Skip to content

[fix][broker] Log replicated subscription update ack failures #26109

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Denovo1998 wants to merge 3 commits into
base: master
Choose a base branch
from
Open

[fix][broker] Log replicated subscription update ack failures #26109

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
255353e
[fix][broker] Observe replicated subscription update ack failures
Denovo1998 Jun 29, 2026
45bde64
add test for replication subscription update failure handling
Denovo1998 Jun 30, 2026
591502d
Merge remote-tracking branch 'origin/master' into observe_replicated_…
lhotari Jul 26, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
32 changes: 28 additions & 4 deletions ...n/java/org/apache/pulsar/broker/service/persistent/ReplicatedSubscriptionsController.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,6 +41,7 @@
import org.apache.bookkeeper.mledger.PositionFactory;
import org.apache.commons.lang3.mutable.MutableBoolean;
import org.apache.pulsar.broker.service.Replicator;
import org.apache.pulsar.broker.service.Subscription;
import org.apache.pulsar.broker.service.Topic;
import org.apache.pulsar.broker.stats.OpenTelemetryReplicatedSubscriptionStats;
import org.apache.pulsar.common.api.proto.ClusterMessageId;
Expand All @@ -52,6 +53,7 @@
import org.apache.pulsar.common.api.proto.ReplicatedSubscriptionsSnapshotResponse;
import org.apache.pulsar.common.api.proto.ReplicatedSubscriptionsUpdate;
import org.apache.pulsar.common.protocol.Markers;
import org.apache.pulsar.common.util.FutureUtil;
import org.apache.pulsar.opentelemetry.annotations.PulsarDeprecatedMetric;

/**
Expand Down Expand Up @@ -214,7 +216,7 @@ private void receiveSubscriptionUpdated(ReplicatedSubscriptionsUpdate update) {

PersistentSubscription sub = topic.getSubscription(update.getSubscriptionName());
if (sub != null) {
sub.acknowledgeMessageAsync(Collections.singletonList(pos), AckType.Cumulative, Collections.emptyMap());
acknowledgeSubscriptionUpdate(update.getSubscriptionName(), sub, pos);
} else {
// Subscription doesn't exist. We need to force the creation of the subscription in this cluster.
log.info()
Expand All @@ -224,13 +226,35 @@ private void receiveSubscriptionUpdated(ReplicatedSubscriptionsUpdate update) {
.log("Creating subscription at: after receiving update from replicated subscription");
topic.createSubscription(update.getSubscriptionName(), InitialPosition.Earliest,
true /* replicateSubscriptionState */, Collections.emptyMap())
.thenAccept(subscriptionCreated -> {
subscriptionCreated.acknowledgeMessageAsync(Collections.singletonList(pos),
AckType.Cumulative, Collections.emptyMap());
.thenAccept(subscriptionCreated ->
Comment thread
lhotari marked this conversation as resolved.
acknowledgeSubscriptionUpdate(update.getSubscriptionName(), subscriptionCreated, pos))
.exceptionally(e -> {
if (e != null) {
log.warn()
.attr("subscriptionName", update.getSubscriptionName())
.attr("pos", pos)
.exception(FutureUtil.unwrapCompletionException(e))
.log("Failed to create replicated subscription");
}
return null;
});
}
}

private CompletableFuture<Void> acknowledgeSubscriptionUpdate(String subscriptionName, Subscription sub,
Position pos) {
return sub.acknowledgeMessageAsync(Collections.singletonList(pos), AckType.Cumulative, Collections.emptyMap())
.whenComplete((__, e) -> {
if (e != null) {
log.warn()
.attr("subscriptionName", subscriptionName)
.attr("pos", pos)
.exception(FutureUtil.unwrapCompletionException(e))
.log("Failed to update replicated subscription");
}
});
}

private void startNewSnapshot() {
cleanupTimedOutSnapshots();

