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

   1 /*
   2  * Copyright (c) 2015 Brocade Communications 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 package org.opendaylight.controller.cluster.raft;
   9
  10 import akka.persistence.RecoveryCompleted;
  11 import akka.persistence.SnapshotOffer;
  12 import com.google.common.base.Stopwatch;
  13 import java.util.Collections;
  14 import org.opendaylight.controller.cluster.PersistentDataProvider;
  15 import org.opendaylight.controller.cluster.raft.base.messages.ApplySnapshot;
  16 import org.opendaylight.controller.cluster.raft.persisted.ApplyJournalEntries;
  17 import org.opendaylight.controller.cluster.raft.persisted.DeleteEntries;
  18 import org.opendaylight.controller.cluster.raft.persisted.EmptyState;
  19 import org.opendaylight.controller.cluster.raft.persisted.MigratedSerializable;
  20 import org.opendaylight.controller.cluster.raft.persisted.ServerConfigurationPayload;
  21 import org.opendaylight.controller.cluster.raft.persisted.Snapshot;
  22 import org.opendaylight.controller.cluster.raft.persisted.UpdateElectionTerm;
  23 import org.opendaylight.controller.cluster.raft.protobuff.client.messages.PersistentPayload;
  24 import org.slf4j.Logger;
  25
  26 /**
  27  * Support class that handles persistence recovery for a RaftActor.
  28  *
  29  * @author Thomas Pantelis
  30  */
  31 class RaftActorRecoverySupport {
  32     private final RaftActorContext context;
  33     private final RaftActorRecoveryCohort cohort;
  34
  35     private int currentRecoveryBatchCount;
  36     private boolean dataRecoveredWithPersistenceDisabled;
  37     private boolean anyDataRecovered;
  38     private boolean hasMigratedDataRecovered;
  39
  40     private Stopwatch recoveryTimer;
  41     private final Logger log;
  42
  43     RaftActorRecoverySupport(final RaftActorContext context, final RaftActorRecoveryCohort cohort) {
  44         this.context = context;
  45         this.cohort = cohort;
  46         this.log = context.getLogger();
  47     }
  48
  49     boolean handleRecoveryMessage(final Object message, final PersistentDataProvider persistentProvider) {
  50         log.trace("{}: handleRecoveryMessage: {}", context.getId(), message);
  51
  52         anyDataRecovered = anyDataRecovered || !(message instanceof RecoveryCompleted);
  53
  54         if (isMigratedSerializable(message)) {
  55             hasMigratedDataRecovered = true;
  56         }
  57
  58         boolean recoveryComplete = false;
  59         if (message instanceof UpdateElectionTerm) {
  60             context.getTermInformation().update(((UpdateElectionTerm) message).getCurrentTerm(),
  61                     ((UpdateElectionTerm) message).getVotedFor());
  62         } else if (message instanceof SnapshotOffer) {
  63             onRecoveredSnapshot((SnapshotOffer) message);
  64         } else if (message instanceof ReplicatedLogEntry) {
  65             onRecoveredJournalLogEntry((ReplicatedLogEntry) message);
  66         } else if (message instanceof ApplyJournalEntries) {
  67             onRecoveredApplyLogEntries(((ApplyJournalEntries) message).getToIndex());
  68         } else if (message instanceof DeleteEntries) {
  69             onDeleteEntries((DeleteEntries) message);
  70         } else if (message instanceof ServerConfigurationPayload) {
  71             context.updatePeerIds((ServerConfigurationPayload)message);
  72         } else if (message instanceof RecoveryCompleted) {
  73             recoveryComplete = true;
  74             onRecoveryCompletedMessage(persistentProvider);
  75         }
  76
  77         return recoveryComplete;
  78     }
  79
  80     @SuppressWarnings("checkstyle:IllegalCatch")
  81     private void possiblyRestoreFromSnapshot() {
  82         Snapshot restoreFromSnapshot = cohort.getRestoreFromSnapshot();
  83         if (restoreFromSnapshot == null) {
  84             return;
  85         }
  86
  87         if (anyDataRecovered) {
  88             log.warn("{}: The provided restore snapshot was not applied because the persistence store is not empty",
  89                     context.getId());
  90             return;
  91         }
  92
  93         log.debug("{}: Restore snapshot: {}", context.getId(), restoreFromSnapshot);
  94
  95         context.getSnapshotManager().apply(new ApplySnapshot(restoreFromSnapshot));
  96     }
  97
  98     private ReplicatedLog replicatedLog() {
  99         return context.getReplicatedLog();
 100     }
 101
 102     private void initRecoveryTimer() {
 103         if (recoveryTimer == null) {
 104             recoveryTimer = Stopwatch.createStarted();
 105         }
 106     }
 107
 108     private void onRecoveredSnapshot(final SnapshotOffer offer) {
 109         log.debug("{}: SnapshotOffer called.", context.getId());
 110
 111         initRecoveryTimer();
 112
 113         Snapshot snapshot = (Snapshot) offer.snapshot();
 114
 115         for (ReplicatedLogEntry entry: snapshot.getUnAppliedEntries()) {
 116             if (isMigratedPayload(entry)) {
 117                 hasMigratedDataRecovered = true;
 118             }
 119         }
 120
 121         if (!context.getPersistenceProvider().isRecoveryApplicable()) {
 122             // We may have just transitioned to disabled and have a snapshot containing state data and/or log
 123             // entries - we don't want to preserve these, only the server config and election term info.
