Vert.x Reference Guide
Vert.x is a toolkit for building reactive applications. As described in the Quarkus Reactive Architecture, Quarkus uses Vert.x underneath.
This guide is the companion to the Using Eclipse Vert.x API from a Quarkus Application guide. It provides more advanced details about the usage and the configuration of the Vert.x instance used by Quarkus.
Access the Vert.x instance
To access the managed Vert.x instance, add the quarkus-vertx extension to your project.
This dependency might already be available in your project (as a transitive dependency).
With this extension, you can retrieve the managed instance of Vert.x using either field or constructor injection:
@ApplicationScoped
public class MyBean {
// Field injection
@Inject Vertx vertx;
// Constructor injection
MyBean(Vertx vertx) {
// ...
}
}
You can inject either the:
-
io.vertx.core.Vertxinstance exposing the bare Vert.x API -
io.vertx.mutiny.core.Vertxinstance exposing the Mutiny API
We recommend using the Mutiny variant as it integrates with the other reactive APIs provided by Quarkus.
|
Mutiny
If you are not familiar with Mutiny, check Mutiny - an intuitive reactive programming library. |
Documentation about the Vert.x Mutiny variant is available on https://smallrye.io/smallrye-mutiny-vertx-bindings.
Configure the Vert.x instance
You can configure the Vert.x instance from the application.properties file.
The following table lists the supported properties:
Configuration property fixed at build time - All other configuration properties are overridable at runtime
Configuration property |
Type |
Default |
|---|---|---|
Whether and how strictly native transport should be used. Supported values:
Environment variable: Show more |
|
|
The type of native transport to use. When set to a value other than Setting this to a specific transport type implicitly enables native transport (i.e., Supported values:
Note that, if not set to Environment variable: Show more |
|
|
Enables or disables the Vert.x cache. Environment variable: Show more |
boolean |
|
Configure the file cache directory. When not set, the cache is stored in the system temporary directory (read from the Note that this property is ignored if the Environment variable: Show more |
string |
|
Enables or disabled the Vert.x classpath resource resolver. Environment variable: Show more |
boolean |
|
The number of event loops. By default, it matches the number of CPUs detected on the system. Environment variable: Show more |
int |
|
The maximum amount of time the event loop can be blocked. Environment variable: Show more |
|
|
The amount of time before a warning is displayed if the event loop is blocked. Environment variable: Show more |
|
|
The maximum amount of time the worker thread can be blocked. Environment variable: Show more |
|
|
The size of the internal thread pool (used for the file system). Environment variable: Show more |
int |
|
The queue size. For most applications this should be unbounded Environment variable: Show more |
int |
|
The executor growth resistance. A resistance factor applied after the core pool is full; values applied here will cause that fraction of submissions to create new threads when no idle thread is available. A value of Environment variable: Show more |
float |
|
The amount of time a thread will stay alive with no work. Environment variable: Show more |
|
|
Prefill thread pool when creating a new Executor. When Environment variable: Show more |
boolean |
|
Enables the async DNS resolver. Environment variable: Show more |
boolean |
|
The maximum amount of time that a successfully resolved address will be cached. If not set explicitly, resolved addresses may be cached forever. The value must be a duration of at least 1 second. Environment variable: Show more |
|
|
The minimum amount of time that a successfully resolved address will be cached. Environment variable: Show more |
|
|
The amount of time that an unsuccessful attempt to resolve an address will be cached. Environment variable: Show more |
|
|
The maximum number of queries to be sent during a resolution. Environment variable: Show more |
int |
|
The duration after which a DNS query is considered to be failed. Environment variable: Show more |
|
|
Set the path of an alternate hosts configuration file to use instead of the one provided by the os. The default value is Environment variable: Show more |
string |
|
The hosts configuration refresh period. The resolver caches the hosts configuration (configured using A value of Environment variable: Show more |
|
|
Set the list of DNS server addresses, an address is the IP of the dns server, followed by an optional colon and a port, e.g Environment variable: Show more |
list of string |
|
Set to true to enable the automatic inclusion in DNS queries of an optional record that hints the remote DNS server about how much data the resolver can read per response. Environment variable: Show more |
boolean |
|
Set the DNS queries Recursion Desired flag value. Environment variable: Show more |
boolean |
|
Set the lists of DNS search domains. When the search domain list is null, the effective search domain list will be populated using the system DNS search domains. Environment variable: Show more |
list of string |
|
Set the ndots value used when resolving using search domains, the default value is Environment variable: Show more |
int |
|
Set to Environment variable: Show more |
boolean |
|
Set to Environment variable: Show more |
boolean |
|
The interval of the blocked thread checker. The blocked thread checker periodically checks if any event loop or worker thread has been blocked for longer than the configured thresholds ( Environment variable: Show more |
|
|
This property is deprecated: use The size of the worker pool size. Environment variable: Show more |
int |
|
|
About the Duration format
To write duration values, use the standard You can also use a simplified format, starting with a number:
In other cases, the simplified format is translated to the
|
See Customize the Vert.x configuration to configure the Vert.x instance using a programmatic approach.
