opendaylight/md-sal/sal-akka-raft/src/main/java/org/opendaylight/controller/cluster/raft/behaviors/Follower.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 com.google.common.annotations.VisibleForTesting;
  13 import java.util.ArrayList;
  14 import org.opendaylight.controller.cluster.raft.RaftActorContext;
  15 import org.opendaylight.controller.cluster.raft.RaftState;
  16 import org.opendaylight.controller.cluster.raft.ReplicatedLogEntry;
  17 import org.opendaylight.controller.cluster.raft.Snapshot;
  18 import org.opendaylight.controller.cluster.raft.base.messages.ApplySnapshot;
  19 import org.opendaylight.controller.cluster.raft.base.messages.ElectionTimeout;
  20 import org.opendaylight.controller.cluster.raft.messages.AppendEntries;
  21 import org.opendaylight.controller.cluster.raft.messages.AppendEntriesReply;
  22 import org.opendaylight.controller.cluster.raft.messages.InstallSnapshot;
  23 import org.opendaylight.controller.cluster.raft.messages.InstallSnapshotReply;
  24 import org.opendaylight.controller.cluster.raft.messages.RaftRPC;
  25 import org.opendaylight.controller.cluster.raft.messages.RequestVote;
  26 import org.opendaylight.controller.cluster.raft.messages.RequestVoteReply;
  27
  28 /**
  29  * The behavior of a RaftActor in the Follower state
  30  * <p/>
  31  * <ul>
  32  * <li> Respond to RPCs from candidates and leaders
  33  * <li> If election timeout elapses without receiving AppendEntries
  34  * RPC from current leader or granting vote to candidate:
  35  * convert to candidate
  36  * </ul>
  37  */
  38 public class Follower extends AbstractRaftActorBehavior {
  39
  40     private SnapshotTracker snapshotTracker = null;
  41
  42     private final SyncStatusTracker initialSyncStatusTracker;
  43
  44     private static final int SYNC_THRESHOLD = 10;
  45
  46     public Follower(RaftActorContext context) {
  47         super(context, RaftState.Follower);
  48
  49         initialSyncStatusTracker = new SyncStatusTracker(context.getActor(), getId(), SYNC_THRESHOLD);
  50
  51         if(context.getRaftPolicy().automaticElectionsEnabled()) {
  52             if (context.getPeerAddresses().isEmpty()) {
  53                 actor().tell(ELECTION_TIMEOUT, actor());
  54             } else {
  55                 scheduleElection(electionDuration());
  56             }
  57         }
  58
  59     }
  60
  61     private boolean isLogEntryPresent(long index){
  62         if(index == context.getReplicatedLog().getSnapshotIndex()){
  63             return true;
  64         }
  65
  66         ReplicatedLogEntry previousEntry = context.getReplicatedLog()
  67                 .get(index);
  68
  69         return previousEntry != null;
  70
  71     }
  72
  73     private long getLogEntryTerm(long index){
  74         if(index == context.getReplicatedLog().getSnapshotIndex()){
  75             return context.getReplicatedLog().getSnapshotTerm();
  76         }
  77
  78         ReplicatedLogEntry previousEntry = context.getReplicatedLog()
  79                 .get(index);
  80
  81         if(previousEntry != null){
  82             return previousEntry.getTerm();
  83         }
  84
  85         return -1;
  86     }
  87
  88     private void updateInitialSyncStatus(long currentLeaderCommit, String leaderId){
  89         initialSyncStatusTracker.update(leaderId, currentLeaderCommit, context.getCommitIndex());
  90     }
  91
  92     @Override protected RaftActorBehavior handleAppendEntries(ActorRef sender,
  93                                                               AppendEntries appendEntries) {
  94
  95         int numLogEntries = appendEntries.getEntries() != null ? appendEntries.getEntries().size() : 0;
  96         if(LOG.isTraceEnabled()) {
  97             LOG.trace("{}: handleAppendEntries: {}", logName(), appendEntries);
  98         } else if(LOG.isDebugEnabled() && numLogEntries > 0) {
  99             LOG.debug("{}: handleAppendEntries: {}", logName(), appendEntries);
 100         }
 101
 102         // TODO : Refactor this method into a bunch of smaller methods
 103         // to make it easier to read. Before refactoring ensure tests
 104         // cover the code properly
 105
 106         if (snapshotTracker != null) {
 107             // if snapshot install is in progress, follower should just acknowledge append entries with a reply.
