Add Payload.serializedSize()
[controller.git] / opendaylight / md-sal / sal-akka-raft / src / main / java / org / opendaylight / controller / cluster / raft / behaviors / AbstractLeader.java
index ec465935039eb99a0c34341e05aa93bd2340e34d..0188a6df1ac387e603962d838def314a7dab1c8d 100644 (file)
@@ -13,7 +13,6 @@ import akka.actor.ActorRef;
 import akka.actor.ActorSelection;
 import akka.actor.Cancellable;
 import com.google.common.annotations.VisibleForTesting;
-import com.google.common.base.Optional;
 import com.google.common.io.ByteSource;
 import edu.umd.cs.findbugs.annotations.SuppressFBWarnings;
 import java.io.IOException;
@@ -26,6 +25,8 @@ import java.util.LinkedList;
 import java.util.List;
 import java.util.Map;
 import java.util.Map.Entry;
+import java.util.Optional;
+import java.util.OptionalInt;
 import java.util.Queue;
 import java.util.concurrent.TimeUnit;
 import org.eclipse.jdt.annotation.Nullable;
@@ -40,14 +41,17 @@ import org.opendaylight.controller.cluster.raft.RaftActorContext;
 import org.opendaylight.controller.cluster.raft.RaftState;
 import org.opendaylight.controller.cluster.raft.ReplicatedLogEntry;
 import org.opendaylight.controller.cluster.raft.VotingState;
+import org.opendaylight.controller.cluster.raft.base.messages.ApplyState;
 import org.opendaylight.controller.cluster.raft.base.messages.CheckConsensusReached;
 import org.opendaylight.controller.cluster.raft.base.messages.Replicate;
 import org.opendaylight.controller.cluster.raft.base.messages.SendHeartBeat;
 import org.opendaylight.controller.cluster.raft.base.messages.SendInstallSnapshot;
 import org.opendaylight.controller.cluster.raft.messages.AppendEntries;
 import org.opendaylight.controller.cluster.raft.messages.AppendEntriesReply;
+import org.opendaylight.controller.cluster.raft.messages.IdentifiablePayload;
 import org.opendaylight.controller.cluster.raft.messages.InstallSnapshot;
 import org.opendaylight.controller.cluster.raft.messages.InstallSnapshotReply;
+import org.opendaylight.controller.cluster.raft.messages.Payload;
 import org.opendaylight.controller.cluster.raft.messages.RaftRPC;
 import org.opendaylight.controller.cluster.raft.messages.RequestVote;
 import org.opendaylight.controller.cluster.raft.messages.RequestVoteReply;
@@ -98,7 +102,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
     private final MessageSlicer appendEntriesMessageSlicer;
 
     private Cancellable heartbeatSchedule = null;
-    private Optional<SnapshotHolder> snapshotHolder = Optional.absent();
+    private Optional<SnapshotHolder> snapshotHolder = Optional.empty();
     private int minReplicationCount;
 
     protected AbstractLeader(final RaftActorContext context, final RaftState state,
@@ -162,7 +166,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
         followerToLog.remove(followerId);
     }
 
-    public void updateMinReplicaCount() {
+    public final void updateMinReplicaCount() {
         int numVoting = 0;
         for (PeerInfo peer: context.getPeers()) {
             if (peer.isVoting()) {
@@ -186,7 +190,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
 
     @VisibleForTesting
     void setSnapshotHolder(final @Nullable SnapshotHolder snapshotHolder) {
-        this.snapshotHolder = Optional.fromNullable(snapshotHolder);
+        this.snapshotHolder = Optional.ofNullable(snapshotHolder);
     }
 
     @VisibleForTesting
@@ -432,8 +436,12 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
         super.performSnapshotWithoutCapture(minReplicatedToAllIndex);
     }
 
-    @Override
-    protected ClientRequestTracker removeClientRequestTracker(final long logIndex) {
+    /**
+     * Removes and returns the ClientRequestTracker for the specified log index.
+     * @param logIndex the log index
+     * @return the ClientRequestTracker or null if none available
+     */
+    private ClientRequestTracker removeClientRequestTracker(final long logIndex) {
         final Iterator<ClientRequestTracker> it = trackers.iterator();
         while (it.hasNext()) {
             final ClientRequestTracker t = it.next();
@@ -447,12 +455,35 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
     }
 