Use Vert.x clients
In addition to Vert.x core, you can use most Vert.x ecosystem libraries. Some Quarkus extension already wraps Vert.x libraries.
Available APIs
The following table lists the most used libraries from the Vert.x ecosystem. To access these APIs, add the indicated extension or dependency to your project. Check the associated documentation to learn how to use them.
API |
Extension or Dependency |
Documentation |
AMQP Client |
|
|
Circuit Breaker |
|
|
Consul Client |
|
|
DB2 Client |
|
|
Kafka Client |
|
|
Mail Client |
|
|
MQTT Client |
|
No guide yet |
MS SQL Client |
|
|
MySQL Client |
|
|
Oracle Client |
|
|
PostgreSQL Client |
|
|
RabbitMQ Client |
|
|
Redis Client |
|
|
Web Client |
|
To learn more about the usage of the Vert.x Mutiny API, refer to https://smallrye.io/smallrye-mutiny-vertx-bindings.
Use the Vert.x Web Client
This section gives an example using the Vert.x WebClient in the context of a Quarkus REST (formerly RESTEasy Reactive) application.
As indicated in the table above, add the following dependency to your project:
<dependency>
<groupId>io.smallrye.reactive</groupId>
<artifactId>smallrye-mutiny-vertx-web-client</artifactId>
</dependency>
implementation("io.smallrye.reactive:smallrye-mutiny-vertx-web-client")
Now, in your code, you can create an instance of WebClient:
package org.acme.vertx;
import jakarta.annotation.PostConstruct;
import jakarta.inject.Inject;
import jakarta.ws.rs.GET;
import jakarta.ws.rs.Path;
import jakarta.ws.rs.Produces;
import jakarta.ws.rs.core.MediaType;
import io.smallrye.mutiny.Uni;
import io.vertx.mutiny.core.Vertx;
import io.vertx.mutiny.ext.web.client.WebClient;
import io.vertx.core.json.JsonObject;
import io.vertx.ext.web.client.WebClientOptions;
@Path("/fruit-data")
public class ResourceUsingWebClient {
private final WebClient client;
@Inject
VertxResource(Vertx vertx) {
this.client = WebClient.create(vertx);
}
@GET
@Produces(MediaType.APPLICATION_JSON)
@Path("/{name}")
public Uni<JsonObject> getFruitData(String name) {
return client.getAbs("https://.../api/fruit/" + name)
.send()
.onItem().transform(resp -> {
if (resp.statusCode() == 200) {
return resp.bodyAsJsonObject();
} else {
return new JsonObject()
.put("code", resp.statusCode())
.put("message", resp.bodyAsString());
}
});
}
}
This resource creates a WebClient and, upon request, uses this client to invoke a remote HTTP API.
Depending on the result, the response is forwarded as received, or it creates a JSON object wrapping the error.
The WebClient is asynchronous (and non-blocking), to the endpoint returns a Uni.
The application can also run as a native executable.
But, first, we need to instruct Quarkus to enable ssl (if the remote API uses HTTPS).
