Expected Answer (Go 1.26.5)

context.Context propagates cancellation, deadlines, and request-scoped metadata across API boundaries. A function should accept it as its first parameter and regularly select on ctx.Done() when performing work that can be stopped. Derive contexts with WithCancel or WithTimeout, and call the returned cancel function even when the timeout will eventually fire.

Context values are for request-scoped cross-cutting metadata such as a trace ID, not optional function parameters or general application state.

package main

import (
	"context"
	"fmt"
	"time"
)

func waitForWork(ctx context.Context) error {
	select {
	case <-time.After(time.Second):
		return nil
	case <-ctx.Done():
		return ctx.Err()
	}
}

func main() {
	ctx, cancel := context.WithTimeout(context.Background(), 10*time.Millisecond)
	defer cancel()

	fmt.Println(waitForWork(ctx)) // context deadline exceeded
}

Why It Matters

Cancellation prevents abandoned HTTP requests and background jobs from continuing to consume connections, CPU, and memory. Correct propagation improves service latency and keeps shutdown behavior predictable.

Common Mistakes

• Creating a new context.Background() inside a request path: This discards the caller’s deadline and cancellation signal.
• Forgetting to call cancel: Resources associated with a derived context can remain until its parent ends or its timeout fires.
• Using context values for required dependencies: It hides the function’s real API and weakens type safety.

Follow-up Questions

• What does ctx.Err() return after a timeout? (Answer: context.DeadlineExceeded.)
• How should an HTTP handler obtain its context? (Answer: From r.Context(), then pass it to downstream calls.)