Expand Down
137 changes: 137 additions & 0 deletions ...va/org/apache/pulsar/broker/service/persistent/ReplicatedSubscriptionsControllerTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,7 @@
*/
package org.apache.pulsar.broker.service.persistent;

import static org.assertj.core.api.Assertions.assertThat;
import static org.mockito.ArgumentMatchers.any;
import static org.mockito.ArgumentMatchers.anyLong;
import static org.mockito.ArgumentMatchers.anyString;
Expand All @@ -42,6 +43,9 @@
import java.util.concurrent.TimeUnit;
import java.util.concurrent.atomic.AtomicBoolean;
import java.util.concurrent.atomic.AtomicReference;
import java.util.function.BiConsumer;
import java.util.function.BiFunction;
import java.util.function.Function;
import lombok.Cleanup;
import org.apache.bookkeeper.mledger.AsyncCallbacks.DeleteCursorCallback;
import org.apache.bookkeeper.mledger.ManagedLedger;
Expand All @@ -54,7 +58,10 @@
import org.apache.pulsar.broker.service.Replicator;
import org.apache.pulsar.broker.service.Topic;
import org.apache.pulsar.broker.stats.OpenTelemetryReplicatedSubscriptionStats;
import org.apache.pulsar.common.api.proto.CommandAck.AckType;
import org.apache.pulsar.common.api.proto.CommandSubscribe.InitialPosition;
import org.apache.pulsar.common.api.proto.MarkerType;
import org.apache.pulsar.common.api.proto.MarkersMessageIdData;
import org.apache.pulsar.common.policies.data.BacklogQuota;
import org.apache.pulsar.common.policies.data.impl.BacklogQuotaImpl;
import org.apache.pulsar.common.protocol.Commands;
Expand All @@ -66,6 +73,100 @@
@Test(groups = "broker-replication")
public class ReplicatedSubscriptionsControllerTest {

@Test
@SuppressWarnings({"rawtypes", "unchecked"})
public void testReplicatedSubscriptionUpdateAckFutureIsObserved() throws Exception {
PulsarService pulsar = mock(PulsarService.class);
ScheduledExecutorService executor = mock(ScheduledExecutorService.class);
ScheduledFuture timer = mock(ScheduledFuture.class);
ServiceConfiguration config = new ServiceConfiguration();
config.setEnableReplicatedSubscriptions(true);
OpenTelemetryReplicatedSubscriptionStats stats = mock(OpenTelemetryReplicatedSubscriptionStats.class);
BrokerService brokerService = mock(BrokerService.class);
PersistentTopic topic = mock(PersistentTopic.class);
PersistentSubscription subscription = mock(PersistentSubscription.class);
ObservableFuture<Void> ackFuture = new ObservableFuture<>();

when(brokerService.pulsar()).thenReturn(pulsar);
when(brokerService.getPulsar()).thenReturn(pulsar);
when(pulsar.getExecutor()).thenReturn(executor);
when(pulsar.getConfiguration()).thenReturn(config);
when(pulsar.getOpenTelemetryReplicatedSubscriptionStats()).thenReturn(stats);
when(executor.scheduleAtFixedRate(any(Runnable.class), anyLong(), anyLong(), any(TimeUnit.class)))
.thenReturn(timer);

when(topic.getName()).thenReturn("persistent://public/default/t1");
when(topic.getBrokerService()).thenReturn(brokerService);
when(topic.getSubscription("sub")).thenReturn(subscription);
when(subscription.acknowledgeMessageAsync(any(), eq(AckType.Cumulative), any())).thenReturn(ackFuture);

ReplicatedSubscriptionsController controller = new ReplicatedSubscriptionsController(topic, "local");
ByteBuf marker = Markers.newReplicatedSubscriptionsUpdate("sub",
Map.of("local", new MarkersMessageIdData().setLedgerId(1).setEntryId(2)));
try {
Commands.skipMessageMetadata(marker);

controller.receivedReplicatedSubscriptionMarker(PositionFactory.create(3, 4),
MarkerType.REPLICATED_SUBSCRIPTION_UPDATE_VALUE, marker);

assertThat(ackFuture.isObserved())
.as("replicated subscription update ack failures should not be dropped")
.isTrue();
} finally {
marker.release();
controller.close();
}
}

