Getting Started with Go for Web Services (Part 3): Testing, Deployment, Scaling

Part 3 closes the series with quality gates, release strategy, and scale patterns.

Testing pyramid for Go APIs

1. Unit tests for service logic.
2. Integration tests for DB/repository behavior.
3. Contract tests for external APIs.
4. End-to-end smoke tests after deploy.

func TestCreateUser(t *testing.T) {
    store := newFakeStore()
    svc := NewUserService(store)

    got, err := svc.Create(context.Background(), CreateUserInput{Email: "a@b.com"})
    if err != nil {
        t.Fatalf("unexpected err: %v", err)
    }
    if got.Email != "a@b.com" {
        t.Fatalf("unexpected email: %s", got.Email)
    }
}

Deployment safety checks

Deploy pipeline should block on:

• go test ./...
• static analysis (go vet, lint)
• migration checks
• container scan
• smoke test in staging

Zero-downtime release model

Horizontal scale and bottlenecks

Scale sequence to follow:

1. Fix hot queries and N+1 patterns.
2. Add caching where read-heavy.
3. Queue background work.
4. Then add replicas.

When Python still belongs in the stack

In mature systems, Python can remain useful for:

• offline analytics and ML pipelines
• experimentation workloads
• rule iteration before Go hardening

Keep serving path latency-critical code in Go, and treat Python as adjacent tooling unless requirements dictate otherwise.

Final production checklist

• Timeouts everywhere
• Idempotency keys for unsafe retries
• Backpressure and rate limits
• Budgeted SLOs with alerting
• Disaster recovery runbook