Open the src/main/resources/application.properties and add:
quarkus.ssl.native=true
Then, create the native executable with:
quarkus build --native
./mvnw install -Dnative
./gradlew build -Dquarkus.native.enabled=true
Use Vert.x JSON
Vert.x APIs often rely on JSON.
Vert.x provides two convenient classes to manipulate JSON document: io.vertx.core.json.JsonObject and io.vertx.core.json.JsonArray.
JsonObject can be used to map an object into its JSON representation and build an object from a JSON document:
// Map an object into JSON
Person person = ...;
JsonObject json = JsonObject.mapFrom(person);
// Build an object from JSON
json = new JsonObject();
person = json.mapTo(Person.class);
Note that these features use the mapper managed by the quarkus-jackson extension.
Refer to Jackson configuration to customize the mapping.
JSON Object and JSON Array are both supported as Quarkus HTTP endpoint requests and response bodies (using classic RESTEasy and Quarkus REST). Consider these endpoints:
package org.acme.vertx;
import io.vertx.core.json.JsonObject;
import io.vertx.core.json.JsonArray;
import jakarta.ws.rs.GET;
import jakarta.ws.rs.Path;
import jakarta.ws.rs.Produces;
import jakarta.ws.rs.core.MediaType;
@Path("/hello")
@Produces(MediaType.APPLICATION_JSON)
public class VertxJsonResource {
@GET
@Path("{name}/object")
public JsonObject jsonObject(String name) {
return new JsonObject().put("Hello", name);
}
@GET
@Path("{name}/array")
public JsonArray jsonArray(String name) {
return new JsonArray().add("Hello").add(name);
}
}
{"Hello":"Quarkus"}
["Hello","Quarkus"]
This works equally well when the JSON content is a request body or is wrapped in a Uni, Multi, CompletionStage or Publisher.
Use Verticles
Verticles is "a simple, scalable, actor-like deployment and concurrency model" provided by _Vert.x. This model does not claim to be a strict actor-model implementation, but it shares similarities, especially concerning concurrency, scaling, and deployment. To use this model, you write and deploy verticles, communicating by sending messages on the event bus.
You can deploy verticles in Quarkus. It supports:
-
bare verticle - Java classes extending
io.vertx.core.AbstractVerticle -
Mutiny verticle - Java classes extending
io.smallrye.mutiny.vertx.core.AbstractVerticle
Deploy Verticles
To deploy verticles, use the deployVerticle method:
@Inject Vertx vertx;
// ...
vertx.deployVerticle(MyVerticle.class.getName(), ar -> { });
vertx.deployVerticle(new MyVerticle(), ar -> { });
If you use the Mutiny-variant of Vert.x, be aware that the deployVerticle method returns a Uni, and you would need to trigger a subscription to make the actual deployment.
| An example explaining how to deploy verticles during the initialization of the application will follow. |
Use @ApplicationScoped beans as Verticle
In general, Vert.x verticles are not CDI beans. And so cannot use injection. However, in Quarkus, you can deploy verticles as beans. Note that in this case, CDI (Arc in Quarkus) is responsible for creating the instance.
The following snippet provides an example:
package io.quarkus.vertx.verticles;
import io.smallrye.mutiny.Uni;
import io.smallrye.mutiny.vertx.core.AbstractVerticle;
import org.eclipse.microprofile.config.inject.ConfigProperty;
import jakarta.enterprise.context.ApplicationScoped;
@ApplicationScoped
public class MyBeanVerticle extends AbstractVerticle {
@ConfigProperty(name = "address") String address;
@Override
public Uni<Void> asyncStart() {
return vertx.eventBus().consumer(address)
.handler(m -> m.replyAndForget("hello"))
.completionHandler();
}
}
You don’t have to inject the vertx instance; instead, leverage the protected field from AbstractVerticle.