@Test
@SuppressWarnings("unchecked")
public void testReplicatedSubscriptionUpdateCreatedSubscriptionAckFutureIsObserved() throws Exception {
PulsarService pulsar = mock(PulsarService.class);
ScheduledExecutorService executor = mock(ScheduledExecutorService.class);
@SuppressWarnings("rawtypes")
ScheduledFuture timer = mock(ScheduledFuture.class);
ServiceConfiguration config = new ServiceConfiguration();
config.setEnableReplicatedSubscriptions(true);
OpenTelemetryReplicatedSubscriptionStats stats = mock(OpenTelemetryReplicatedSubscriptionStats.class);
BrokerService brokerService = mock(BrokerService.class);
PersistentTopic topic = mock(PersistentTopic.class);
PersistentSubscription createdSubscription = mock(PersistentSubscription.class);
ObservableFuture<Void> ackFuture = new ObservableFuture<>();

when(brokerService.pulsar()).thenReturn(pulsar);
when(brokerService.getPulsar()).thenReturn(pulsar);
when(pulsar.getExecutor()).thenReturn(executor);
when(pulsar.getConfiguration()).thenReturn(config);
when(pulsar.getOpenTelemetryReplicatedSubscriptionStats()).thenReturn(stats);
when(executor.scheduleAtFixedRate(any(Runnable.class), anyLong(), anyLong(), any(TimeUnit.class)))
.thenReturn(timer);

when(topic.getName()).thenReturn("persistent://public/default/t1");
when(topic.getBrokerService()).thenReturn(brokerService);
when(topic.getSubscription("sub")).thenReturn(null);
when(topic.createSubscription(eq("sub"), eq(InitialPosition.Earliest), eq(true), any()))
.thenReturn(CompletableFuture.completedFuture(createdSubscription));
when(createdSubscription.acknowledgeMessageAsync(any(), eq(AckType.Cumulative), any())).thenReturn(ackFuture);

ReplicatedSubscriptionsController controller = new ReplicatedSubscriptionsController(topic, "local");
ByteBuf marker = Markers.newReplicatedSubscriptionsUpdate("sub",
Map.of("local", new MarkersMessageIdData().setLedgerId(1).setEntryId(2)));
try {
Commands.skipMessageMetadata(marker);

controller.receivedReplicatedSubscriptionMarker(PositionFactory.create(3, 4),
MarkerType.REPLICATED_SUBSCRIPTION_UPDATE_VALUE, marker);

verify(topic).createSubscription(eq("sub"), eq(InitialPosition.Earliest), eq(true), any());
assertThat(ackFuture.isObserved())
.as("replicated subscription update ack failures should be observed after subscription creation")
.isTrue();
} finally {
marker.release();
controller.close();
}
}

@Test
@SuppressWarnings("unchecked")
public void testFinalSnapshotMarkerPublishFailureKeepsSnapshotPending() {
Expand Down Expand Up @@ -306,4 +407,40 @@ public Set<Entry<String, Replicator>> entrySet() {
controller.close();
}
}

private static class ObservableFuture<T> extends CompletableFuture<T> {
private boolean observed;

boolean isObserved() {
return observed;
}

private void markObserved() {
observed = true;
}

@Override
public CompletableFuture<T> whenComplete(BiConsumer<? super T, ? super Throwable> action) {
markObserved();
return super.whenComplete(action);
}

@Override
public CompletableFuture<T> whenCompleteAsync(BiConsumer<? super T, ? super Throwable> action) {
markObserved();
return super.whenCompleteAsync(action);
}

@Override
public CompletableFuture<T> exceptionally(Function<Throwable, ? extends T> fn) {
markObserved();
return super.exceptionally(fn);
}

@Override
public <R> CompletableFuture<R> handle(BiFunction<? super T, Throwable, ? extends R> fn) {
markObserved();
return super.handle(fn);
}
}
}
Loading