     @Override
-    protected RaftActorBehavior handleRequestVoteReply(final ActorRef sender,
-        final RequestVoteReply requestVoteReply) {
+    final ApplyState getApplyStateFor(final ReplicatedLogEntry entry) {
+        // first check whether a ClientRequestTracker exists for this entry.
+        // If it does that means the leader wasn't dropped before the transaction applied.
+        // That means that this transaction can be safely applied as a local transaction since we
+        // have the ClientRequestTracker.
+        final ClientRequestTracker tracker = removeClientRequestTracker(entry.getIndex());
+        if (tracker != null) {
+            return new ApplyState(tracker.getClientActor(), tracker.getIdentifier(), entry);
+        }
+
+        // Tracker is missing, this means that we switched behaviours between replicate and applystate
+        // and became the leader again,. We still want to apply this as a local modification because
+        // we have resumed leadership with that log entry having been committed.
+        final Payload payload = entry.getData();
+        if (payload instanceof IdentifiablePayload) {
+            return new ApplyState(null, ((IdentifiablePayload<?>) payload).getIdentifier(), entry);
+        }
+
+        return new ApplyState(null, null, entry);
+    }
+
+    @Override
+    protected RaftActorBehavior handleRequestVoteReply(final ActorRef sender, final RequestVoteReply requestVoteReply) {
         return this;
     }
 
-    protected void beforeSendHeartbeat(){}
+    protected void beforeSendHeartbeat() {
+        // No-op
+    }
 
     @Override
     public RaftActorBehavior handleMessage(final ActorRef sender, final Object message) {
@@ -462,12 +493,11 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
             return this;
         }
 
-        if (message instanceof RaftRPC) {
-            RaftRPC rpc = (RaftRPC) message;
+        if (message instanceof RaftRPC rpc) {
             // If RPC request or response contains term T > currentTerm:
             // set currentTerm = T, convert to follower (ยง5.1)
             // This applies to all RPC messages and responses
-            if (rpc.getTerm() > context.getTermInformation().getCurrentTerm()) {
+            if (rpc.getTerm() > context.getTermInformation().getCurrentTerm() && shouldUpdateTerm(rpc)) {
                 log.info("{}: Term {} in \"{}\" message is greater than leader's term {} - switching to Follower",
                         logName(), rpc.getTerm(), rpc, context.getTermInformation().getCurrentTerm());
 
@@ -481,9 +511,9 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
                 // start a new election due to lack of responses. This case would only occur if there isn't a majority
                 // of other nodes available that can elect the requesting candidate. Since we're transferring
                 // leadership, we should make every effort to get the requesting node elected.
-                if (message instanceof RequestVote && context.getRaftActorLeadershipTransferCohort() != null) {
+                if (rpc instanceof RequestVote && context.getRaftActorLeadershipTransferCohort() != null) {
                     log.debug("{}: Leadership transfer in progress - processing RequestVote", logName());
-                    super.handleMessage(sender, message);
+                    super.handleMessage(sender, rpc);
                 }
 
                 return internalSwitchBehavior(RaftState.Follower);
@@ -494,8 +524,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
             beforeSendHeartbeat();
             sendHeartBeat();
             scheduleHeartBeat(context.getConfigParams().getHeartBeatInterval());
-        } else if (message instanceof SendInstallSnapshot) {
-            SendInstallSnapshot sendInstallSnapshot = (SendInstallSnapshot) message;
+        } else if (message instanceof SendInstallSnapshot sendInstallSnapshot) {
             setSnapshotHolder(new SnapshotHolder(sendInstallSnapshot.getSnapshot(),
                 sendInstallSnapshot.getSnapshotBytes()));
             sendInstallSnapshot();
@@ -677,7 +706,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
                     } else if (installSnapshotState.canSendNextChunk()) {
                         sendSnapshotChunk(followerActor, followerLogInformation);
                     }
-                } else if (sendHeartbeat) {
+                } else if (sendHeartbeat || followerLogInformation.hasStaleCommitIndex(context.getCommitIndex())) {
                     // we send a heartbeat even if we have not received a reply for the last chunk
                     sendAppendEntries = true;
                 }
@@ -698,7 +727,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
                     log.debug("{}: sendAppendEntries: {} is present for follower {}", logName(),
                             followerNextIndex, followerId);
 
