opendaylight.controller.cluster.raft.RaftActorContext;
  34 import org.opendaylight.controller.cluster.raft.RaftState;
  35 import org.opendaylight.controller.cluster.raft.ReplicatedLogEntry;
  36 import org.opendaylight.controller.cluster.raft.base.messages.Replicate;
  37 import org.opendaylight.controller.cluster.raft.base.messages.SendHeartBeat;
  38 import org.opendaylight.controller.cluster.raft.base.messages.SendInstallSnapshot;
  39 import org.opendaylight.controller.cluster.raft.messages.AppendEntries;
  40 import org.opendaylight.controller.cluster.raft.messages.AppendEntriesReply;
  41 import org.opendaylight.controller.cluster.raft.messages.InstallSnapshot;
  42 import org.opendaylight.controller.cluster.raft.messages.InstallSnapshotReply;
  43 import org.opendaylight.controller.cluster.raft.messages.RaftRPC;
  44 import org.opendaylight.controller.cluster.raft.messages.RequestVoteReply;
  45 import scala.concurrent.duration.FiniteDuration;
  46
  47 /**
  48  * The behavior of a RaftActor when it is in the Leader state
  49  * <p/>
  50  * Leaders:
  51  * <ul>
  52  * <li> Upon election: send initial empty AppendEntries RPCs
  53  * (heartbeat) to each server; repeat during idle periods to
  54  * prevent election timeouts (§5.2)
  55  * <li> If command received from client: append entry to local log,
  56  * respond after entry applied to state machine (§5.3)
  57  * <li> If last log index ≥ nextIndex for a follower: send
  58  * AppendEntries RPC with log entries starting at nextIndex
  59  * <ul>
  60  * <li> If successful: update nextIndex and matchIndex for
  61  * follower (§5.3)
  62  * <li> If AppendEntries fails because of log inconsistency:
  63  * decrement nextIndex and retry (§5.3)
  64  * </ul>
  65  * <li> If there exists an N such that N > commitIndex, a majority
  66  * of matchIndex[i] ≥ N, and log[N].term == currentTerm:
  67  * set commitIndex = N (§5.3, §5.4).
  68  */
  69 public abstract class AbstractLeader extends AbstractRaftActorBehavior {
  70
  71     // The index of the first chunk that is sent when installing a snapshot
  72     public static final int FIRST_CHUNK_INDEX = 1;
  73
  74     // The index that the follower should respond with if it needs the install snapshot to be reset
  75     public static final int INVALID_CHUNK_INDEX = -1;
  76
  77     // This would be passed as the hash code of the last chunk when sending the first chunk
  78     public static final int INITIAL_LAST_CHUNK_HASH_CODE = -1;
  79
  80     private final Map<String, FollowerLogInformation> followerToLog;
  81     private final Map<String, FollowerToSnapshot> mapFollowerToSnapshot = new HashMap<>();
  82
  83     private Cancellable heartbeatSchedule = null;
  84
  85     private final Collection<ClientRequestTracker> trackerList = new LinkedList<>();
  86
  87     protected final int minReplicationCount;
  88
  89     protected final int minIsolatedLeaderPeerCount;
  90
  91     private Optional<ByteString> snapshot;
  92
  93     public AbstractLeader(RaftActorContext context) {
  94         super(context, RaftState.Leader);
  95
  96         setLeaderPayloadVersion(context.getPayloadVersion());
  97
  98         final Builder<String, FollowerLogInformation> ftlBuilder = ImmutableMap.builder();
  99         for (String followerId : context.getPeerAddresses().keySet()) {
 100             FollowerLogInformation followerLogInformation =
 101                 new FollowerLogInformationImpl(followerId, -1, context);
 102
 103             ftlBuilder.put(followerId, followerLogInformation);
 104         }
 105         followerToLog = ftlBuilder.build();
 106
 107         leaderId = context.getId();
 108
 109         LOG.debug("{}: Election: Leader has following peers: {}", logName(), getFollowerIds());
 110
 111         minReplicationCount = getMajorityVoteCount(getFollowerIds().size());
 112
 113         // the isolated Leader peer count will be 1 less than the majority vote count.
 114         // this is because the vote count has the self vote counted in it
 115         // for e.g
 116         // 0 peers = 1 votesRequired , minIsolatedLeaderPeerCount = 0
 117         // 2 peers = 2 votesRequired , minIsolatedLeaderPeerCount = 1
 118         // 4 peers = 3 votesRequired, minIsolatedLeaderPeerCount = 2
 119         minIsolatedLeaderPeerCount = minReplicationCount > 0 ? (minReplicationCount - 1) : 0;
 120
 121         snapshot = Optional.absent();
 122
 123         // Immediately schedule a heartbeat
 124         // Upon election: send initial empty AppendEntries RPCs
 125         // (heartbeat) to each server; repeat during idle periods to
 126         // prevent election timeouts (§5.2)
 127         sendAppendEntries(0, false);
 128
 129         // It is important to schedule this heartbeat here
 130         scheduleHeartBeat(context.getConfigParams().getHeartBeatInterval());
 131     }
 132
 133     /**
 134      * Return an immutable collection of follower identifiers.
