Cloud Pub/Sub
Spring Cloud GCP provides an abstraction layer to publish to and subscribe from Google Cloud Pub/Sub topics and to create, list or delete Google Cloud Pub/Sub topics and subscriptions.
A Spring Boot starter is provided to auto-configure the various required Pub/Sub components.
Maven coordinates, using Spring Cloud GCP BOM:
<dependency>
<groupId>com.google.cloud</groupId>
<artifactId>spring-cloud-gcp-starter-pubsub</artifactId>
</dependency>
Gradle coordinates:
dependencies {
implementation("com.google.cloud:spring-cloud-gcp-starter-pubsub")
}
This starter is also available from Spring Initializr through the GCP Messaging
entry.
Configuration
The Spring Boot starter for Google Cloud Pub/Sub provides the following configuration options.
Spring Cloud GCP Pub/Sub API Configuration
This section describes options for enabling the integration, specifying the GCP project and credentials, and setting whether the APIs should connect to an emulator for local testing.
Name |
Description |
Required |
Default value |
|
Enables or disables Pub/Sub auto-configuration |
No |
|
|
GCP project ID where the Google Cloud Pub/Sub API is hosted, if different from the one in the Spring Cloud GCP Core Module |
No |
|
|
OAuth2 credentials for authenticating with the Google Cloud Pub/Sub API, if different from the ones in the Spring Cloud GCP Core Module |
No |
|
|
The host and port of the local running emulator. If provided, this will setup the client to connect against a running Google Cloud Pub/Sub Emulator. |
No |
|
|
Base64-encoded contents of OAuth2 account private key for authenticating with the Google Cloud Pub/Sub API, if different from the ones in the Spring Cloud GCP Core Module |
No |
|
|
OAuth2 scope for Spring Cloud GCP Pub/Sub credentials |
No |
Publisher/Subscriber Configuration
This section describes configuration options to customize the behavior of the application’s Pub/Sub publishers and subscribers. Subscriber settings can be either global or subscription-specific.
A custom configuration (injected through a setter in DefaultSubscriberFactory or a custom bean) will take precedence over auto-configuration.
Hence, if one wishes to use per-subscription configuration for a Pub/Sub setting, there must not be a custom bean for that setting.
When using auto-configuration, if both global and per-subscription configurations are provided, then the per-subscription configuration will be used.
However, if a per-subscription configuration is not set then the global or default configuration will be used.
|
Name |
Description |
Required |
Default value |
|
The number of pull workers |
No |
1 |
|
The maximum period a message ack deadline will be extended, in seconds |
No |
0 |
|
The endpoint for synchronous pulling messages |
No |
pubsub.googleapis.com:443 |
|
Number of threads used by |
No |
4 |
|
Maximum number of outstanding elements to keep in memory before enforcing flow control. |
No |
unlimited |
|
Maximum number of outstanding bytes to keep in memory before enforcing flow control. |
No |
unlimited |
|
The behavior when the specified limits are exceeded. |
No |
Block |
|
The element count threshold to use for batching. |
No |
1 (batching off) |
|
The request byte threshold to use for batching. |
No |
1 byte (batching off) |
|
The delay threshold to use for batching. After this amount of time has elapsed (counting from the first element added), the elements will be wrapped up in a batch and sent. |
No |
1 ms (batching off) |
|
Enables batching. |
No |
false |
|
Enables message ordering. |
No |
false |
|
The publisher endpoint.
Example: |
No |
pubsub.googleapis.com:443 |
Subscription-specific Configurations
Name |
Description |
Required |
Default value |
|
The fully-qualified subscription name in the |
No |
1 |
|
The number of pull workers. |
No |
1 |
|
The maximum period a message ack deadline will be extended, in seconds. |
No |
0 |
|
The endpoint for synchronous pulling messages. |
No |
pubsub.googleapis.com:443 |
|
Number of threads used by |
No |
4 |
|
Maximum number of outstanding elements to keep in memory before enforcing flow control. |
No |
unlimited |
|
Maximum number of outstanding bytes to keep in memory before enforcing flow control. |
No |
unlimited |
|
The behavior when the specified limits are exceeded. |
No |
Block |
GRPC Connection Settings
The Pub/Sub API uses the GRPC protocol to send API requests to the Pub/Sub service. This section describes configuration options for customizing the GRPC behavior.
