Connect GraphQL Java to Bugsink

When a GraphQL resolver fails, Bugsink can show the original exception and application frame together with the operation and data fetcher which were being executed.

This guide adds Bugsink to an existing GraphQL Java service. You’ll initialize the SDK, attach its GraphQL instrumentation, and send a named query which fails inside a product data fetcher while the service returns its normal GraphQL error response.

To follow along, create a Bugsink project and copy its DSN. The quickstart guide shows where to find it.

Install the GraphQL integration

Add the open-source Sentry integration which matches your GraphQL Java line. GraphQL Java 22 and newer use sentry-graphql-22; GraphQL Java 17 through 21 use sentry-graphql instead. Bugsink accepts the SDK’s error events directly:

<dependency>
  <groupId>io.sentry</groupId>
  <artifactId>sentry-graphql-22</artifactId>
  <version>${sentry.version}</version>
</dependency>

Initialize the SDK

Initialize the SDK once when the service starts. Reading the DSN from the environment keeps it out of source control, while addInAppInclude marks your own package as application code in the stack trace:

Sentry.init(options -> {
    options.setDsn(System.getenv("SENTRY_DSN"));
    options.addInAppInclude("com.example.inventory");
});

Set the DSN in the environment where the GraphQL service runs:

export SENTRY_DSN=https://key@bugsink.example/1

Instrument GraphQL execution

The exception handler retains failures raised by data fetchers, while the instrumentation reports them after GraphQL has assembled the result. Together they preserve GraphQL’s normal errors response and add the named operation, field path, resolver details, and unhandled mechanism to the Bugsink event:

return GraphQL.newGraphQL(executableSchema)
        .defaultDataFetcherExceptionHandler(new SentryGenericDataFetcherExceptionHandler(
                new SimpleDataFetcherExceptionHandler()))
        .instrumentation(new SentryInstrumentation(
                NoOpSubscriptionHandler.getInstance(), true))
        .build();

NoOpSubscriptionHandler is the normal choice for a standalone query and mutation service. Applications built on Spring GraphQL or Netflix DGS use the subscription handler supplied for that framework.

Send a failing query

Start the service and send a named query for a product which does not exist:

mvn package
java -jar target/inventory-graphql-1.0.0.jar
:::bash
curl http://127.0.0.1:8080/graphql \
  -H 'content-type: application/json' \
  --data '{"query":"query FindProduct($sku: ID!) { product(sku: $sku) { sku name } }","variables":{"sku":"graphql-missing"}}'

GraphQL returns HTTP 200 with its usual errors array and a null data result. The integration reports the underlying data-fetcher exception without changing that application response.

See the query failure in Bugsink

Open your Bugsink project and select the new ProductNotFound issue. The event shows the application frame, while its breadcrumbs identify the FindProduct query and /product data fetcher.

Bugsink preview

That’s it: unexpected data-fetcher failures will now reach Bugsink with the GraphQL execution details needed to identify the operation and field.

For handled exceptions, messages, Java logging integrations, tags, structured context, and breadcrumbs, continue with the Java guide.