 135      *
 136      * @return Collection of follower IDs
 137      */
 138     public final Collection<String> getFollowerIds() {
 139         return followerToLog.keySet();
 140     }
 141
 142     @VisibleForTesting
 143     void setSnapshot(Optional<ByteString> snapshot) {
 144         this.snapshot = snapshot;
 145     }
 146
 147     @Override
 148     protected RaftActorBehavior handleAppendEntries(ActorRef sender,
 149         AppendEntries appendEntries) {
 150
 151         LOG.debug("{}: handleAppendEntries: {}", logName(), appendEntries);
 152
 153         return this;
 154     }
 155
 156     @Override
 157     protected RaftActorBehavior handleAppendEntriesReply(ActorRef sender,
 158         AppendEntriesReply appendEntriesReply) {
 159
 160         if(LOG.isTraceEnabled()) {
 161             LOG.trace("{}: handleAppendEntriesReply: {}", logName(), appendEntriesReply);
 162         }
 163
 164         // Update the FollowerLogInformation
 165         String followerId = appendEntriesReply.getFollowerId();
 166         FollowerLogInformation followerLogInformation =
 167             followerToLog.get(followerId);
 168
 169         if(followerLogInformation == null){
 170             LOG.error("{}: handleAppendEntriesReply - unknown follower {}", logName(), followerId);
 171             return this;
 172         }
 173
 174         if(followerLogInformation.timeSinceLastActivity() >
 175                 context.getConfigParams().getElectionTimeOutInterval().toMillis()) {
 176             LOG.warn("{} : handleAppendEntriesReply delayed beyond election timeout, " +
 177                             "appendEntriesReply : {}, timeSinceLastActivity : {}, lastApplied : {}, commitIndex : {}",
 178                     logName(), appendEntriesReply, followerLogInformation.timeSinceLastActivity(),
 179                     context.getLastApplied(), context.getCommitIndex());
 180         }
 181
 182         followerLogInformation.markFollowerActive();
 183         followerLogInformation.setPayloadVersion(appendEntriesReply.getPayloadVersion());
 184
 185         boolean updated = false;
 186         if (appendEntriesReply.isSuccess()) {
 187             updated = updateFollowerLogInformation(followerLogInformation, appendEntriesReply);
 188         } else {
 189             LOG.debug("{}: handleAppendEntriesReply: received unsuccessful reply: {}", logName(), appendEntriesReply);
 190
 191             long followerLastLogIndex = appendEntriesReply.getLogLastIndex();
 192             ReplicatedLogEntry followersLastLogEntry = context.getReplicatedLog().get(followerLastLogIndex);
 193             if(followerLastLogIndex < 0 || (followersLastLogEntry != null &&
 194                     followersLastLogEntry.getTerm() == appendEntriesReply.getLogLastTerm())) {
 195                 // The follower's log is empty or the last entry is present in the leader's journal
 196                 // and the terms match so the follower is just behind the leader's journal from
 197                 // the last snapshot, if any. We'll catch up the follower quickly by starting at the
 198                 // follower's last log index.
 199
 200                 updated = updateFollowerLogInformation(followerLogInformation, appendEntriesReply);
 201             } else {
 202                 // TODO: When we find that the follower is out of sync with the
 203                 // Leader we simply decrement that followers next index by 1.
 204                 // Would it be possible to do better than this? The RAFT spec
 205                 // does not explicitly deal with it but may be something for us to
 206                 // think about.
 207
 208                 followerLogInformation.decrNextIndex();
 209             }
 210         }
 211
 212         // Now figure out if this reply warrants a change in the commitIndex
 213         // If there exists an N such that N > commitIndex, a majority
 214         // of matchIndex[i] ≥ N, and log[N].term == currentTerm:
 215         // set commitIndex = N (§5.3, §5.4).