 108             AppendEntriesReply reply = new AppendEntriesReply(context.getId(), currentTerm(), true,
 109                     lastIndex(), lastTerm(), context.getPayloadVersion());
 110
 111             if(LOG.isDebugEnabled()) {
 112                 LOG.debug("{}: snapshot install is in progress, replying immediately with {}", logName(), reply);
 113             }
 114             sender.tell(reply, actor());
 115
 116             return this;
 117         }
 118
 119         // If we got here then we do appear to be talking to the leader
 120         leaderId = appendEntries.getLeaderId();
 121
 122         setLeaderPayloadVersion(appendEntries.getPayloadVersion());
 123
 124         updateInitialSyncStatus(appendEntries.getLeaderCommit(), appendEntries.getLeaderId());
 125         // First check if the logs are in sync or not
 126         long lastIndex = lastIndex();
 127
 128         if (isOutOfSync(appendEntries)) {
 129             // We found that the log was out of sync so just send a negative
 130             // reply and return
 131
 132             LOG.debug("{}: Follower is out-of-sync, so sending negative reply, lastIndex: {}, lastTerm: {}",
 133                         logName(), lastIndex, lastTerm());
 134
 135             sender.tell(new AppendEntriesReply(context.getId(), currentTerm(), false, lastIndex,
 136                     lastTerm(), context.getPayloadVersion()), actor());
 137             return this;
 138         }
 139
 140         if (appendEntries.getEntries() != null && appendEntries.getEntries().size() > 0) {
 141
 142             LOG.debug("{}: Number of entries to be appended = {}", logName(),
 143                         appendEntries.getEntries().size());
 144
 145             // 3. If an existing entry conflicts with a new one (same index
 146             // but different terms), delete the existing entry and all that
 147             // follow it (§5.3)
 148             int addEntriesFrom = 0;
 149             if (context.getReplicatedLog().size() > 0) {
 150
 151                 // Find the entry up until the one that is not in the follower's log
 152                 for (int i = 0;i < appendEntries.getEntries().size(); i++, addEntriesFrom++) {
 153                     ReplicatedLogEntry matchEntry = appendEntries.getEntries().get(i);
 154                     ReplicatedLogEntry newEntry = context.getReplicatedLog().get(matchEntry.getIndex());
 155
 156                     if (newEntry == null) {
 157                         //newEntry not found in the log
 158                         break;
 159                     }
 160
 161                     if (newEntry.getTerm() == matchEntry.getTerm()) {
 162                         continue;
 163                     }
 164
 165                     if(!context.getRaftPolicy().applyModificationToStateBeforeConsensus()) {
 166
 167                         LOG.debug("{}: Removing entries from log starting at {}", logName(),
 168                                 matchEntry.getIndex());
 169
 170                         // Entries do not match so remove all subsequent entries
 171                         context.getReplicatedLog().removeFromAndPersist(matchEntry.getIndex());
 172                         break;
 173                     } else {
 174                         sender.tell(new AppendEntriesReply(context.getId(), currentTerm(), false, lastIndex,
 175                                 lastTerm(), context.getPayloadVersion(), true), actor());
 176                         return this;
 177                     }
 178                 }
 179             }
 180
 181             lastIndex = lastIndex();
 182             LOG.debug("{}: After cleanup entries to be added from = {}", logName(),
 183                         (addEntriesFrom + lastIndex));
 184
 185             // 4. Append any new entries not already in the log
 186             for (int i = addEntriesFrom; i < appendEntries.getEntries().size(); i++) {
 187                 ReplicatedLogEntry entry = appendEntries.getEntries().get(i);
 188
 189                 LOG.debug("{}: Append entry to log {}", logName(), entry.getData());
 190
 191                 context.getReplicatedLog().appendAndPersist(entry);
 192             }
 193
 194             LOG.debug("{}: Log size is now {}", logName(), context.getReplicatedLog().size());
 195         }
 196
 197         // 5. If leaderCommit > commitIndex, set commitIndex =
 198         // min(leaderCommit, index of last new entry)
 199
 200         lastIndex = lastIndex();
 201         long prevCommitIndex = context.getCommitIndex();
 202
 203         context.setCommitIndex(Math.min(appendEntries.getLeaderCommit(), lastIndex));
 204
 205         if (prevCommitIndex != context.getCommitIndex()) {
 206             LOG.debug("{}: Commit index set to {}", logName(), context.getCommitIndex());