Then, deploy the verticle instances with:
package io.quarkus.vertx.verticles;
import io.quarkus.runtime.StartupEvent;
import io.vertx.mutiny.core.Vertx;
import jakarta.enterprise.context.ApplicationScoped;
import jakarta.enterprise.event.Observes;
@ApplicationScoped
public class VerticleDeployer {
public void init(@Observes StartupEvent e, Vertx vertx, MyBeanVerticle verticle) {
vertx.deployVerticle(verticle).await().indefinitely();
}
}
If you want to deploy every exposed AbstractVerticle, you can use:
public void init(@Observes StartupEvent e, Vertx vertx, Instance<AbstractVerticle> verticles) {
for (AbstractVerticle verticle : verticles) {
vertx.deployVerticle(verticle).await().indefinitely();
}
}
Create multiple verticles instances
When using @ApplicationScoped, you will get a single instance for your verticle.
Having multiple instances of verticles can be helpful to share the load among them.
Each of them will be associated with a different I/O thread (Vert.x event loop).
To deploy multiple instances of your verticle, use the @Dependent scope instead of @ApplicationScoped:
package org.acme.verticle;
import io.smallrye.mutiny.Uni;
import io.smallrye.mutiny.vertx.core.AbstractVerticle;
import jakarta.enterprise.context.Dependent;
import jakarta.inject.Inject;
@Dependent
public class MyVerticle extends AbstractVerticle {
@Override
public Uni<Void> asyncStart() {
return vertx.eventBus().consumer("address")
.handler(m -> m.reply("Hello from " + this))
.completionHandler();
}
}
Then, deploy your verticle as follows:
package org.acme.verticle;
import io.quarkus.runtime.StartupEvent;
import io.vertx.core.DeploymentOptions;
import io.vertx.mutiny.core.Vertx;
import jakarta.enterprise.context.ApplicationScoped;
import jakarta.enterprise.event.Observes;
import jakarta.enterprise.inject.Instance;
import jakarta.inject.Inject;
@ApplicationScoped
public class MyApp {
void init(@Observes StartupEvent ev, Vertx vertx, Instance<MyVerticle> verticles) {
vertx
.deployVerticle(verticles::get, new DeploymentOptions().setInstances(2))
.await().indefinitely();
}
}
The init method receives an Instance<MyVerticle>.
Then, you pass a supplier to the deployVerticle method.
The supplier is just calling the get() method.
Thanks to the @Dependent scope, it returns a new instance on every call.
Finally, you pass the desired number of instances to the DeploymentOptions, such as two in the previous example.
It will call the supplier twice, which will create two instances of your verticle.
Use the Event Bus
Vert.x comes with a built-in event bus that you can use from your Quarkus application. So, your application components (CDI beans, resources…) can interact using asynchronous events, thus promoting loose-coupling.
With the event bus, you send messages to virtual addresses. The event bus offers three types of delivery mechanisms:
-
point-to-point - send the message, one consumer receives it. If several consumers listen to the address, a round-robin is applied;
-
publish/subscribe - publish a message; all the consumers listening to the address are receiving the message;
-
request/reply - send the message and expect a response. The receiver can respond to the message in an asynchronous fashion.
All these delivery mechanisms are non-blocking and are providing one of the fundamental bricks to build reactive applications.
Consume events
While you can use the Vert.x API to register consumers, Quarkus comes with declarative support.
To consume events, use the io.quarkus.vertx.ConsumeEvent annotation:
package org.acme.vertx;
import io.quarkus.vertx.ConsumeEvent;
import jakarta.enterprise.context.ApplicationScoped;
@ApplicationScoped
public class GreetingService {
@ConsumeEvent (1)
public String consume(String name) { (2)
return name.toUpperCase();
}
}
| 1 | If not set, the address is the fully qualified name of the bean; for instance, in this snippet, it’s org.acme.vertx.GreetingService. |
| 2 | The method parameter is the message body. If the method returns something, it’s the message response. |
Configure the address
The @ConsumeEvent annotation can be configured to set the address:
@ConsumeEvent("greeting") (1)
public String consume(String name) {
return name.toUpperCase();
}
| 1 | Receive the messages sent to the greeting address |
The address value can be a property expression.
In this case, the configured value is used instead: @ConsumeEvent("${my.consumer.address}").
Additionally, the property expression can specify a default value: @ConsumeEvent("${my.consumer.address:defaultAddress}").