The properties that refer to retry control the RPC retries for transient failures during the gRPC call to Cloud Pub/Sub server.
They do not control message redelivery; only message acknowledgement deadline can be used to extend or shorten the amount of time until Pub/Sub attempts redelivery.
|
Name |
Description |
Required |
Default value |
|
Determines frequency of keepalive gRPC ping |
No |
|
|
RPC status codes that should be retried when pulling messages. |
No |
UNKNOWN,ABORTED,UNAVAILABLE |
|
TotalTimeout has ultimate control over how long the logic should keep trying the remote call until it gives up completely. The higher the total timeout, the more retries can be attempted. |
No |
0 |
|
InitialRetryDelay controls the delay before the first retry. Subsequent retries will use this value adjusted according to the RetryDelayMultiplier. |
No |
0 |
|
RetryDelayMultiplier controls the change in retry delay. The retry delay of the previous call is multiplied by the RetryDelayMultiplier to calculate the retry delay for the next call. |
No |
1 |
|
MaxRetryDelay puts a limit on the value of the retry delay, so that the RetryDelayMultiplier can’t increase the retry delay higher than this amount. |
No |
0 |
|
MaxAttempts defines the maximum number of attempts to perform. If this value is greater than 0, and the number of attempts reaches this limit, the logic will give up retrying even if the total retry time is still lower than TotalTimeout. |
No |
0 |
|
Jitter determines if the delay time should be randomized. |
No |
true |
|
InitialRpcTimeout controls the timeout for the initial RPC. Subsequent calls will use this value adjusted according to the RpcTimeoutMultiplier. |
No |
0 |
|
RpcTimeoutMultiplier controls the change in RPC timeout. The timeout of the previous call is multiplied by the RpcTimeoutMultiplier to calculate the timeout for the next call. |
No |
1 |
|
MaxRpcTimeout puts a limit on the value of the RPC timeout, so that the RpcTimeoutMultiplier can’t increase the RPC timeout higher than this amount. |
No |
0 |
Subscription-specific Configuration
Name |
Description |
Required |
Default value |
|
RPC status codes that should be retried when pulling messages. |
No |
UNKNOWN,ABORTED,UNAVAILABLE |
|
TotalTimeout has ultimate control over how long the logic should keep trying the remote call until it gives up completely. The higher the total timeout, the more retries can be attempted. |
No |
0 |
|
InitialRetryDelay controls the delay before the first retry. Subsequent retries will use this value adjusted according to the RetryDelayMultiplier. |
No |
0 |
|
RetryDelayMultiplier controls the change in retry delay. The retry delay of the previous call is multiplied by the RetryDelayMultiplier to calculate the retry delay for the next call. |
No |
1 |
|
MaxRetryDelay puts a limit on the value of the retry delay, so that the RetryDelayMultiplier can’t increase the retry delay higher than this amount. |
No |
0 |
|
MaxAttempts defines the maximum number of attempts to perform. If this value is greater than 0, and the number of attempts reaches this limit, the logic will give up retrying even if the total retry time is still lower than TotalTimeout. |
No |
0 |
|
Jitter determines if the delay time should be randomized. |
No |
true |
|
InitialRpcTimeout controls the timeout for the initial RPC. Subsequent calls will use this value adjusted according to the RpcTimeoutMultiplier. |
No |
0 |
|
RpcTimeoutMultiplier controls the change in RPC timeout. The timeout of the previous call is multiplied by the RpcTimeoutMultiplier to calculate the timeout for the next call. |
No |
1 |
|
MaxRpcTimeout puts a limit on the value of the RPC timeout, so that the RpcTimeoutMultiplier can’t increase the RPC timeout higher than this amount. |
No |
0 |
Programmatic Configuration
To apply publishing customizations not covered by the properties above, you may provide custom beans of type PublisherCustomizer
to post-process the Publisher.Builder
object right before it is built into a Publisher
.
The PublisherCustomizer
beans may be annotated with Spring Framework’s @Order
annotation to ensure they are applied in a particular sequence.