 207         }
 208
 209         // If commitIndex > lastApplied: increment lastApplied, apply
 210         // log[lastApplied] to state machine (§5.3)
 211         // check if there are any entries to be applied. last-applied can be equal to last-index
 212         if (appendEntries.getLeaderCommit() > context.getLastApplied() &&
 213             context.getLastApplied() < lastIndex) {
 214             if(LOG.isDebugEnabled()) {
 215                 LOG.debug("{}: applyLogToStateMachine, " +
 216                         "appendEntries.getLeaderCommit(): {}," +
 217                         "context.getLastApplied(): {}, lastIndex(): {}", logName(),
 218                     appendEntries.getLeaderCommit(), context.getLastApplied(), lastIndex);
 219             }
 220
 221             applyLogToStateMachine(appendEntries.getLeaderCommit());
 222         }
 223
 224         AppendEntriesReply reply = new AppendEntriesReply(context.getId(), currentTerm(), true,
 225             lastIndex, lastTerm(), context.getPayloadVersion());
 226
 227         if(LOG.isTraceEnabled()) {
 228             LOG.trace("{}: handleAppendEntries returning : {}", logName(), reply);
 229         } else if(LOG.isDebugEnabled() && numLogEntries > 0) {
 230             LOG.debug("{}: handleAppendEntries returning : {}", logName(), reply);
 231         }
 232
 233         sender.tell(reply, actor());
 234
 235         if (!context.getSnapshotManager().isCapturing()) {
 236             super.performSnapshotWithoutCapture(appendEntries.getReplicatedToAllIndex());
 237         }
 238
 239         return this;
 240     }
 241
 242     private boolean isOutOfSync(AppendEntries appendEntries) {
 243
 244         long prevLogTerm = getLogEntryTerm(appendEntries.getPrevLogIndex());
 245         boolean prevEntryPresent = isLogEntryPresent(appendEntries.getPrevLogIndex());
 246         long lastIndex = lastIndex();
 247         int numLogEntries = appendEntries.getEntries() != null ? appendEntries.getEntries().size() : 0;
 248         boolean outOfSync = true;
 249
 250         if (lastIndex == -1 && appendEntries.getPrevLogIndex() != -1) {
 251
 252             // The follower's log is out of sync because the leader does have
 253             // an entry at prevLogIndex and this follower has no entries in
 254             // it's log.
 255
 256             LOG.debug("{}: The followers log is empty and the senders prevLogIndex is {}",
 257                         logName(), appendEntries.getPrevLogIndex());
 258         } else if (lastIndex > -1 && appendEntries.getPrevLogIndex() != -1 && !prevEntryPresent) {
 259
 260             // The follower's log is out of sync because the Leader's
 261             // prevLogIndex entry was not found in it's log
 262
 263             LOG.debug("{}: The log is not empty but the prevLogIndex {} was not found in it",
 264                         logName(), appendEntries.getPrevLogIndex());
 265         } else if (lastIndex > -1 && prevEntryPresent && prevLogTerm != appendEntries.getPrevLogTerm()) {
 266
 267             // The follower's log is out of sync because the Leader's
 268             // prevLogIndex entry does exist in the follower's log but it has
 269             // a different term in it
 270
 271             LOG.debug(
 272                     "{}: Cannot append entries because previous entry term {}  is not equal to append entries prevLogTerm {}",
 273                     logName(), prevLogTerm, appendEntries.getPrevLogTerm());
 274         } else if(appendEntries.getPrevLogIndex() == -1 && appendEntries.getPrevLogTerm() == -1
 275                 && appendEntries.getReplicatedToAllIndex() != -1
 276                 && !isLogEntryPresent(appendEntries.getReplicatedToAllIndex())) {
 277             // This append entry comes from a leader who has it's log aggressively trimmed and so does not have
 278             // the previous entry in it's in-memory journal
 279
 280             LOG.debug(
 281                     "{}: Cannot append entries because the replicatedToAllIndex {} does not appear to be in the in-memory journal",
 282                     logName(), appendEntries.getReplicatedToAllIndex());
 283         } else if(appendEntries.getPrevLogIndex() == -1 && appendEntries.getPrevLogTerm() == -1
 284                 && appendEntries.getReplicatedToAllIndex() != -1 && numLogEntries > 0 &&
 285                 !isLogEntryPresent(appendEntries.getEntries().get(0).getIndex() - 1)){
 286             LOG.debug(
 287                     "{}: Cannot append entries because the calculated previousIndex {} was not found in the in-memory journal",
 288                     logName(), appendEntries.getEntries().get(0).getIndex() - 1);
 289         } else {
 290             outOfSync = false;
 291         }
