Golang / GoLang System Architecture and Testing Interview Questions

How do you implement event-driven communication between Go microservices using message queues?

Synchronous RPC (REST/gRPC) creates tight coupling — if service B is down, service A fails. Message queues (Kafka, NATS, RabbitMQ) decouple producers from consumers: A publishes an event and continues; B processes it when ready. This improves resilience and enables fan-out.

// NATS JetStream producer
import "github.com/nats-io/nats.go"

type EventPublisher struct {
    js nats.JetStreamContext
}

type UserCreatedEvent struct {
    UserID    int       `json:"user_id"`
    Email     string    `json:"email"`
    CreatedAt time.Time `json:"created_at"`
}

func (p *EventPublisher) PublishUserCreated(ctx context.Context, e UserCreatedEvent) error {
    data, err := json.Marshal(e)
    if err != nil {
        return fmt.Errorf("marshal event: %w", err)
    }
    msg := &nats.Msg{
        Subject: "users.created",
        Data:    data,
        Header:  make(nats.Header),
    }
    // Propagate trace context in headers
    otel.GetTextMapPropagator().Inject(ctx, propagation.HeaderCarrier(msg.Header))

    _, err = p.js.PublishMsg(msg)
    return err
}

// Consumer with idempotency
type EmailConsumer struct {
    email  EmailSender
    dedup  *DeduplicationCache // prevent double-sending
}

func (c *EmailConsumer) HandleUserCreated(msg *nats.Msg) {
    var event UserCreatedEvent
    if err := json.Unmarshal(msg.Data, &event); err != nil {
        msg.Nak() // negative ack: requeue
        return
    }

    // Idempotency check â process each event exactly once
    key := fmt.Sprintf("email:welcome:%d", event.UserID)
    if c.dedup.Has(key) {
        msg.Ack() // already processed â ack without re-sending
        return
    }

    if err := c.email.SendWelcome(event.Email); err != nil {
        msg.Nak()
        return
    }
    c.dedup.Set(key)
    msg.Ack()
}

Exactly-once delivery: message queues provide at-least-once delivery (messages may be re-delivered on failure). Consumers must be idempotent — processing the same message twice produces the same result. Use a deduplication cache (Redis SET NX with TTL) keyed on the event ID.

Take quiz

What is the key advantage of event-driven (message queue) communication over synchronous gRPC for microservices?Message queues are always faster than gRPC

✗ Try again.

It decouples producer and consumer — the producer continues if the consumer is down, and multiple consumers can independently process the same event

✓ Correct! Well done.

Message queues provide stronger type safety than gRPC

✗ Try again.

Events are automatically encrypted by the message broker

✗ Try again.

Why must event consumers be idempotent in an at-least-once message queue?The broker guarantees message ordering only for idempotent consumers

✗ Try again.

Messages may be redelivered on consumer failure — processing the same event twice must produce the same result to prevent double-sending emails, double-charging, etc.

✓ Correct! Well done.

Idempotency is required by the AMQP protocol

✗ Try again.

Non-idempotent consumers cause message queue deadlock

✗ Try again.

More Related questions...

Compare REST/JSON with gRPC/Protocol Buffers. When would you choose gRPC for a Go microservice? How do you implement a gRPC server in Go, including error handling and interceptors? How do you build a production-ready gRPC client in Go with connection reuse and resilience? What microservice design patterns are most important to understand for Go interviews? How do you implement distributed tracing and observability in a Go microservice system? How do you implement event-driven communication between Go microservices using message queues? How do you manage database connections and sharding in a high-scale Go service? What caching strategies do you use in Go microservices and how do you prevent cache stampede? What are table-driven tests in Go and why are they the standard testing pattern? How do you write Go benchmarks and what does -benchmem tell you? How do you find and fix memory allocation hotspots in a Go service using profiling? How do you structure integration tests in Go that require real databases or external services? Explain the difference between mocks, stubs, and fakes in Go testing. When do you use each? How does Go's built-in fuzzing work and when should you use property-based testing? How do you test concurrent Go code correctly — including data races and timing issues? How do you decide where to draw service boundaries when decomposing a Go monolith into microservices? How do you version gRPC APIs in Go without breaking existing clients? How do you write unit and integration tests for gRPC services in Go? How do you load test a Go microservice and interpret the results? How does service discovery and client-side load balancing work in a Go microservice system? How do you design a consistent error model across multiple Go microservices? How do you implement the Saga pattern for distributed transactions in Go? What testing.T methods do experienced Go engineers use to write cleaner tests? How do you benchmark concurrent code with testing.B and what insights does it provide? How do you manage dependency injection at scale in a large Go service — wire, dig, or manual? How do you achieve zero-downtime deployments for a Go microservice in Kubernetes? How do generics in Go 1.18+ enable better system design and what are the trade-offs? How do you use test coverage meaningfully in Go — beyond just a percentage? What are the best practices for designing Protocol Buffer schemas in Go microservices? How do you implement safe retries in Go microservices? What are golden file tests in Go and when should you use them? How do you ensure data consistency across Go microservices without distributed transactions? What is the API Gateway pattern and how does it complement Go microservices? What memory leak patterns in Go are not goroutine leaks and how do you detect them? How do CQRS and event sourcing apply to Go microservice architecture? What is chaos engineering and how do Go teams apply it to test microservice resilience? What is contract testing and how does it apply to Go microservices? What makes a Go microservice horizontally scalable and what patterns break scaling? How do you implement configuration hot-reloading in a Go service without restart? How do you architect Go services for maximum testability at the package level? How do you implement feature flags and canary deployments in a Go microservice? How do you design a multi-tenant Go microservice? What is mutation testing and how does it evaluate test suite quality beyond coverage? How do you manage the full lifecycle of a Go microservice from startup to shutdown? How do you test Go code that processes streaming data or works with channels? Summarise the key principles for designing scalable Go microservices that senior engineers demonstrate.

Show more question and Answers...

SAP

	Interviews Questions Java Spring Hibernate DevOps Testing API BigData Web Erlang AI Database Integration Cloud Python Scala Tools Golang SAP	About Javapedia.net Javapedia.net is for Java and J2EE developers, technologist and college students who prepare of interview. Also this site includes many practical examples. This site is developed using J2EE technologies by Steve Antony, a senior Developer/lead at one of the logistics based company. Follow Javapedia updates
	contact: javatutorials2016[at]gmail[dot]com
Kindly consider donating for maintaining this website. Thanks.
	Copyright © 2026, javapedia.net, all rights reserved. Privacy Policy · Terms of Service.

Golang / GoLang System Architecture and Testing Interview Questions

How do you implement event-driven communication between Go microservices using message queues?

Comments & Discussions

Recently added questions