 216         for (long N = context.getCommitIndex() + 1; ; N++) {
 217             int replicatedCount = 1;
 218
 219             for (FollowerLogInformation info : followerToLog.values()) {
 220                 if (info.getMatchIndex() >= N) {
 221                     replicatedCount++;
 222                 }
 223             }
 224
 225             if (replicatedCount >= minReplicationCount) {
 226                 ReplicatedLogEntry replicatedLogEntry = context.getReplicatedLog().get(N);
 227                 if (replicatedLogEntry != null &&
 228                     replicatedLogEntry.getTerm() == currentTerm()) {
 229                     context.setCommitIndex(N);
 230                 }
 231             } else {
 232                 break;
 233             }
 234         }
 235
 236         // Apply the change to the state machine
 237         if (context.getCommitIndex() > context.getLastApplied()) {
 238             if(LOG.isDebugEnabled()) {
 239                 LOG.debug("{}: handleAppendEntriesReply from {}: applying to log - commitIndex: {}, lastAppliedIndex: {}",
 240                         logName(), followerId, context.getCommitIndex(), context.getLastApplied());
 241             }
 242
 243             applyLogToStateMachine(context.getCommitIndex());
 244         }
 245
 246         if (!context.getSnapshotManager().isCapturing()) {
 247             purgeInMemoryLog();
 248         }
 249
 250         //Send the next log entry immediately, if possible, no need to wait for heartbeat to trigger that event
 251         sendUpdatesToFollower(followerId, followerLogInformation, false, !updated);
 252         return this;
 253     }
 254
 255     private boolean updateFollowerLogInformation(FollowerLogInformation followerLogInformation,
 256             AppendEntriesReply appendEntriesReply) {
 257         boolean updated = followerLogInformation.setMatchIndex(appendEntriesReply.getLogLastIndex());
 258         updated = followerLogInformation.setNextIndex(appendEntriesReply.getLogLastIndex() + 1) || updated;
 259
 260         if(updated && LOG.isDebugEnabled()) {
 261             LOG.debug("{}: handleAppendEntriesReply - FollowerLogInformation for {} updated: matchIndex: {}, nextIndex: {}",
 262                     logName(), followerLogInformation.getId(), followerLogInformation.getMatchIndex(),
 263                     followerLogInformation.getNextIndex());
 264         }
 265         return updated;
 266     }
 267
 268     private void purgeInMemoryLog() {
 269         //find the lowest index across followers which has been replicated to all.
 270         // lastApplied if there are no followers, so that we keep clearing the log for single-node
 271         // we would delete the in-mem log from that index on, in-order to minimize mem usage
 272         // we would also share this info thru AE with the followers so that they can delete their log entries as well.
 273         long minReplicatedToAllIndex = followerToLog.isEmpty() ? context.getLastApplied() : Long.MAX_VALUE;
 274         for (FollowerLogInformation info : followerToLog.values()) {
 275             minReplicatedToAllIndex = Math.min(minReplicatedToAllIndex, info.getMatchIndex());
 276         }
 277
 278         super.performSnapshotWithoutCapture(minReplicatedToAllIndex);
 279     }
 280
 281     @Override
 282     protected ClientRequestTracker removeClientRequestTracker(long logIndex) {
 283         final Iterator<ClientRequestTracker> it = trackerList.iterator();
 284         while (it.hasNext()) {
 285             final ClientRequestTracker t = it.next();
 286             if (t.getIndex() == logIndex) {
 287                 it.remove();
 288                 return t;
 289             }
 290         }
 291
 292         return null;
 293     }
 294
 295     @Override
 296     protected ClientRequestTracker findClientRequestTracker(long logIndex) {
 297         for (ClientRequestTracker tracker : trackerList) {
 298             if (tracker.getIndex() == logIndex) {
 299                 return tracker;
 300             }
 301         }
 302         return null;
 303     }
 304
 305     @Override
 306     protected RaftActorBehavior handleRequestVoteReply(ActorRef sender,
 307         RequestVoteReply requestVoteReply) {
 308         return this;
 309     }
 310
 311     protected void beforeSendHeartbeat(){}
 312
 313     @Override
 314     public RaftActorBehavior handleMessage(ActorRef sender, Object originalMessage) {
 315         Preconditions.checkNotNull(sender, "sender should not be null");
 316
 317         Object message = fromSerializableMessage(originalMessage);
 318
 319         if (message instanceof RaftRPC) {
