Fixes for status endpoint

Nullable fields are not included in status response
Improved flaky tests around code

enterprise/causal-clustering/src/main/java/org/neo4j/causalclustering/catchup/tx/BatchingTxApplier.java

@@ -53,7 +53,6 @@ public class BatchingTxApplier extends LifecycleAdapter
     private final PullRequestMonitor monitor;
     private final PageCursorTracerSupplier pageCursorTracerSupplier;
     private final VersionContextSupplier versionContextSupplier;
-    private final ReadReplicaLastAppliedTransactionMonitor lastAppliedTransactionMonitor;
     private final CommandIndexTracker commandIndexTracker;
     private final Log log;
 
@@ -75,7 +74,6 @@ public BatchingTxApplier( int maxBatchSize, Supplier<TransactionIdStore> txIdSto
         this.pageCursorTracerSupplier = pageCursorTracerSupplier;
         this.log = logProvider.getLog( getClass() );
         this.monitor = monitors.newMonitor( PullRequestMonitor.class );
-        this.lastAppliedTransactionMonitor = monitors.newMonitor( ReadReplicaLastAppliedTransactionMonitor.class );
         this.versionContextSupplier = versionContextSupplier;
         this.commandIndexTracker = commandIndexTracker;
     }
@@ -89,7 +87,6 @@ public void start()
         {
             commitProcess.commit( first, NULL, EXTERNAL );
             pageCursorTracerSupplier.get().reportEvents();  // Report paging metrics for the commit
-            lastAppliedTransactionMonitor.applyTransaction( last.transactionId() );
             long lastAppliedRaftLogIndex = LogIndexTxHeaderEncoding.decodeLogIndexFromTxHeader( last.transactionRepresentation().additionalHeader() );
             commandIndexTracker.setAppliedCommandIndex( lastAppliedRaftLogIndex );
         } );

enterprise/causal-clustering/src/main/java/org/neo4j/causalclustering/core/consensus/ConsensusModule.java

@@ -149,7 +149,7 @@ public ConsensusModule( MemberId myself, final PlatformModule platformModule,
                 config.get( refuse_to_be_leader ),
                 supportsPreVoting, platformModule.monitors );
 
-        DurationSinceLastMessageMonitor durationSinceLastMessageMonitor = new DurationSinceLastMessageMonitor( logProvider );
+        DurationSinceLastMessageMonitor durationSinceLastMessageMonitor = new DurationSinceLastMessageMonitor();
         platformModule.monitors.addMonitorListener( durationSinceLastMessageMonitor );
         platformModule.dependencies.satisfyDependency( durationSinceLastMessageMonitor );
 

enterprise/causal-clustering/src/main/java/org/neo4j/causalclustering/core/consensus/DurationSinceLastMessageMonitor.java

@@ -30,24 +30,15 @@
 public class DurationSinceLastMessageMonitor implements RaftMessageTimerResetMonitor
 {
     private long lastMessageNanos = -1;
-    private final Log log;
-
-    public DurationSinceLastMessageMonitor( LogProvider logProvider )
-    {
-        log = logProvider.getLog( DurationSinceLastMessageMonitor.class );
-        log.info( "Created duration tracker" );
-    }
 
     @Override
     public void timerReset()
     {
         lastMessageNanos = System.nanoTime();
-        log.info( "Timer reset" );
     }
 
     public Duration durationSinceLastMessage()
     {
-        log.info( "Duration retrieved" );
         return Duration.ofNanos( System.nanoTime() - lastMessageNanos );
     }
 }

enterprise/causal-clustering/src/main/java/org/neo4j/causalclustering/readreplica/EnterpriseReadReplicaEditionModule.java

@@ -49,8 +49,6 @@
 import org.neo4j.causalclustering.catchup.storecopy.StoreFiles;
 import org.neo4j.causalclustering.catchup.tx.BatchingTxApplier;
 import org.neo4j.causalclustering.catchup.tx.CatchupPollingProcess;
-import org.neo4j.causalclustering.catchup.tx.ReadReplicaLastAppliedTransactionMonitor;
-import org.neo4j.causalclustering.catchup.tx.TrackingLastAppliedTransactionMonitor;
 import org.neo4j.causalclustering.catchup.tx.TransactionLogCatchUpFactory;
 import org.neo4j.causalclustering.catchup.tx.TxPullClient;
 import org.neo4j.causalclustering.core.CausalClusteringSettings;
@@ -276,11 +274,6 @@ public EnterpriseReadReplicaEditionModule( final PlatformModule platformModule,
         LifeSupport txPulling = new LifeSupport();
         int maxBatchSize = config.get( CausalClusteringSettings.read_replica_transaction_applier_batch_size );
 