Spring Boot Actuator Support
Cloud Pub/Sub Health Indicator
If you are using Spring Boot Actuator, you can take advantage of the Cloud Pub/Sub health indicator called pubsub
.
The health indicator will verify whether Cloud Pub/Sub is up and accessible by your application.
To enable it, all you need to do is add the Spring Boot Actuator to your project.
The pubsub
indicator will then roll up to the overall application status visible at http://localhost:8080/actuator/health (use the management.endpoint.health.show-details
property to view per-indicator details).
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-actuator</artifactId>
</dependency>
If your application already has actuator and Cloud Pub/Sub starters, this health indicator is enabled by default.
To disable the Cloud Pub/Sub indicator, set management.health.pubsub.enabled to false .
|
The health indicator validates the connection to Pub/Sub by pulling messages from a Pub/Sub subscription.
If no subscription has been specified via spring.cloud.gcp.pubsub.health.subscription
, it will pull messages from a random subscription that is expected not to exist.
It will signal "up" if it is able to connect to GCP Pub/Sub APIs, i.e. the pull results in a response of NOT_FOUND
or PERMISSION_DENIED
.
If a custom subscription has been specified, this health indicator will signal "up" if messages are successfully pulled and (optionally) acknowledged, or when a successful pull is performed but no messages are returned from Pub/Sub.
Note that messages pulled from the subscription will not be acknowledged, unless you set the spring.cloud.gcp.pubsub.health.acknowledge-messages
option to true
.
So, take care not to configure a subscription that has a business impact, or instead leave the custom subscription out completely.
Name |
Description |
Required |
Default value |
|
Whether to enable the Pub/Sub health indicator |
No |
|
|
Subscription to health check against by pulling a message |
No |
Random non-existent |
|
Milliseconds to wait for response from Pub/Sub before timing out |
No |
|
|
Whether to acknowledge messages pulled from the optionally specified subscription |
No |
|
Cloud Pub/Sub Subscription Health Indicator
If you are using Spring Boot Actuator, you can take advantage of the Cloud Pub/Sub subscription health indicator called pubsub-subscriber
.
The subscription health indicator will verify whether Pub/Sub subscriptions are actively processing messages from the subscription’s backlog.
To enable it, you need to add the Spring Boot Actuator to your project and the GCP Monitoring.
Also you need to set the following properties spring.cloud.gcp.pubsub.health.lagThreshold
, spring.cloud.gcp.pubsub.health.backlogThreshold
.
The pubsub-subscriber
indicator will then roll up to the overall application status visible at http://localhost:8080/actuator/health (use the management.endpoint.health.show-details
property to view per-indicator details).
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-actuator</artifactId>
</dependency>
<dependency>
<groupId>com.google.cloud</groupId>
<artifactId>google-cloud-monitoring</artifactId>
</dependency>
The health indicator validates a subscriber’s health by checking the subscription’s message backlog and the last processed message.
A subscription’s backlog is retrieved using Google Cloud’s Monitoring Metrics. The metric used is the num_undelivered_messages
for a subscription.
If a message has been recently processed in a reasonable time threshold, then the subscriber is healthy. If the backlog of messages for a subscription is big but the subscriber consumes messages then subscriber is still healthy. If there hasn’t been any processing of recent messages but the backlog increases, then the subscriber is unhealthy.
The health indicator will not behave entirely as expected if Dead Letter Queueing is enabled on the subscription being checked, num_undelivered_messages will drop down by itself after DLQ threshold is reached. |
Name |
Description |
Required |
Default value |
|
Whether to enable the Pub/Sub Subscription health indicator |
No |
|
|
Threshold in seconds over message processing lag |
Yes |
Provided |
|
The threshold number of messages for a subscription backlog |
Yes |
Provided |
|
The optional interval in seconds for subscription backlog lookup |
No |
|
|
Number of threads used for Health Check Executors |
No |
|
Pub/Sub Operations & Template
PubSubOperations
is an abstraction that allows Spring users to use Google Cloud Pub/Sub without depending on any Google Cloud Pub/Sub API semantics.
It provides the common set of operations needed to interact with Google Cloud Pub/Sub.
PubSubTemplate
is the default implementation of PubSubOperations
and it uses the Google Cloud Java Client for Pub/Sub to interact with Google Cloud Pub/Sub.