@ConsumeEvent("${my.consumer.address}") (1)
public String consume(String name) {
return name.toLowerCase();
}
| 1 | Receive the messages sent to the address configured with the my.consumer.address key. |
| If no config property with the specified key exists and no default value is set then the application startup fails. |
Process events asynchronously
The previous examples use synchronous processing.
Asynchronous processing is also possible by returning either an io.smallrye.mutiny.Uni or a java.util.concurrent.CompletionStage:
package org.acme.vertx;
import io.quarkus.vertx.ConsumeEvent;
import jakarta.enterprise.context.ApplicationScoped;
import java.util.concurrent.CompletableFuture;
import java.util.concurrent.CompletionStage;
import io.smallrye.mutiny.Uni;
@ApplicationScoped
public class GreetingService {
@ConsumeEvent
public Uni<String> process(String name) {
// return an Uni completed when the processing is finished.
// You can also fail the Uni explicitly
}
}
|
Mutiny
The previous example uses Mutiny reactive types. If you are not familiar with Mutiny, check Mutiny - an intuitive reactive programming library. |
Blocking processing of events
By default, the code consuming the event must be non-blocking, as it’s called on an I/O thread.
If your processing is blocking, use the @io.smallrye.common.annotation.Blocking annotation:
@ConsumeEvent(value = "blocking-consumer")
@Blocking
void consumeBlocking(String message) {
// Something blocking
}
Alternatively, you can use the blocking attribute from the @ConsumeEvent annotation:
@ConsumeEvent(value = "blocking-consumer", blocking = true)
void consumeBlocking(String message) {
// Something blocking
}
When using @Blocking, it ignores the value of the blocking attribute of @ConsumeEvent.
Reply to events
The return value of a method annotated with @ConsumeEvent is used to respond to the incoming message.
For instance, in the following snippet, the returned String is the response.
@ConsumeEvent("greeting")
public String consume(String name) {
return name.toUpperCase();
}
You can also return a Uni<T> or a CompletionStage<T> to handle asynchronous reply:
@ConsumeEvent("greeting")
public Uni<String> consume2(String name) {
return Uni.createFrom().item(() -> name.toUpperCase()).emitOn(executor);
}
|
You can inject an
|
Implement fire-and-forget interactions
You don’t have to reply to received messages.
Typically, for a fire and forget interaction, the messages are consumed, and the sender does not need to know about it.
To implement this pattern, your consumer method returns void.
@ConsumeEvent("greeting")
public void consume(String event) {
// Do something with the event
}
Consume messages (instead of events)
Unlike the previous example using the payloads directly, you can also use Message directly:
@ConsumeEvent("greeting")
public void consume(Message<String> msg) {
System.out.println(msg.address());
System.out.println(msg.body());
}
Handle failures
If a method annotated with @ConsumeEvent throws an exception, then:
-
if a reply handler is set, then the failure is propagated back to the sender via an
io.vertx.core.eventbus.ReplyExceptionwith codeConsumeEvent#FAILURE_CODEand the exception message, -
if no reply handler is set, then the exception is rethrown (and wrapped in a
RuntimeExceptionif necessary) and can be handled by the default exception handler, i.e.io.vertx.core.Vertx#exceptionHandler().
Send messages
Sending and publishing messages use the Vert.x event bus:
package org.acme.vertx;
import io.smallrye.mutiny.Uni;
import io.vertx.mutiny.core.eventbus.EventBus;
import io.vertx.mutiny.core.eventbus.Message;
import jakarta.inject.Inject;
import jakarta.ws.rs.GET;
import jakarta.ws.rs.Path;
import jakarta.ws.rs.Produces;
import jakarta.ws.rs.core.MediaType;
@Path("/async")
public class EventResource {
@Inject
EventBus bus; (1)
@GET
@Produces(MediaType.TEXT_PLAIN)
@Path("{name}")
public Uni<String> greeting(String name) {
return bus.<String>request("greeting", name) (2)
.onItem().transform(Message::body);
}
}
| 1 | Inject the Event bus |
| 2 | Send a message to the address greeting. Message payload is name |
The EventBus object provides methods to:
-
senda message to a specific address - one single consumer receives the message. -
publisha message to a specific address - all consumers receive the messages. -
requesta message and expect a reply
// Case 1
bus.send("greeting", name)
// Case 2
bus.publish("greeting", name)
// Case 3
Uni<String> response = bus.<String>request("address", "hello, how are you?")