-        // Used in the web interface to correctly report the transaction id
-        TrackingLastAppliedTransactionMonitor readReplicaLastAppliedTransactionMonitor = new TrackingLastAppliedTransactionMonitor();
-        dependencies.satisfyDependency( readReplicaLastAppliedTransactionMonitor );
-        platformModule.monitors.addMonitorListener( readReplicaLastAppliedTransactionMonitor );
-
         CommandIndexTracker commandIndexTracker = platformModule.dependencies.satisfyDependency( new CommandIndexTracker() );
         BatchingTxApplier batchingTxApplier = new BatchingTxApplier(
                 maxBatchSize, () -> localDatabase.dataSource().getDependencyResolver().resolveDependency( TransactionIdStore.class ), writableCommitProcess,

enterprise/server-enterprise/src/main/java/org/neo4j/server/rest/causalclustering/BaseStatus.java

@@ -23,8 +23,10 @@
 package org.neo4j.server.rest.causalclustering;
 
 import org.codehaus.jackson.map.ObjectMapper;
+import org.codehaus.jackson.map.annotate.JsonSerialize;
 
 import java.io.IOException;
+import java.time.Duration;
 import java.util.Collection;
 import javax.ws.rs.core.MediaType;
 import javax.ws.rs.core.Response;
@@ -54,7 +56,7 @@ public Response discover()
     }
 
     Response statusResponse( long lastAppliedRaftIndex, boolean isParticipatingInRaftGroup, Collection<MemberId> votingMembers, boolean isHealthy,
-            MemberId memberId, MemberId leader, Long millisSinceLastLeaderMessage, boolean isCore )
+            MemberId memberId, MemberId leader, Duration millisSinceLastLeaderMessage, boolean isCore )
     {
         String jsonObject;
         ObjectMapper objectMapper = new ObjectMapper();

enterprise/server-enterprise/src/main/java/org/neo4j/server/rest/causalclustering/ClusterStatusResponse.java

@@ -22,11 +22,17 @@
  */
 package org.neo4j.server.rest.causalclustering;
 
+import org.codehaus.jackson.map.annotate.JsonSerialize;
+
+import java.time.Duration;
 import java.util.Collection;
+import java.util.Optional;
+import java.util.UUID;
 import java.util.stream.Collectors;
 
 import org.neo4j.causalclustering.identity.MemberId;
 
+@JsonSerialize( include = JsonSerialize.Inclusion.NON_NULL )
 public class ClusterStatusResponse
 {
     private final boolean isCore;
@@ -39,22 +45,15 @@ public class ClusterStatusResponse
     private final Long millisSinceLastLeaderMessage;
 
     ClusterStatusResponse( long lastAppliedRaftIndex, boolean isParticipatingInRaftGroup, Collection<MemberId> votingMembers, boolean isHealthy,
-            MemberId memberId, MemberId leader, Long millisSinceLastLeaderMessage, boolean isCore )
+            MemberId memberId, MemberId leader, Duration millisSinceLastLeaderMessage, boolean isCore )
     {
         this.lastAppliedRaftIndex = lastAppliedRaftIndex;
         this.isParticipatingInRaftGroup = isParticipatingInRaftGroup;
         this.votingMembers = votingMembers.stream().map( member -> member.getUuid().toString() ).sorted().collect( Collectors.toList() );
         this.isHealthy = isHealthy;
         this.memberId = memberId.getUuid().toString();
-        if ( leader != null )
-        {
-            this.leader = leader.getUuid().toString();
-        }
-        else
-        {
-            this.leader = "";
-        }
-        this.millisSinceLastLeaderMessage = millisSinceLastLeaderMessage;
+        this.leader = Optional.ofNullable( leader ).map( MemberId::getUuid ).map( UUID::toString ).orElse( null );
+        this.millisSinceLastLeaderMessage = Optional.ofNullable( millisSinceLastLeaderMessage ).map( Duration::toMillis ).orElse( null );
         this.isCore = isCore;
     }
 

enterprise/server-enterprise/src/main/java/org/neo4j/server/rest/causalclustering/CoreStatus.java

@@ -22,6 +22,7 @@
  */
 package org.neo4j.server.rest.causalclustering;
 
+import java.time.Duration;
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.List;
@@ -105,7 +106,7 @@ public Response description()
         boolean participatingInRaftGroup = votingMembers.contains( myself ) && Objects.nonNull( leader );
 
         long lastAppliedRaftIndex = commandIndexTracker.getAppliedCommandIndex();
-        long millisSinceLastLeaderMessage = raftMessageTimerResetMonitor.durationSinceLastMessage().toMillis();
+        Duration millisSinceLastLeaderMessage = raftMessageTimerResetMonitor.durationSinceLastMessage();
 
         return statusResponse( lastAppliedRaftIndex, participatingInRaftGroup, votingMembers, databaseHealth.isHealthy(), myself, leader,
                 millisSinceLastLeaderMessage, true );

enterprise/server-enterprise/src/main/java/org/neo4j/server/rest/causalclustering/ReadReplicaStatus.java

@@ -22,6 +22,7 @@
  */
 package org.neo4j.server.rest.causalclustering;
 
+import java.time.Duration;
 import java.util.Collection;
 import javax.ws.rs.core.Response;
 