 292         return outOfSync;
 293     }
 294
 295     @Override protected RaftActorBehavior handleAppendEntriesReply(ActorRef sender,
 296         AppendEntriesReply appendEntriesReply) {
 297         return this;
 298     }
 299
 300     @Override protected RaftActorBehavior handleRequestVoteReply(ActorRef sender,
 301         RequestVoteReply requestVoteReply) {
 302         return this;
 303     }
 304
 305     @Override public RaftActorBehavior handleMessage(ActorRef sender, Object originalMessage) {
 306
 307         Object message = fromSerializableMessage(originalMessage);
 308
 309         if (message instanceof RaftRPC) {
 310             RaftRPC rpc = (RaftRPC) message;
 311             // If RPC request or response contains term T > currentTerm:
 312             // set currentTerm = T, convert to follower (§5.1)
 313             // This applies to all RPC messages and responses
 314             if (rpc.getTerm() > context.getTermInformation().getCurrentTerm()) {
 315                 LOG.debug("{}: Term {} in \"{}\" message is greater than follower's term {} - updating term",
 316                         logName(), rpc.getTerm(), rpc, context.getTermInformation().getCurrentTerm());
 317
 318                 context.getTermInformation().updateAndPersist(rpc.getTerm(), null);
 319             }
 320         }
 321
 322         if (message instanceof ElectionTimeout) {
 323             LOG.debug("{}: Received ElectionTimeout - switching to Candidate", logName());
 324             return internalSwitchBehavior(RaftState.Candidate);
 325
 326         } else if (message instanceof InstallSnapshot) {
 327             InstallSnapshot installSnapshot = (InstallSnapshot) message;
 328             handleInstallSnapshot(sender, installSnapshot);
 329         }
 330
 331         if(message instanceof RaftRPC && (!(message instanceof RequestVote) || (canGrantVote((RequestVote) message)))){
 332             scheduleElection(electionDuration());
 333         }
 334
 335         return super.handleMessage(sender, message);
 336     }
 337
 338     private void handleInstallSnapshot(ActorRef sender, InstallSnapshot installSnapshot) {
 339
 340         LOG.debug("{}: InstallSnapshot received from leader {}, datasize: {} , Chunk: {}/{}",
 341                     logName(), installSnapshot.getLeaderId(), installSnapshot.getData().size(),
 342                     installSnapshot.getChunkIndex(), installSnapshot.getTotalChunks());
 343
 344         if(snapshotTracker == null){
 345             snapshotTracker = new SnapshotTracker(LOG, installSnapshot.getTotalChunks());
 346         }
 347
 348         updateInitialSyncStatus(installSnapshot.getLastIncludedIndex(), installSnapshot.getLeaderId());
 349
 350         try {
 351             if(snapshotTracker.addChunk(installSnapshot.getChunkIndex(), installSnapshot.getData(),
 352                     installSnapshot.getLastChunkHashCode())){
 353                 Snapshot snapshot = Snapshot.create(snapshotTracker.getSnapshot(),
 354                         new ArrayList<ReplicatedLogEntry>(),
 355                         installSnapshot.getLastIncludedIndex(),
 356                         installSnapshot.getLastIncludedTerm(),
 357                         installSnapshot.getLastIncludedIndex(),
 358                         installSnapshot.getLastIncludedTerm());
 359
 360                 actor().tell(new ApplySnapshot(snapshot), actor());
 361
 362                 snapshotTracker = null;
 363
 364             }
 365
 366             InstallSnapshotReply reply = new InstallSnapshotReply(
 367                     currentTerm(), context.getId(), installSnapshot.getChunkIndex(), true);
 368
 369             LOG.debug("{}: handleInstallSnapshot returning: {}", logName(), reply);
 370
 371             sender.tell(reply, actor());
 372
 373         } catch (SnapshotTracker.InvalidChunkException e) {
 374             LOG.debug("{}: Exception in InstallSnapshot of follower", logName(), e);
 375
 376             sender.tell(new InstallSnapshotReply(currentTerm(), context.getId(),
 377                     -1, false), actor());
 378             snapshotTracker = null;
 379
 380         } catch (Exception e){
 381             LOG.error("{}: Exception in InstallSnapshot of follower", logName(), e);
 382
 383             //send reply with success as false. The chunk will be sent again on failure
 384             sender.tell(new InstallSnapshotReply(currentTerm(), context.getId(),
 385                     installSnapshot.getChunkIndex(), false), actor());
 386
 387         }
 388     }
 389
 390     @Override
 391     public void close() throws Exception {
 392         stopElection();
 393     }
 394
 395     @VisibleForTesting
 396     SnapshotTracker getSnapshotTracker(){
 397         return snapshotTracker;
 398     }
 399 }