gRPC RESOURCE_EXHAUSTED with quarkus.grpc.server.use-separate-server=false

[ncabanis]

Describe the bug

When using quarkus.grpc.server.use-separate-server=false sending a gRPC message of 256 KB results in RESOURCE_EXHAUSTED. This does not happen with quarkus.grpc.server.use-separate-server=true.
Furthermore, quarkus.grpc.server.max-inbound-message-size has no effect. Even though the default is 4 MB anyways for Java gRPC.

Expected behavior

The integrated gRPC server should behave the same as the separate gRPC server and the configuration should be the same.

Actual behavior

No response

How to Reproduce?

1. Generate the default starter with gRPC.
2. In application.properties, set quarkus.grpc.server.use-separate-server=false
3. Update the generated test case and execute:

    @Test
    void testHello() {
        var sizeInChars = 256 * 1024;
        helloGrpc.sayHello(
                HelloRequest.newBuilder().setName("a".repeat(sizeInChars)).build()
        ).await().atMost(Duration.ofSeconds(5));
    }

code-with-quarkus.zip

Output of uname -a or ver

No response

Output of java -version

No response

Quarkus version or git rev

No response

Build tool (ie. output of mvnw --version or gradlew --version)

No response

Additional information

No response

[quarkus-bot]

/cc @alesj (grpc), @cescoffier (grpc)

[cescoffier]

This is hitting the max message size configured in the Vert.x gRPC server.
As far as I can see, we cannot configure it (@alesj can you confirm). We definitely should have a configuration property for this.

In the GrpcServerRecorder.java file, line 149, I don't see the GrpcServerOptions passed to the server. We should have some options here, typically to configure the max message size. We do have a property about the max inbound message size that we could reuse (this one comes from gRPC Java), but we would still need to configure the server.

[cescoffier]

BTW, it's the only option of the GRPC server options right now, but more is coming in Vert.x 5.

[cescoffier]

BTW, thanks for your report @ncabanis! I've opened a PR fixing this issue.