 124
 125             snapshot = Snapshot.create(
 126                     EmptyState.INSTANCE, Collections.emptyList(), -1, -1, -1, -1,
 127                     snapshot.getElectionTerm(), snapshot.getElectionVotedFor(), snapshot.getServerConfiguration());
 128         }
 129
 130         // Create a replicated log with the snapshot information
 131         // The replicated log can be used later on to retrieve this snapshot
 132         // when we need to install it on a peer
 133
 134         context.setReplicatedLog(ReplicatedLogImpl.newInstance(snapshot, context));
 135         context.setLastApplied(snapshot.getLastAppliedIndex());
 136         context.setCommitIndex(snapshot.getLastAppliedIndex());
 137         context.getTermInformation().update(snapshot.getElectionTerm(), snapshot.getElectionVotedFor());
 138
 139         Stopwatch timer = Stopwatch.createStarted();
 140
 141         // Apply the snapshot to the actors state
 142         if (!(snapshot.getState() instanceof EmptyState)) {
 143             cohort.applyRecoverySnapshot(snapshot.getState());
 144         }
 145
 146         if (snapshot.getServerConfiguration() != null) {
 147             context.updatePeerIds(snapshot.getServerConfiguration());
 148         }
 149
 150         timer.stop();
 151         log.info("Recovery snapshot applied for {} in {}: snapshotIndex={}, snapshotTerm={}, journal-size={}",
 152                 context.getId(), timer.toString(), replicatedLog().getSnapshotIndex(),
 153                 replicatedLog().getSnapshotTerm(), replicatedLog().size());
 154     }
 155
 156     private void onRecoveredJournalLogEntry(final ReplicatedLogEntry logEntry) {
 157         if (log.isDebugEnabled()) {
 158             log.debug("{}: Received ReplicatedLogEntry for recovery: index: {}, size: {}", context.getId(),
 159                     logEntry.getIndex(), logEntry.size());
 160         }
 161
 162         if (isServerConfigurationPayload(logEntry)) {
 163             context.updatePeerIds((ServerConfigurationPayload)logEntry.getData());
 164         }
 165
 166         if (isMigratedPayload(logEntry)) {
 167             hasMigratedDataRecovered = true;
 168         }
 169
 170         if (context.getPersistenceProvider().isRecoveryApplicable()) {
 171             replicatedLog().append(logEntry);
 172         } else if (!isPersistentPayload(logEntry)) {
 173             dataRecoveredWithPersistenceDisabled = true;
 174         }
 175     }
 176
 177     private void onRecoveredApplyLogEntries(final long toIndex) {
 178         if (!context.getPersistenceProvider().isRecoveryApplicable()) {
 179             dataRecoveredWithPersistenceDisabled = true;
 180             return;
 181         }
 182
 183         long lastUnappliedIndex = context.getLastApplied() + 1;
 184
 185         if (log.isDebugEnabled()) {
 186             // it can happen that lastUnappliedIndex > toIndex, if the AJE is in the persistent journal
 187             // but the entry itself has made it to that state and recovered via the snapshot
 188             log.debug("{}: Received apply journal entries for recovery, applying to state: {} to {}",
 189                     context.getId(), lastUnappliedIndex, toIndex);
 190         }
 191
 192         long lastApplied = lastUnappliedIndex - 1;
 193         for (long i = lastUnappliedIndex; i <= toIndex; i++) {
 194             ReplicatedLogEntry logEntry = replicatedLog().get(i);
 195             if (logEntry != null) {
 196                 lastApplied++;
 197                 batchRecoveredLogEntry(logEntry);
 198             } else {
 199                 // Shouldn't happen but cover it anyway.