-                    if (followerLogInformation.okToReplicate()) {
+                    if (followerLogInformation.okToReplicate(context.getCommitIndex())) {
                         entries = getEntriesToSend(followerLogInformation, followerActor);
                         sendAppendEntries = true;
                     }
@@ -726,7 +755,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
                                 context.getReplicatedLog().size());
                     }
 
-                } else if (sendHeartbeat) {
+                } else if (sendHeartbeat || followerLogInformation.hasStaleCommitIndex(context.getCommitIndex())) {
                     // we send an AppendEntries, even if the follower is inactive
                     // in-order to update the followers timestamp, in case it becomes active again
                     sendAppendEntries = true;
@@ -752,7 +781,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
 
         // If the first entry's size exceeds the max data size threshold, it will be returned from the call above. If
         // that is the case, then we need to slice it into smaller chunks.
-        if (!(entries.size() == 1 && entries.get(0).getData().size() > maxDataSize)) {
+        if (entries.size() != 1 || entries.get(0).getData().serializedSize() <= maxDataSize) {
             // Don't need to slice.
             return entries;
         }
@@ -837,6 +866,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
                     appendEntries);
         }
 
+        followerLogInformation.setSentCommitIndex(leaderCommitIndex);
         followerActor.tell(appendEntries, actor());
     }
 
@@ -872,7 +902,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
         }
 
         boolean captureInitiated = context.getSnapshotManager().captureToInstall(context.getReplicatedLog().last(),
-            this.getReplicatedToAllIndex(), followerId);
+            getReplicatedToAllIndex(), followerId);
         if (captureInitiated) {
             followerLogInfo.setLeaderInstallSnapshotState(new LeaderInstallSnapshotState(
                 context.getConfigParams().getSnapshotChunkSize(), logName()));
@@ -936,19 +966,22 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
                         nextSnapshotChunk.length);
 
                 int nextChunkIndex = installSnapshotState.incrementChunkIndex();
-                Optional<ServerConfigurationPayload> serverConfig = Optional.absent();
+                Optional<ServerConfigurationPayload> serverConfig = Optional.empty();
                 if (installSnapshotState.isLastChunk(nextChunkIndex)) {
-                    serverConfig = Optional.fromNullable(context.getPeerServerInfo(true));
+                    serverConfig = Optional.ofNullable(context.getPeerServerInfo(true));
                 }
 
                 sendSnapshotChunk(followerActor, followerLogInfo, nextSnapshotChunk, nextChunkIndex, serverConfig);
 
+                log.debug("{}: InstallSnapshot sent to follower {}, Chunk: {}/{}", logName(), followerActor.path(),
+                        installSnapshotState.getChunkIndex(), installSnapshotState.getTotalChunks());
+
             } catch (IOException e) {
-                throw new RuntimeException(e);
+                log.warn("{}: Unable to send chunk: {}/{}. Reseting snapshot progress. Snapshot state: {}", logName(),
+                        installSnapshotState.getChunkIndex(), installSnapshotState.getTotalChunks(),
+                        installSnapshotState);
+                installSnapshotState.reset();
             }
-
-            log.debug("{}: InstallSnapshot sent to follower {}, Chunk: {}/{}", logName(), followerActor.path(),
-                installSnapshotState.getChunkIndex(), installSnapshotState.getTotalChunks());
         }
     }
 
@@ -965,7 +998,7 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
                         snapshotChunk,
                         chunkIndex,
                         installSnapshotState.getTotalChunks(),
-                        Optional.of(installSnapshotState.getLastChunkHashCode()),
+                        OptionalInt.of(installSnapshotState.getLastChunkHashCode()),
                         serverConfig
                 ).toSerializable(followerLogInfo.getRaftVersion()),
                 actor()
@@ -1089,8 +1122,8 @@ public abstract class AbstractLeader extends AbstractRaftActorBehavior {
         private final ByteSource snapshotBytes;
 
         SnapshotHolder(final Snapshot snapshot, final ByteSource snapshotBytes) {
-            this.lastIncludedTerm = snapshot.getLastAppliedTerm();
-            this.lastIncludedIndex = snapshot.getLastAppliedIndex();
+            lastIncludedTerm = snapshot.getLastAppliedTerm();
+            lastIncludedIndex = snapshot.getLastAppliedIndex();
             this.snapshotBytes = snapshotBytes;
         }