.onItem().transform(Message::body);
Process events on virtual threads
Methods annotated with @ConsumeEvent can also be annotated with @RunOnVirtualThread.
In this case, the method is invoked on a virtual thread.
Each event is invoked on a different virtual thread.
To use this feature, make sure:
-
Your Java runtime supports virtual threads.
-
Your method uses a blocking signature.
The second point means only methods returning an object or void can use @RunOnVirtualThread.
Methods returning a Uni or a CompletionStage cannot run on virtual threads.
Read the virtual thread guide for more details.
Use codecs
The Vert.x Event Bus uses codecs to serialize and deserialize message objects.
Quarkus provides a default codec for local delivery.
This codec is automatically used for return types and message body parameters of methods annotated with @ConsumeEvent.
So that you can exchange the message objects as follows:
@GET
@Produces(MediaType.TEXT_PLAIN)
@Path("{name}")
public Uni<String> greeting(String name) {
return bus.<String>request("greeting", new MyName(name))
.onItem().transform(Message::body);
}
@ConsumeEvent(value = "greeting")
Uni<String> greeting(MyName name) {
return Uni.createFrom().item(() -> "Hello " + name.getName());
}
If you want to use a specific codec, you need to set it on both ends explicitly:
@GET
@Produces(MediaType.TEXT_PLAIN)
@Path("{name}")
public Uni<String> greeting(String name) {
return bus.<String>request("greeting", name,
new DeliveryOptions().setCodecName(MyNameCodec.class.getName())) (1)
.onItem().transform(Message::body);
}
@ConsumeEvent(value = "greeting", codec = MyNameCodec.class) (2)
Uni<String> greeting(MyName name) {
return Uni.createFrom().item(() -> "Hello "+name.getName());
}
| 1 | Set the name of the codec to use to send the message |
| 2 | Set the codec to use to receive the message |
Combine HTTP and the Event Bus
Let’s revisit a greeting HTTP endpoint and use asynchronous message passing to delegate the call to a separated bean. It uses the request/reply dispatching mechanism. Instead of implementing the business logic inside the Jakarta REST endpoint, we are sending a message. Another bean consumes this message, and the response is sent using the reply mechanism.
In your HTTP endpoint class, inject the event bus and uses the request method to send a message to the event bus and expect a response:
package org.acme.vertx;
import io.smallrye.mutiny.Uni;
import io.vertx.mutiny.core.eventbus.EventBus;
import io.vertx.mutiny.core.eventbus.Message;
import jakarta.inject.Inject;
import jakarta.ws.rs.GET;
import jakarta.ws.rs.Path;
import jakarta.ws.rs.Produces;
import jakarta.ws.rs.core.MediaType;
@Path("/bus")
public class EventResource {
@Inject
EventBus bus;
@GET
@Produces(MediaType.TEXT_PLAIN)
@Path("{name}")
public Uni<String> greeting(String name) {
return bus.<String>request("greeting", name) (1)
.onItem().transform(Message::body); (2)
}
}
| 1 | send the name to the greeting address and request a response |
| 2 | when we get the response, extract the body and send it to the user |
the HTTP method returns a Uni.
If you are using Quarkus REST, Uni support is built-in.
If you are using classic RESTEasy, you need to add the quarkus resteasy-mutiny extension to your project.
|
We need a consumer listening on the greeting address.
This consumer can be in the same class or another bean such as:
package org.acme.vertx;
import io.quarkus.vertx.ConsumeEvent;
import jakarta.enterprise.context.ApplicationScoped;
@ApplicationScoped
public class GreetingService {
@ConsumeEvent("greeting")
public String greeting(String name) {
return "Hello " + name;
}
}
This bean receives the name and returns the greeting message.