 320             RaftRPC rpc = (RaftRPC) message;
 321             // If RPC request or response contains term T > currentTerm:
 322             // set currentTerm = T, convert to follower (§5.1)
 323             // This applies to all RPC messages and responses
 324             if (rpc.getTerm() > context.getTermInformation().getCurrentTerm()) {
 325                 LOG.debug("{}: Term {} in \"{}\" message is greater than leader's term {} - switching to Follower",
 326                         logName(), rpc.getTerm(), rpc, context.getTermInformation().getCurrentTerm());
 327
 328                 context.getTermInformation().updateAndPersist(rpc.getTerm(), null);
 329
 330                 return switchBehavior(new Follower(context));
 331             }
 332         }
 333
 334         if (message instanceof SendHeartBeat) {
 335             beforeSendHeartbeat();
 336             sendHeartBeat();
 337             scheduleHeartBeat(context.getConfigParams().getHeartBeatInterval());
 338             return this;
 339
 340         } else if(message instanceof SendInstallSnapshot) {
 341             // received from RaftActor
 342             setSnapshot(Optional.of(((SendInstallSnapshot) message).getSnapshot()));
 343             sendInstallSnapshot();
 344
 345         } else if (message instanceof Replicate) {
 346             replicate((Replicate) message);
 347
 348         } else if (message instanceof InstallSnapshotReply){
 349             handleInstallSnapshotReply((InstallSnapshotReply) message);
 350
 351         }
 352
 353
 354         return super.handleMessage(sender, message);
 355     }
 356
 357     private void handleInstallSnapshotReply(InstallSnapshotReply reply) {
 358         LOG.debug("{}: handleInstallSnapshotReply: {}", logName(), reply);
 359
 360         String followerId = reply.getFollowerId();
 361         FollowerToSnapshot followerToSnapshot = mapFollowerToSnapshot.get(followerId);
 362
 363         if (followerToSnapshot == null) {
 364             LOG.error("{}: FollowerId {} in InstallSnapshotReply not known to Leader",
 365                     logName(), followerId);
 366             return;
 367         }
 368
 369         FollowerLogInformation followerLogInformation = followerToLog.get(followerId);
 370         followerLogInformation.markFollowerActive();
 371
 372         if (followerToSnapshot.getChunkIndex() == reply.getChunkIndex()) {
 373             boolean wasLastChunk = false;
 374             if (reply.isSuccess()) {
 375                 if(followerToSnapshot.isLastChunk(reply.getChunkIndex())) {
 376                     //this was the last chunk reply
 377                     if(LOG.isDebugEnabled()) {
 378                         LOG.debug("{}: InstallSnapshotReply received, " +
 379                                 "last chunk received, Chunk: {}. Follower: {} Setting nextIndex: {}",
 380                                 logName(), reply.getChunkIndex(), followerId,
 381                             context.getReplicatedLog().getSnapshotIndex() + 1
 382                         );
 383                     }
 384
 385                     followerLogInformation.setMatchIndex(
 386                         context.getReplicatedLog().getSnapshotIndex());
 387                     followerLogInformation.setNextIndex(
 388                         context.getReplicatedLog().getSnapshotIndex() + 1);
 389                     mapFollowerToSnapshot.remove(followerId);
 390
 391                     LOG.debug("{}: follower: {}, matchIndex set to {}, nextIndex set to {}",
 392                                 logName(), followerId, followerLogInformation.getMatchIndex(),
 393                                 followerLogInformation.getNextIndex());
 394
 395                     if (mapFollowerToSnapshot.isEmpty()) {
 396                         // once there are no pending followers receiving snapshots
 397                         // we can remove snapshot from the memory
 398                         setSnapshot(Optional.<ByteString>absent());
 399                     }
 400                     wasLastChunk = true;
 401
 402                 } else {
 403                     followerToSnapshot.markSendStatus(true);
 404                 }
 405             } else {
 406                 LOG.info("{}: InstallSnapshotReply received sending snapshot chunk failed, Will retry, Chunk: {}",
 407                         logName(), reply.getChunkIndex());
 408
 409                 followerToSnapshot.markSendStatus(false);
 410             }
 411
 412             if (wasLastChunk && !context.getSnapshotManager().isCapturing()) {
 413                 // Since the follower is now caught up try to purge the log.