@@ -94,7 +95,7 @@ public Response description()
                 .orElse( null );
         long lastAppliedRaftIndex = commandIndexTracker.getAppliedCommandIndex();
         // leader message duration is meaningless for replicas since communication is not guaranteed with leader and transactions are streamed periodically
-        Long millisSinceLastLeaderMessage = null;
+        Duration millisSinceLastLeaderMessage = null;
         return statusResponse( lastAppliedRaftIndex, false, votingMembers, isHealthy, memberId, leader, millisSinceLastLeaderMessage, false );
     }
 }

enterprise/server-enterprise/src/test/java/org/neo4j/server/rest/causalclustering/CoreStatusTest.java

@@ -61,6 +61,7 @@
 import static javax.ws.rs.core.Response.Status.OK;
 import static org.hamcrest.Matchers.greaterThan;
 import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertThat;
 import static org.mockito.Mockito.mock;
 import static org.mockito.Mockito.when;
@@ -102,7 +103,7 @@ public void setup() throws Exception
         topologyService = dependencyResolver.satisfyDependency(
                 new FakeTopologyService( Arrays.asList( core2, core3 ), Collections.singleton( replica ), myself, RoleInfo.FOLLOWER ) );
 
-        raftMessageTimerResetMonitor = dependencyResolver.satisfyDependency( new DurationSinceLastMessageMonitor( logProvider ) );
+        raftMessageTimerResetMonitor = dependencyResolver.satisfyDependency( new DurationSinceLastMessageMonitor() );
         raftMachine = dependencyResolver.satisfyDependency( mock( RaftMachine.class ) );
         commandIndexTracker = dependencyResolver.satisfyDependency( new CommandIndexTracker() );
 
@@ -176,12 +177,13 @@ public void testAnswersWhenFollower()
     }
 
     @Test
-    public void expectedStatusFieldsAreIncluded() throws IOException, NoLeaderFoundException
+    public void expectedStatusFieldsAreIncluded() throws IOException, NoLeaderFoundException, InterruptedException
     {
         // given ideal normal conditions
         commandIndexTracker.setAppliedCommandIndex( 123 );
         when( raftMachine.getLeader() ).thenReturn( core2 );
         raftMessageTimerResetMonitor.timerReset();
+        Thread.sleep( 1 ); // Sometimes the test can be fast. This guarantees at least 1 ms since message received
 
         // and helpers
         List<String> votingMembers =
@@ -206,7 +208,7 @@ public void expectedStatusFieldsAreIncluded() throws IOException, NoLeaderFoundE
     public void notParticipatingInRaftGroupWhenNotInVoterSet() throws IOException
     {
         // given not in voting set
-        topologyService.replaceWithRole( core2, RoleInfo.LEADER ); // TODO necessary?
+        topologyService.replaceWithRole( core2, RoleInfo.LEADER );
         when( raftMembershipManager.votingMembers() ).thenReturn( new HashSet<>( Arrays.asList( core2, core3 ) ) );
 
         // when
@@ -236,7 +238,6 @@ public void databaseHealthIsReflected() throws IOException
     {
         // given database is not healthy
         databaseHealth.panic( new RuntimeException() );
-        topologyService.replaceWithRole( myself, RoleInfo.LEADER ); // TODO necessary?
 
         // when
         Response description = status.description();
@@ -247,7 +248,7 @@ public void databaseHealthIsReflected() throws IOException
     }
 
     @Test
-    public void leaderIsEmptyStringIfNonExistent() throws IOException
+    public void leaderNotIncludedIfUnknown() throws IOException
     {
         // given no leader
         topologyService.replaceWithRole( null, RoleInfo.LEADER );
@@ -257,7 +258,7 @@ public void leaderIsEmptyStringIfNonExistent() throws IOException
 
         // then
         Map<String,Object> response = responseAsMap( description );
-        assertThat( response, containsAndEquals( "leader", "" ) );
+        assertFalse( description.getEntity().toString(), response.containsKey( "leader" ) );
     }
 
     static RaftMembershipManager fakeRaftMembershipManager( Set<MemberId> votingMembers )

enterprise/server-enterprise/src/test/java/org/neo4j/server/rest/causalclustering/ReadReplicaStatusTest.java

@@ -54,6 +54,7 @@
 import static javax.ws.rs.core.Response.Status.NOT_FOUND;
 import static javax.ws.rs.core.Response.Status.OK;
 import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertTrue;
 import static org.mockito.Mockito.mock;
 import static org.mockito.Mockito.when;
@@ -157,10 +158,10 @@ public void includesMemberId() throws IOException
     }
 
     @Test
-    public void leaderIsReported() throws IOException
+    public void leaderIsOptional() throws IOException
     {
         Response description = status.description();
-        assertEquals( "", responseAsMap( description ).get( "leader" ) );
+        assertFalse( responseAsMap( description ).containsKey( "leader" ) );
 
         MemberId selectedLead = topologyService.allCoreServers()
                 .members()