Publishing to a topic
PubSubTemplate
provides asynchronous methods to publish messages to a Google Cloud Pub/Sub topic.
The publish()
method takes in a topic name to post the message to, a payload of a generic type and, optionally, a map with the message headers.
The topic name could either be a canonical topic name within the current project, or the fully-qualified name referring to a topic in a different project using the projects/<project_name>/topics/<topic_name>
format.
Here is an example of how to publish a message to a Google Cloud Pub/Sub topic:
Map<String, String> headers = Collections.singletonMap("key1", "val1");
pubSubTemplate.publish(topicName, "message", headers).get();
By default, the SimplePubSubMessageConverter
is used to convert payloads of type byte[]
, ByteString
, ByteBuffer
, and String
to Pub/Sub messages.
Ordering messages
If you are relying on message converters and would like to provide an ordering key, use the GcpPubSubHeaders.ORDERING_KEY
header.
You will also need to make sure to enable message ordering on the publisher via the spring.cloud.gcp.pubsub.publisher.enable-message-ordering
property.
Additionally, if you are using multiple publishers, you will want to set the spring.cloud.gcp.pubsub.publisher.endpoint
to a regional endpoint such as "us-east1-pubsub.googleapis.com:443"
so that messages are sent to the same region and received in order.
Map<String, String> headers =
Collections.singletonMap(GcpPubSubHeaders.ORDERING_KEY, "key1");
pubSubTemplate.publish(topicName, "message1", headers).get();
pubSubTemplate.publish(topicName, "message2", headers).get();
Subscribing to a subscription
Google Cloud Pub/Sub allows many subscriptions to be associated to the same topic.
PubSubTemplate
allows you to listen to subscriptions via the subscribe()
method.
When listening to a subscription, messages will be pulled from Google Cloud Pub/Sub asynchronously and passed to a user provided message handler.
The subscription name could either be a canonical subscription name within the current project, or the fully-qualified name referring to a subscription in a different project using the projects/<project_name>/subscriptions/<subscription_name>
format.
Example
Subscribe to a subscription with a message handler:
Subscriber subscriber =
pubSubTemplate.subscribe(
subscriptionName,
message -> {
logger.info(
"Message received from "
+ subscriptionName
+ " subscription: "
+ message.getPubsubMessage().getData().toStringUtf8());
message.ack();
});
Subscribe methods
PubSubTemplate
provides the following subscribe methods:
subscribe(String subscription, Consumer<BasicAcknowledgeablePubsubMessage> messageConsumer) |
asynchronously pulls messages and passes them to |
---|---|
subscribeAndConvert(String subscription, Consumer<ConvertedBasicAcknowledgeablePubsubMessage<T>> messageConsumer, Class<T> payloadType) |
same as |
As of version 1.2, subscribing by itself is not enough to keep an application running.
For a command-line application, you may want to provide your own ThreadPoolTaskScheduler bean named pubsubSubscriberThreadPool , which by default creates non-daemon threads that will keep an application from stopping.
This default behavior has been overridden in Spring Cloud GCP for consistency with Cloud Pub/Sub client library, and to avoid holding up command-line applications that would like to shut down once their work is done.
|
Pulling messages from a subscription
Google Cloud Pub/Sub supports synchronous pulling of messages from a subscription. This is different from subscribing to a subscription, in the sense that subscribing is an asynchronous task.
Example
Pull up to 10 messages:
int maxMessages = 10;
boolean returnImmediately = false;
List<AcknowledgeablePubsubMessage> messages =
pubSubTemplate.pull(subscriptionName, maxMessages, returnImmediately);
// acknowledge the messages
pubSubTemplate.ack(messages);
messages.forEach(
message ->
logger.info(message.getPubsubMessage().getData().toStringUtf8()));
Pull methods
PubsubTemplate
provides the following pull methods:
pull(String subscription, Integer maxMessages, Boolean returnImmediately) |
Pulls a number of messages from a subscription, allowing for the retry settings to be configured.
Any messages received by The If |
---|---|
pullAndAck |
Works the same as the |
pullNext |
Allows for a single message to be pulled and automatically acknowledged from a subscription. |
pullAndConvert |
Works the same as the |
We do not recommend setting returnImmediately to true , as it may result in delayed message delivery.
