ℹ️ dropwizard-health has been promoted to a core module and is part of the Dropwizard main project since Dropwizard 2.1.0.

dropwizard-health

Provides a health check implementation that performs ongoing monitoring of an application's dependencies and includes an endpoint that can be called by a load balancer to determine if the application is healthy and thus able to receive traffic.

Dropwizard Version Support Matrix

Usage

Add dependency on library.

<dependency>
  <groupId>io.dropwizard.modules</groupId>
  <artifactId>dropwizard-health</artifactId>
  <version>${dropwizard-health.version}</version>
</dependency>

Health Checks

In your application's Configuration class, add a HealthConfiguration object:

public class ExampleConfiguration extends Configuration {
    ...

    @Valid
    @NotNull
    @JsonProperty("health")
    private HealthConfiguration healthConfiguration = new HealthConfiguration();

    public HealthConfiguration getHealthConfiguration() {
        return healthConfiguration;
    }

    public void setHealthConfiguration(final HealthConfiguration healthConfiguration) {
        this.healthConfiguration = healthConfiguration;
    }
}

Add a HealthCheckBundle to the Bootstrap object in your initialize method:

bootstrap.addBundle(new HealthCheckBundle<ExampleConfiguration>() {
    @Override
    protected HealthConfiguration getHealthConfiguration(final ExampleConfiguration configuration) {
        return configuration.getHealthConfiguration();
    }
});

Configure health checks for any dependencies your application has, such as any databases, caches, queues, etc. Dropwizard modules that support connecting to a dependency will often register a health check automatically.

Define the following health check configurations in your config.yml file:

health:
  delayedShutdownHandlerEnabled: true
  shutdownWaitPeriod: 30s
  healthChecks:
    - name: UserDatabase
      critical: true
    - name: UserNotificationTopic
      critical: false
      schedule:
        checkInterval: 2500ms
        downtimeInterval: 10s
        failureAttempts: 2
        successAttempts: 1
    - name: UserCache
      critical: false

Configuration Reference

Health Configuration

Health Check Configuration

Schedule

Default Health Servlet

Application Status

There are two types of status that are supported: Alive and Ready

alive

An alive status indicates the application is operating normally and does not need to be restarted to recover from a stuck state.

Long-running applications can eventually reach a broken state and cannot recover except by being restarted (e.g. deadlocked threads).

ready

A ready status indicates the application is ready to serve traffic.

Applications can temporarily be unable to serve traffic due to a variety of reasons, for example, an application might need to build/compute large caches during startup or can critically depend on an external service.

Query Application Status

https://:/health-check?type= (replace <type> with ready or alive; defaults to ready)

Healthy

$ curl -v https://<hostname>:<port>/health-check?type=ready
> GET /health-check?type=ready HTTP/1.1
...
>
< HTTP/1.1 200 OK
< Content-Type: application/json
< Cache-Control: no-store
...
<
{"status": "healthy"}

Not Healthy

$ curl -v https://<hostname>:<port>/health-check?type=ready
> GET /health-check?type=ready HTTP/1.1
...
>
< HTTP/1.1 503 Service Unavailable
< Content-Type: application/json
< Cache-Control: no-store
...
<
{"status": "unhealthy"}

HTTP and TCP Health Checks

Should your service have any dependencies that it needs to perform health checks against that expose either an HTTP or TCP health check interface, you can use the HttpHealthCheck or TcpHealthCheck classes to do so easily.

Usage

You will need to register your health check(s) in your Application class run() method.

HTTP

@Override
public void run(final AppConfiguration configuration, final Environment environment) {
    ...
    environment.healthChecks().register("http-service-dependency", new HttpHealthCheck("http://some-http-dependency.com:8080/health-check"));
}

TCP

@Override
public void run(final AppConfiguration configuration, final Environment environment) {
    ...
    environment.healthChecks().register("tcp-service-dependency", new TcpHealthCheck("some-tcp-dependency.com", 443));
}

Composite Health Checks

You might find you need a health check that is a composite of more than one other health check. For instance, consider the case where you have a database and a cache, and if only one of those two are unhealthy, your service can still fulfill a subset of functionality, and thus should not necessarily be marked down.

Let's say that you have a database health check registered under the name UserDatabase and a cache health check under the name UserCache. Below, see an example of how you might create a health check that is a composite of these two checks. Note that there does not exist a CompositeHealthCheck class currently, but it might be a nice addition to this library.

Example Composite Health Check

private void registerCompositeHealthCheck(final HealthCheckRegistry healthChecks) {
    final HealthCheck databaseCheck = healthChecks.getHealthCheck("UserDatabase");
    final HealthCheck cacheCheck = healthChecks.getHealthCheck("UserCache");

    final UserHealthCheck userHealthCheck = new UserHealthCheck(databaseCheck, cacheCheck);
    healthChecks.register("UserComposite", userHealthCheck);
}

Example Composite Health Check YAML

health:
  delayedShutdownHandlerEnabled: true
  shutdownWaitPeriod: 30s
  healthChecks:
    - name: UserComposite
      critical: true

Future improvements:

• Hooks for health status change events.
• The ability to expose health check data to other modules (for instance, an Admin page may want to show historical health check results).
• More out-of-the-box generally useful health checks implementations, like file-system health checks.