 414                 purgeInMemoryLog();
 415             } else if (!wasLastChunk && followerToSnapshot.canSendNextChunk()) {
 416                 ActorSelection followerActor = context.getPeerActorSelection(followerId);
 417                 if(followerActor != null) {
 418                     sendSnapshotChunk(followerActor, followerId);
 419                 }
 420             }
 421
 422         } else {
 423             LOG.error("{}: Chunk index {} in InstallSnapshotReply from follower {} does not match expected index {}",
 424                     logName(), reply.getChunkIndex(), followerId,
 425                     followerToSnapshot.getChunkIndex());
 426
 427             if(reply.getChunkIndex() == INVALID_CHUNK_INDEX){
 428                 // Since the Follower did not find this index to be valid we should reset the follower snapshot
 429                 // so that Installing the snapshot can resume from the beginning
 430                 followerToSnapshot.reset();
 431             }
 432         }
 433     }
 434
 435     private void replicate(Replicate replicate) {
 436         long logIndex = replicate.getReplicatedLogEntry().getIndex();
 437
 438         LOG.debug("{}: Replicate message: identifier: {}, logIndex: {}", logName(),
 439                 replicate.getIdentifier(), logIndex);
 440
 441         // Create a tracker entry we will use this later to notify the
 442         // client actor
 443         trackerList.add(
 444             new ClientRequestTrackerImpl(replicate.getClientActor(),
 445                 replicate.getIdentifier(),
 446                 logIndex)
 447         );
 448
 449         if (followerToLog.isEmpty()) {
 450             context.setCommitIndex(logIndex);
 451             applyLogToStateMachine(logIndex);
 452         } else {
 453             sendAppendEntries(0, false);
 454         }
 455     }
 456
 457     private void sendAppendEntries(long timeSinceLastActivityInterval, boolean isHeartbeat) {
 458         // Send an AppendEntries to all followers
 459         for (Entry<String, FollowerLogInformation> e : followerToLog.entrySet()) {
 460             final String followerId = e.getKey();
 461             final FollowerLogInformation followerLogInformation = e.getValue();
 462             // This checks helps not to send a repeat message to the follower
 463             if(!followerLogInformation.isFollowerActive() ||
 464                     followerLogInformation.timeSinceLastActivity() >= timeSinceLastActivityInterval) {
 465                 sendUpdatesToFollower(followerId, followerLogInformation, true, isHeartbeat);
 466             }
 467         }
 468     }
 469
 470     /**
 471      *
 472      * This method checks if any update needs to be sent to the given follower. This includes append log entries,
 473      * sending next snapshot chunk, and initiating a snapshot.
 474      * @return true if any update is sent, false otherwise
 475      */
 476
 477     private void sendUpdatesToFollower(String followerId, FollowerLogInformation followerLogInformation,
 478                                        boolean sendHeartbeat, boolean isHeartbeat) {
 479
 480         ActorSelection followerActor = context.getPeerActorSelection(followerId);
 481         if (followerActor != null) {
 482             long followerNextIndex = followerLogInformation.getNextIndex();
 483             boolean isFollowerActive = followerLogInformation.isFollowerActive();
 484             boolean sendAppendEntries = false;
 485             List<ReplicatedLogEntry> entries = Collections.emptyList();
 486
 487             if (mapFollowerToSnapshot.get(followerId) != null) {
 488                 // if install snapshot is in process , then sent next chunk if possible
 489                 if (isFollowerActive && mapFollowerToSnapshot.get(followerId).canSendNextChunk()) {
 490                     sendSnapshotChunk(followerActor, followerId);
 491                 } else if(sendHeartbeat) {
 492                     // we send a heartbeat even if we have not received a reply for the last chunk
 493                     sendAppendEntries = true;
 494                 }
 495             } else {
 496                 long leaderLastIndex = context.getReplicatedLog().lastIndex();
 497                 long leaderSnapShotIndex = context.getReplicatedLog().getSnapshotIndex();
 498
 499                 if((!isHeartbeat && LOG.isDebugEnabled()) || LOG.isTraceEnabled()) {
 500                     LOG.debug("{}: Checking sendAppendEntries for follower {}: active: {}, followerNextIndex: {}, leaderLastIndex: {}, leaderSnapShotIndex: {}",
 501                             logName(), followerId, isFollowerActive, followerNextIndex, leaderLastIndex, leaderSnapShotIndex);
 502                 }
 503
 504                 if (isFollowerActive && context.getReplicatedLog().isPresent(followerNextIndex)) {
 505
 506                     LOG.debug("{}: sendAppendEntries: {} is present for follower {}", logName(),
 507                             followerNextIndex, followerId);
 508
 509                     if(followerLogInformation.okToReplicate()) {
 510                         // Try to send all the entries in the journal but not exceeding the max data size
 511                         // for a single AppendEntries message.
 512                         int maxEntries = (int) context.getReplicatedLog().size();
 513                         entries = context.getReplicatedLog().getFrom(followerNextIndex, maxEntries,
 514                                 context.getConfigParams().getSnapshotChunkSize());
 515                         sendAppendEntries = true;
 516                     }
 517                 } else if (isFollowerActive && followerNextIndex >= 0 &&
 518                     leaderLastIndex > followerNextIndex && !context.getSnapshotManager().isCapturing()) {
 519                     // if the followers next index is not present in the leaders log, and
 520                     // if the follower is just not starting and if leader's index is more than followers index
 521                     // then snapshot should be sent
 522
 523                     if (LOG.isDebugEnabled()) {
 524                         LOG.debug(String.format("%s: InitiateInstallSnapshot to follower: %s," +
 525                                     "follower-nextIndex: %d, leader-snapshot-index: %d,  " +
 526                                     "leader-last-index: %d", logName(), followerId,
 527                                     followerNextIndex, leaderSnapShotIndex, leaderLastIndex));
 528                     }
 529
 530                     // Send heartbeat to follower whenever install snapshot is initiated.