 200                 log.error("{}: Log entry not found for index {}", context.getId(), i);
 201                 break;
 202             }
 203         }
 204
 205         context.setLastApplied(lastApplied);
 206         context.setCommitIndex(lastApplied);
 207     }
 208
 209     private void onDeleteEntries(final DeleteEntries deleteEntries) {
 210         if (context.getPersistenceProvider().isRecoveryApplicable()) {
 211             replicatedLog().removeFrom(deleteEntries.getFromIndex());
 212         } else {
 213             dataRecoveredWithPersistenceDisabled = true;
 214         }
 215     }
 216
 217     private void batchRecoveredLogEntry(final ReplicatedLogEntry logEntry) {
 218         initRecoveryTimer();
 219
 220         int batchSize = context.getConfigParams().getJournalRecoveryLogBatchSize();
 221         if (!isServerConfigurationPayload(logEntry)) {
 222             if (currentRecoveryBatchCount == 0) {
 223                 cohort.startLogRecoveryBatch(batchSize);
 224             }
 225
 226             cohort.appendRecoveredLogEntry(logEntry.getData());
 227
 228             if (++currentRecoveryBatchCount >= batchSize) {
 229                 endCurrentLogRecoveryBatch();
 230             }
 231         }
 232     }
 233
 234     private void endCurrentLogRecoveryBatch() {
 235         cohort.applyCurrentLogRecoveryBatch();
 236         currentRecoveryBatchCount = 0;
 237     }
 238
 239     private void onRecoveryCompletedMessage(final PersistentDataProvider persistentProvider) {
 240         if (currentRecoveryBatchCount > 0) {
 241             endCurrentLogRecoveryBatch();
 242         }
 243
 244         String recoveryTime = "";
 245         if (recoveryTimer != null) {
 246             recoveryTimer.stop();
 247             recoveryTime = " in " + recoveryTimer.toString();
 248             recoveryTimer = null;
 249         }
 250
 251         log.info("{}: Recovery completed {} - Switching actor to Follower - last log index = {}, last log term = {}, "
 252                 + "snapshot index = {}, snapshot term = {}, journal size = {}", context.getId(), recoveryTime,
 253                 replicatedLog().lastIndex(), replicatedLog().lastTerm(), replicatedLog().getSnapshotIndex(),
 254                 replicatedLog().getSnapshotTerm(), replicatedLog().size());
 255
 256         if (dataRecoveredWithPersistenceDisabled
 257                 || hasMigratedDataRecovered && !context.getPersistenceProvider().isRecoveryApplicable()) {
 258             if (hasMigratedDataRecovered) {
 259                 log.info("{}: Saving snapshot after recovery due to migrated messages", context.getId());
 260             } else {
 261                 log.info("{}: Saving snapshot after recovery due to data persistence disabled", context.getId());
 262             }
 263
 264             // Either data persistence is disabled and we recovered some data entries (ie we must have just
 265             // transitioned to disabled or a persistence backup was restored) or we recovered migrated
 266             // messages. Either way, we persist a snapshot and delete all the messages from the akka journal
 267             // to clean out unwanted messages.
 268
 269             Snapshot snapshot = Snapshot.create(
 270                     EmptyState.INSTANCE, Collections.<ReplicatedLogEntry>emptyList(),
 271                     -1, -1, -1, -1,
 272                     context.getTermInformation().getCurrentTerm(), context.getTermInformation().getVotedFor(),
 273                     context.getPeerServerInfo(true));
 274
 275             persistentProvider.saveSnapshot(snapshot);
 276
 277             persistentProvider.deleteMessages(persistentProvider.getLastSequenceNumber());
 278         } else if (hasMigratedDataRecovered) {
 279             log.info("{}: Snapshot capture initiated after recovery due to migrated messages", context.getId());
 280
 281             context.getSnapshotManager().capture(replicatedLog().last(), -1);
 282         } else {
 283             possiblyRestoreFromSnapshot();
 284         }
 285     }
 286
 287     private static boolean isServerConfigurationPayload(final ReplicatedLogEntry repLogEntry) {
 288         return repLogEntry.getData() instanceof ServerConfigurationPayload;
 289     }
 290
 291     private static boolean isPersistentPayload(final ReplicatedLogEntry repLogEntry) {
 292         return repLogEntry.getData() instanceof PersistentPayload;
 293     }
 294
 295     private static boolean isMigratedPayload(final ReplicatedLogEntry repLogEntry) {
 296         return isMigratedSerializable(repLogEntry.getData());
 297     }
 298
 299     private static boolean isMigratedSerializable(final Object message) {
 300         return message instanceof MigratedSerializable && ((MigratedSerializable)message).isMigrated();
 301     }
 302 }