With this in place, every HTTP request on /bus/quarkus sends a message to the event bus, waits for a reply, and when this one arrives, writes the HTTP response:
Hello Quarkus
To better understand, let’s detail how the HTTP request/response has been handled:
-
The request is received by the
greetingmethod -
a message containing the name is sent to the event bus
-
Another bean receives this message and computes the response
-
This response is sent back using the reply mechanism
-
Once the reply is received by the sender, the content is written to the HTTP response
Bidirectional communication with browsers by using SockJS
The SockJS bridge provided by Vert.x allows browser applications and Quarkus applications to communicate using the event bus. It connects both sides. So, both sides can send messages received on the other side. It supports the three delivery mechanisms.
SockJS negotiates the communication channel between the Quarkus application and the browser. If WebSockets are supported, it uses them; otherwise, it degrades to SSE, long polling, etc.
So use SockJS, you need to configure the bridge, especially the addresses that will be used to communicate:
package org.acme;
import io.vertx.core.Vertx;
import io.vertx.ext.bridge.PermittedOptions;
import io.vertx.ext.web.Router;
import io.vertx.ext.web.handler.sockjs.SockJSBridgeOptions;
import io.vertx.ext.web.handler.sockjs.SockJSHandler;
import jakarta.enterprise.context.ApplicationScoped;
import jakarta.enterprise.event.Observes;
import jakarta.inject.Inject;
import java.util.concurrent.atomic.AtomicInteger;
@ApplicationScoped
public class SockJsExample {
@Inject
Vertx vertx;
public void init(@Observes Router router) {
SockJSHandler sockJSHandler = SockJSHandler.create(vertx);
Router bridge = sockJSHandler.bridge(new SockJSBridgeOptions()
.addOutboundPermitted(new PermittedOptions().setAddress("ticks")));
router.route("/eventbus/*").subRouter(bridge);
AtomicInteger counter = new AtomicInteger();
vertx.setPeriodic(1000,
ignored -> vertx.eventBus().publish("ticks", counter.getAndIncrement()));
}
}
This code configures the SockJS bridge to send all the messages targeting the ticks address to the connected browsers.
More detailed explanations about the configuration can be found on the Vert.x SockJS Bridge documentation.
The browser must use the vertx-eventbus JavaScript library to consume the message:
<!doctype html>
<html>
<head>
<meta charset="utf-8"/>
<title>SockJS example - Quarkus</title>
<script src="https://code.jquery.com/jquery-3.3.1.min.js"
integrity="sha256-FgpCb/KJQlLNfOu91ta32o/NMZxltwRo8QtmkMRdAu8=" crossorigin="anonymous"></script>
<script type="application/javascript" src="https://cdn.jsdelivr.net/sockjs/0.3.4/sockjs.min.js"></script>
<script src="https://cdn.jsdelivr.net/npm/vertx3-eventbus-client@3.8.5/vertx-eventbus.min.js"></script>
</head>
<body>
<h1>SockJS Examples</h1>
<p><strong>Last Tick:</strong> <span id="tick"></span></p>
</body>
<script>
var eb = new EventBus('/eventbus');
eb.onopen = function () {
eb.registerHandler('ticks', function (error, message) {
$("#tick").html(message.body);
});
}
</script>
</html>
Native transports
Quarkus supports Netty’s native transports (epoll, kqueue, io_uring) for improved I/O performance on specific platforms. See the Native Transport Reference for configuration, dependencies, and platform-specific details.
Use a Vert.x context-aware scheduler
Some Mutiny operators need to schedule work on an executor thread pool.
A good example is .onItem().delayIt().by(Duration.ofMillis(10)) as it needs such an executor to delay emissions.
The default executor is returned by io.smallrye.mutiny.infrastructure.Infrastructure and it is already configured and managed by Quarkus.
That being said, there are cases where you need to make sure that an operation is run on a Vert.x (duplicated) context and not just on any random thread.
The io.smallrye.mutiny.vertx.core.ContextAwareScheduler interface offers an API to obtain context-aware schedulers.