 531                     sendAppendEntries = true;
 532                     initiateCaptureSnapshot(followerId, followerNextIndex);
 533
 534                 } else if(sendHeartbeat) {
 535                     // we send an AppendEntries, even if the follower is inactive
 536                     // in-order to update the followers timestamp, in case it becomes active again
 537                     sendAppendEntries = true;
 538                 }
 539
 540             }
 541
 542             if(sendAppendEntries) {
 543                 sendAppendEntriesToFollower(followerActor, followerNextIndex,
 544                         entries, followerId);
 545             }
 546         }
 547     }
 548
 549     private void sendAppendEntriesToFollower(ActorSelection followerActor, long followerNextIndex,
 550         List<ReplicatedLogEntry> entries, String followerId) {
 551         AppendEntries appendEntries = new AppendEntries(currentTerm(), context.getId(),
 552             prevLogIndex(followerNextIndex),
 553             prevLogTerm(followerNextIndex), entries,
 554             context.getCommitIndex(), super.getReplicatedToAllIndex(), context.getPayloadVersion());
 555
 556         if(!entries.isEmpty() || LOG.isTraceEnabled()) {
 557             LOG.debug("{}: Sending AppendEntries to follower {}: {}", logName(), followerId,
 558                     appendEntries);
 559         }
 560
 561         followerActor.tell(appendEntries.toSerializable(), actor());
 562     }
 563
 564     /**
 565      * Install Snapshot works as follows
 566      * 1. Leader initiates the capture snapshot by sending a CaptureSnapshot message to actor
 567      * 2. RaftActor on receipt of the CaptureSnapshotReply (from Shard), stores the received snapshot in the replicated log
 568      * and makes a call to Leader's handleMessage , with SendInstallSnapshot message.
 569      * 3. Leader , picks the snapshot from im-mem ReplicatedLog and sends it in chunks to the Follower
 570      * 4. On complete, Follower sends back a InstallSnapshotReply.
 571      * 5. On receipt of the InstallSnapshotReply for the last chunk, Leader marks the install complete for that follower
 572      * and replenishes the memory by deleting the snapshot in Replicated log.
 573      * 6. If another follower requires a snapshot and a snapshot has been collected (via CaptureSnapshotReply)
 574      * then send the existing snapshot in chunks to the follower.
 575      * @param followerId
 576      * @param followerNextIndex
 577      */
 578     private void initiateCaptureSnapshot(String followerId, long followerNextIndex) {
 579         if (!context.getReplicatedLog().isPresent(followerNextIndex) &&
 580                 context.getReplicatedLog().isInSnapshot(followerNextIndex)) {
 581
 582             if (snapshot.isPresent()) {
 583                 // if a snapshot is present in the memory, most likely another install is in progress
 584                 // no need to capture snapshot.
 585                 // This could happen if another follower needs an install when one is going on.
 586                 final ActorSelection followerActor = context.getPeerActorSelection(followerId);
 587                 sendSnapshotChunk(followerActor, followerId);
 588
 589
 590             } else {
 591                 context.getSnapshotManager().captureToInstall(context.getReplicatedLog().last(),
 592                         this.getReplicatedToAllIndex(), followerId);
 593             }
 594         }
 595     }
 596
 597
 598     private void sendInstallSnapshot() {
 599         LOG.debug("{}: sendInstallSnapshot", logName());
 600         for (Entry<String, FollowerLogInformation> e : followerToLog.entrySet()) {
 601             ActorSelection followerActor = context.getPeerActorSelection(e.getKey());
 602
 603             if (followerActor != null) {
 604                 long nextIndex = e.getValue().getNextIndex();
 605
 606                 if (!context.getReplicatedLog().isPresent(nextIndex) &&
 607                     context.getReplicatedLog().isInSnapshot(nextIndex)) {
 608                     sendSnapshotChunk(followerActor, e.getKey());
 609                 }
 610             }
 611         }
 612     }
 613
 614     /**
 615      *  Sends a snapshot chunk to a given follower
 616      *  InstallSnapshot should qualify as a heartbeat too.
 617      */
 618     private void sendSnapshotChunk(ActorSelection followerActor, String followerId) {
 619         try {
 620             if (snapshot.isPresent()) {
 621                 ByteString nextSnapshotChunk = getNextSnapshotChunk(followerId,snapshot.get());
 622
 623                 // Note: the previous call to getNextSnapshotChunk has the side-effect of adding
 624                 // followerId to the followerToSnapshot map.