"Immediately" really means 1 second, and if Pub/Sub cannot retrieve any messages from the backend in that time, it will return 0 messages, despite having messages queue up on the topic.
Therefore, we recommend setting returnImmediately to false , or using subscribe methods from the previous section.
|
Acknowledging messages
There are two ways to acknowledge messages.
-
To acknowledge multiple messages at once, you can use the
PubSubTemplate.ack()
method. You can also use thePubSubTemplate.nack()
for negatively acknowledging messages. Using these methods for acknowledging messages in batches is more efficient than acknowledging messages individually, but they require the collection of messages to be from the same project. -
To acknowledge messages individually you can use the
ack()
ornack()
method on each of them (to acknowledge or negatively acknowledge, correspondingly).
All ack() , nack() , and modifyAckDeadline() methods on messages, as well as PubSubSubscriberTemplate , are implemented asynchronously, returning a ListenableFuture<Void> to enable asynchronous processing.
|
Dead Letter Topics
Your application may occasionally receive a message it cannot process.
If you create your Subscription
passing the Subscription.Builder
argument, you can specify a DeadLetterPolicy
that will forward all nack()
-ed and non-ack()
-ed messages after a configurable amount of redelivery attempts.
See here for more information.
public Subscription newSubscription() {
// Must use the fully-qualified topic name.
String fullDeadLetterTopic = PubSubTopicUtils
.toTopicName(DEAD_LETTER_TOPIC, gcpProjectIdProvider.getProjectId())
.toString();
return pubSubAdmin.createSubscription(Subscription.newBuilder()
.setName(SUBSCRIPTION_NAME)
.setTopic(TOPIC_NAME)
.setDeadLetterPolicy(DeadLetterPolicy.newBuilder()
.setDeadLetterTopic(fullDeadLetterTopic)
.setMaxDeliveryAttempts(6)
.build()));
}
Dead letter topics are no different than any other topic, though some additional permissions are necessary to ensure the Cloud Pub/Sub service can successfully ack
the original message and re-publish
it on the dead letter topic.
JSON support
For serialization and deserialization of POJOs using Jackson JSON, configure a PubSubMessageConverter
bean, and the Spring Boot starter for GCP Pub/Sub will automatically wire it into the PubSubTemplate
.
// Note: The ObjectMapper is used to convert Java POJOs to and from JSON.
// You will have to configure your own instance if you are unable to depend
// on the ObjectMapper provided by Spring Boot starters.
@Bean
public PubSubMessageConverter pubSubMessageConverter() {
return new JacksonPubSubMessageConverter(new ObjectMapper());
}
Alternatively, you can set it directly by calling the setMessageConverter() method on the PubSubTemplate .
Other implementations of the PubSubMessageConverter can also be configured in the same manner.
|
Assuming you have the following class defined:
static class TestUser {
String username;
String password;
public String getUsername() {
return this.username;
}
void setUsername(String username) {
this.username = username;
}
public String getPassword() {
return this.password;
}
void setPassword(String password) {
this.password = password;
}
}
You can serialize objects to JSON on publish automatically:
TestUser user = new TestUser();
user.setUsername("John");
user.setPassword("password");
pubSubTemplate.publish(topicName, user);
And that’s how you convert messages to objects on pull:
int maxMessages = 1;
boolean returnImmediately = false;
List<ConvertedAcknowledgeablePubsubMessage<TestUser>> messages =
pubSubTemplate.pullAndConvert(
subscriptionName, maxMessages, returnImmediately, TestUser.class);
ConvertedAcknowledgeablePubsubMessage<TestUser> message = messages.get(0);
// acknowledge the message
message.ack();
TestUser receivedTestUser = message.getPayload();
Please refer to our Pub/Sub JSON Payload Sample App as a reference for using this functionality.
Reactive Stream Subscriber
It is also possible to acquire a reactive stream backed by a subscription.
To do so, a Project Reactor dependency (io.projectreactor:reactor-core
) must be added to the project.
The combination of the Pub/Sub starter and the Project Reactor dependencies will then make a PubSubReactiveFactory
bean available, which can then be used to get a Publisher
.