Such a scheduler is configured with:
-
a delegate
ScheduledExecutorServiceof your choice (hint: you can reuseInfrastructure.getDefaultWorkerPool()), and -
a context fetching strategy among:
-
an explicit
Context, or -
calling
Vertx::getOrCreateContext()either on the current thread or later when the scheduling request happens, or -
calling
Vertx::currentContext(), which fails if the current thread is not a Vert.x thread.
-
Here is a sample where ContextAwareScheduler is used:
class MyVerticle extends AbstractVerticle {
@Override
public Uni<Void> asyncStart() {
vertx.getOrCreateContext().put("foo", "bar");
var delegate = Infrastructure.getDefaultWorkerPool();
var scheduler = ContextAwareScheduler.delegatingTo(delegate)
.withCurrentContext();
return Uni.createFrom().voidItem()
.onItem().delayIt().onExecutor(scheduler).by(Duration.ofMillis(10))
.onItem().invoke(() -> {
// Prints "bar"
var ctx = vertx.getOrCreateContext();
System.out.println(ctx.get("foo"));
});
}
}
In this example a scheduler is created by capturing the context of the Vert.x event-loop that calls asyncStart().
The delayIt operator uses that scheduler, and we can check that the context that we get in invoke is a Vert.x duplicated context where the data for key "foo" has been propagated.
Use a Unix Domain Socket
Listening on a Unix Domain Socket allows us to dispense with the overhead of TCP if the connection to the quarkus service is established from the same host. This can happen if access to the service goes through a proxy which is often the case if you’re setting up a service mesh with a proxy like Envoy.
Unix Domain Sockets are supported in both JVM and native modes.
| Unix domain sockets require a native transport dependency. See the Native Transport Reference for details. |
quarkus.http.domain-socket=/var/run/io.quarkus.app.socket quarkus.http.domain-socket-enabled=true
By itself this will not disable the tcp socket which by default will open on
0.0.0.0:8080. It can be explicitly disabled:
quarkus.http.host-enabled=false
These properties can be set through Java’s -D command line parameter or
on application.properties.
| Make sure your application has the right permissions to write to the socket. |
Deploy on read-only environments
In environments with read only file systems you may receive errors of the form:
java.lang.IllegalStateException: Failed to create cache dir
Assuming /tmp/ is writable this can be fixed by setting the vertx.cacheDirBase property to point to a directory in /tmp/ for instance in Kubernetes by creating an environment variable JAVA_OPTS with the value -Dvertx.cacheDirBase=/tmp/vertx, or setting the quarkus.vertx.cache-directory property in application.properties:
quarkus.vertx.cache-directory=/tmp/vertx
Customize the Vert.x configuration
The configuration of the managed Vert.x instance can be provided using the application.properties file, but also using special beans.
CDI beans exposing the io.quarkus.vertx.VertxOptionsCustomizer interface can be used to customize the Vert.x configuration.
For example, the following customizer change the tmp base directory:
@ApplicationScoped
public class MyCustomizer implements VertxOptionsCustomizer {
@Override
public void accept(VertxOptions options) {
options.setFileSystemOptions(new FileSystemOptions().setFileCacheDir("target"));
}
}
The customizer beans received the VertxOptions (coming from the application configuration), and can modify them.
Brotli4J and cross-platform support
Brotli4J is a native library that provides support for the Brotli compression algorithm. By default, Quarkus includes the Brotli native library matching the platform you are running on. But sometimes, you need to include the native library for a different platform.
In this case, you need to explicitly add a dependency to your project.
For example, if you need to include the native library for linux-aarch64, you can add the following dependency:
<dependency>
<groupId>com.aayushatharva.brotli4j</groupId>
<artifactId>native-linux-aarch64</artifactId>
</dependency>
This will include the native library for linux-aarch64 in your project, in addition to the one matching your machine.
Here is the list of available brotli4j artifacts for the different platforms:
-
native-linux-x86_64 -
native-linux-s390x -
native-linux-ppc64le -
native-linux-aarch64 -
native-linux-armv7 -
native-linux-riscv64 -
native-windows-x86_64 -
native-windows-aarch64 -
native-macos-x86_64 -
native-macos-aarch64