 625                 FollowerToSnapshot followerToSnapshot = mapFollowerToSnapshot.get(followerId);
 626
 627                 followerActor.tell(
 628                     new InstallSnapshot(currentTerm(), context.getId(),
 629                         context.getReplicatedLog().getSnapshotIndex(),
 630                         context.getReplicatedLog().getSnapshotTerm(),
 631                         nextSnapshotChunk,
 632                         followerToSnapshot.incrementChunkIndex(),
 633                         followerToSnapshot.getTotalChunks(),
 634                         Optional.of(followerToSnapshot.getLastChunkHashCode())
 635                     ).toSerializable(),
 636                     actor()
 637                 );
 638
 639                 if(LOG.isDebugEnabled()) {
 640                     LOG.debug("{}: InstallSnapshot sent to follower {}, Chunk: {}/{}",
 641                             logName(), followerActor.path(), followerToSnapshot.getChunkIndex(),
 642                             followerToSnapshot.getTotalChunks());
 643                 }
 644             }
 645         } catch (IOException e) {
 646             LOG.error("{}: InstallSnapshot failed for Leader.", logName(), e);
 647         }
 648     }
 649
 650     /**
 651      * Acccepts snaphot as ByteString, enters into map for future chunks
 652      * creates and return a ByteString chunk
 653      */
 654     private ByteString getNextSnapshotChunk(String followerId, ByteString snapshotBytes) throws IOException {
 655         FollowerToSnapshot followerToSnapshot = mapFollowerToSnapshot.get(followerId);
 656         if (followerToSnapshot == null) {
 657             followerToSnapshot = new FollowerToSnapshot(snapshotBytes);
 658             mapFollowerToSnapshot.put(followerId, followerToSnapshot);
 659         }
 660         ByteString nextChunk = followerToSnapshot.getNextChunk();
 661
 662         LOG.debug("{}: next snapshot chunk size for follower {}: {}", logName(), followerId, nextChunk.size());
 663
 664         return nextChunk;
 665     }
 666
 667     private void sendHeartBeat() {
 668         if (!followerToLog.isEmpty()) {
 669             LOG.trace("{}: Sending heartbeat", logName());
 670             sendAppendEntries(context.getConfigParams().getHeartBeatInterval().toMillis(), true);
 671         }
 672     }
 673
 674     private void stopHeartBeat() {
 675         if (heartbeatSchedule != null && !heartbeatSchedule.isCancelled()) {
 676             heartbeatSchedule.cancel();
 677         }
 678     }
 679
 680     private void scheduleHeartBeat(FiniteDuration interval) {
 681         if (followerToLog.isEmpty()) {
 682             // Optimization - do not bother scheduling a heartbeat as there are
 683             // no followers
 684             return;
 685         }
 686
 687         stopHeartBeat();
 688
 689         // Schedule a heartbeat. When the scheduler triggers a SendHeartbeat
 690         // message is sent to itself.
 691         // Scheduling the heartbeat only once here because heartbeats do not
 692         // need to be sent if there are other messages being sent to the remote
 693         // actor.
 694         heartbeatSchedule = context.getActorSystem().scheduler().scheduleOnce(
 695             interval, context.getActor(), new SendHeartBeat(),
 696             context.getActorSystem().dispatcher(), context.getActor());
 697     }
 698
 699     @Override
 700     public void close() throws Exception {
 701         stopHeartBeat();
 702     }
 703
 704     @Override
 705     public String getLeaderId() {
 706         return context.getId();
 707     }
 708
 709     protected boolean isLeaderIsolated() {
 710         int minPresent = minIsolatedLeaderPeerCount;
 711         for (FollowerLogInformation followerLogInformation : followerToLog.values()) {
 712             if (followerLogInformation.isFollowerActive()) {
 713                 --minPresent;
 714                 if (minPresent == 0) {
 715                     break;
 716                 }
 717             }
 718         }
 719         return (minPresent != 0);
 720     }
 721
 722     /**
 723      * Encapsulates the snapshot bytestring and handles the logic of sending
 724      * snapshot chunks
 725      */
 726     protected class FollowerToSnapshot {
 727         private final ByteString snapshotBytes;
 728         private int offset = 0;
 729         // the next snapshot chunk is sent only if the replyReceivedForOffset matches offset
 730         private int replyReceivedForOffset;
 731         // if replyStatus is false, the previous chunk is attempted
 732         private boolean replyStatus = false;
 733         private int chunkIndex;
 734         private final int totalChunks;
 735         private int lastChunkHashCode = AbstractLeader.INITIAL_LAST_CHUNK_HASH_CODE;
 736         private int nextChunkHashCode = AbstractLeader.INITIAL_LAST_CHUNK_HASH_CODE;
 737
 738         public FollowerToSnapshot(ByteString snapshotBytes) {
 739             this.snapshotBytes = snapshotBytes;
 740             int size = snapshotBytes.size();