@Autowired
PubSubReactiveFactory reactiveFactory;
// ...
Flux<AcknowledgeablePubsubMessage> flux
= reactiveFactory.poll("exampleSubscription", 1000);
The Flux
then represents an infinite stream of GCP Pub/Sub messages coming in through the specified subscription.
For unlimited demand, the Pub/Sub subscription will be polled regularly, at intervals determined by pollingPeriodMs
parameter passed in when creating the Flux
.
For bounded demand, the pollingPeriodMs
parameter is unused.
Instead, as many messages as possible (up to the requested number) are delivered immediately, with the remaining messages delivered as they become available.
Any exceptions thrown by the underlying message retrieval logic will be passed as an error to the stream.
The error handling operators (Flux#retry()
, Flux#onErrorResume()
etc.) can be used to recover.
The full range of Project Reactor operations can be applied to the stream.
For example, if you only want to fetch 5 messages, you can use limitRequest
operation to turn the infinite stream into a finite one:
Flux<AcknowledgeablePubsubMessage> fiveMessageFlux = flux.limitRequest(5);
Messages flowing through the Flux
should be manually acknowledged.
flux.doOnNext(AcknowledgeablePubsubMessage::ack);
Pub/Sub management
PubSubAdmin
is the abstraction provided by Spring Cloud GCP to manage Google Cloud Pub/Sub resources.
It allows for the creation, deletion and listing of topics and subscriptions.
Generally when referring to topics and subscriptions, you can either use the short canonical name within the current project, or the fully-qualified name referring to a topic or subscription in a different project using the projects/<project_name>/(topics|subscriptions)/<name> format.
|
The Spring Boot starter for GCP Pub/Sub auto-configures a PubSubAdmin
object using the GcpProjectIdProvider
and the CredentialsProvider
auto-configured by the Spring Boot GCP Core starter.
Creating a topic
PubSubAdmin
implements a method to create topics:
public Topic createTopic(String topicName)
Here is an example of how to create a Google Cloud Pub/Sub topic:
public void newTopic() {
pubSubAdmin.createTopic("topicName");
}
Deleting a topic
PubSubAdmin
implements a method to delete topics:
public void deleteTopic(String topicName)
Here is an example of how to delete a Google Cloud Pub/Sub topic:
public void deleteTopic() {
pubSubAdmin.deleteTopic("topicName");
}
Listing topics
PubSubAdmin
implements a method to list topics:
public List<Topic> listTopics
Here is an example of how to list every Google Cloud Pub/Sub topic name in a project:
List<String> topics =
pubSubAdmin.listTopics().stream().map(Topic::getName).collect(Collectors.toList());
Creating a subscription
PubSubAdmin
implements several methods to create subscriptions to existing topics:
public Subscription createSubscription(String subscriptionName, String topicName)
public Subscription createSubscription(String subscriptionName, String topicName, Integer ackDeadline)
public Subscription createSubscription(String subscriptionName, String topicName, Integer ackDeadline, String pushEndpoint)
public Subscription createSubscription(Subscriber.Builder builder)
The default value for ackDeadline
is 10 seconds.
If pushEndpoint
isn’t specified, the subscription uses message pulling, instead.
You can also pass a Subscription.Builder
for full control over any options or features available in the client library.
Here is an example of how to create a Google Cloud Pub/Sub subscription:
public Subscription newSubscription() {
return pubSubAdmin.createSubscription("subscriptionName", "topicName", 15);
}
Deleting a subscription
PubSubAdmin
implements a method to delete subscriptions:
public void deleteSubscription(String subscriptionName)
Here is an example of how to delete a Google Cloud Pub/Sub subscription:
public void deleteSubscription() {
pubSubAdmin.deleteSubscription("subscriptionName");
}
Listing subscriptions
PubSubAdmin
implements a method to list subscriptions:
public List<Subscription> listSubscriptions()
Here is an example of how to list every subscription name in a project:
List<String> subscriptions =
pubSubAdmin.listSubscriptions().stream()
.map(Subscription::getName)
.collect(Collectors.toList());
Sample
Sample applications for using the template and using a subscription-backed reactive stream are available.