 741             totalChunks = ( size / context.getConfigParams().getSnapshotChunkSize()) +
 742                 ((size % context.getConfigParams().getSnapshotChunkSize()) > 0 ? 1 : 0);
 743             if(LOG.isDebugEnabled()) {
 744                 LOG.debug("{}: Snapshot {} bytes, total chunks to send:{}",
 745                         logName(), size, totalChunks);
 746             }
 747             replyReceivedForOffset = -1;
 748             chunkIndex = AbstractLeader.FIRST_CHUNK_INDEX;
 749         }
 750
 751         public ByteString getSnapshotBytes() {
 752             return snapshotBytes;
 753         }
 754
 755         public int incrementOffset() {
 756             if(replyStatus) {
 757                 // if prev chunk failed, we would want to sent the same chunk again
 758                 offset = offset + context.getConfigParams().getSnapshotChunkSize();
 759             }
 760             return offset;
 761         }
 762
 763         public int incrementChunkIndex() {
 764             if (replyStatus) {
 765                 // if prev chunk failed, we would want to sent the same chunk again
 766                 chunkIndex =  chunkIndex + 1;
 767             }
 768             return chunkIndex;
 769         }
 770
 771         public int getChunkIndex() {
 772             return chunkIndex;
 773         }
 774
 775         public int getTotalChunks() {
 776             return totalChunks;
 777         }
 778
 779         public boolean canSendNextChunk() {
 780             // we only send a false if a chunk is sent but we have not received a reply yet
 781             return replyReceivedForOffset == offset;
 782         }
 783
 784         public boolean isLastChunk(int chunkIndex) {
 785             return totalChunks == chunkIndex;
 786         }
 787
 788         public void markSendStatus(boolean success) {
 789             if (success) {
 790                 // if the chunk sent was successful
 791                 replyReceivedForOffset = offset;
 792                 replyStatus = true;
 793                 lastChunkHashCode = nextChunkHashCode;
 794             } else {
 795                 // if the chunk sent was failure
 796                 replyReceivedForOffset = offset;
 797                 replyStatus = false;
 798             }
 799         }
 800
 801         public ByteString getNextChunk() {
 802             int snapshotLength = getSnapshotBytes().size();
 803             int start = incrementOffset();
 804             int size = context.getConfigParams().getSnapshotChunkSize();
 805             if (context.getConfigParams().getSnapshotChunkSize() > snapshotLength) {
 806                 size = snapshotLength;
 807             } else {
 808                 if ((start + context.getConfigParams().getSnapshotChunkSize()) > snapshotLength) {
 809                     size = snapshotLength - start;
 810                 }
 811             }
 812
 813
 814             LOG.debug("{}: Next chunk: length={}, offset={},size={}", logName(),
 815                     snapshotLength, start, size);
 816
 817             ByteString substring = getSnapshotBytes().substring(start, start + size);
 818             nextChunkHashCode = substring.hashCode();
 819             return substring;
 820         }
 821
 822         /**
 823          * reset should be called when the Follower needs to be sent the snapshot from the beginning
 824          */
 825         public void reset(){
 826             offset = 0;
 827             replyStatus = false;
 828             replyReceivedForOffset = offset;
 829             chunkIndex = AbstractLeader.FIRST_CHUNK_INDEX;
 830             lastChunkHashCode = AbstractLeader.INITIAL_LAST_CHUNK_HASH_CODE;
 831         }
 832
 833         public int getLastChunkHashCode() {
 834             return lastChunkHashCode;
 835         }
 836     }
 837
 838     // called from example-actor for printing the follower-states
 839     public String printFollowerStates() {
 840         final StringBuilder sb = new StringBuilder();
 841
 842         sb.append('[');
 843         for (FollowerLogInformation followerLogInformation : followerToLog.values()) {
 844             sb.append('{');
 845             sb.append(followerLogInformation.getId());
 846             sb.append(" state:");
 847             sb.append(followerLogInformation.isFollowerActive());
 848             sb.append("},");
 849         }
 850         sb.append(']');
 851
 852         return sb.toString();
 853     }
 854
 855     @VisibleForTesting
 856     public FollowerLogInformation getFollower(String followerId) {
 857         return followerToLog.get(followerId);
 858     }
 859
 860     @VisibleForTesting
 861     protected void setFollowerSnapshot(String followerId, FollowerToSnapshot snapshot) {
 862         mapFollowerToSnapshot.put(followerId, snapshot);
 863     }
 864
 865     @VisibleForTesting
 866     public int followerSnapshotSize() {
 867         return mapFollowerToSnapshot.size();
 868     }
 869
 870     @VisibleForTesting
 871     public int followerLogSize() {
 872         return followerToLog.size();
 873     }
 874 }