gRPC Interview Questions

1. What is gRPC and what problems does it solve?

2. What is the difference between gRPC and REST?

3. What are Protocol Buffers (protobuf) and how do they work?

4. What is an .proto file and how do you define a gRPC service?

5. What is the role of HTTP/2 in gRPC and why does it matter?

6. What are the four types of gRPC communication (unary, server streaming, client streaming, bidirectional)?

7. What is unary RPC in gRPC and when do you use it?

8. What is server-side streaming in gRPC?

9. What is client-side streaming in gRPC?

10. What is bidirectional streaming in gRPC and how does it work?

11. How does gRPC handle serialization compared to JSON?

12. What are gRPC stubs and how does code generation work?

13. How does error handling and status codes work in gRPC?

14. What are gRPC deadlines and timeouts and why are they important?

15. How does gRPC handle authentication and TLS?

16. What are gRPC interceptors and how are they used?

17. What is gRPC metadata and how do you use it?

18. How does gRPC handle backward and forward compatibility in protobuf schemas?

19. What is the difference between proto2 and proto3?

20. How does gRPC load balancing work?

21. What is gRPC channel and how does connection management work?

22. What is gRPC-Web and how does it enable browser clients?

23. How does gRPC support health checking and service discovery?

24. What is server reflection in gRPC and why is it useful?

25. How do you handle streaming flow control and backpressure in gRPC?

26. What are the performance advantages of gRPC over REST?

27. How do you test and debug a gRPC service?

28. When should you choose gRPC over REST or GraphQL?

29. How does gRPC handle retries and deadline propagation?

30. What are common gRPC best practices and pitfalls?

31. Why is reusing a deleted protobuf field number dangerous, and how do you prevent it?

32. Which protobuf schema changes are wire-compatible, and which quietly break peers?

33. How do you propagate and enforce gRPC deadlines across a multi-hop call graph?

34. How do you return rich, structured errors from a gRPC service beyond a status code?

35. Why does gRPC traffic become unbalanced behind a load balancer, and how do you fix it?

36. How do you configure safe gRPC retries, and when is hedging appropriate instead?

37. How does HTTP/2 flow control create backpressure in gRPC streams, and how do you tune it?

38. What breaks when gRPC crosses proxies and browsers, and how does grpc-web work around it?

39. How do you set up and operate mutual TLS for gRPC services, including rotation?

40. How do you order and design a gRPC interceptor chain for auth, observability and resilience?

41. Why do idle gRPC connections silently die, and how do you configure keepalive correctly?

42. How do you drain and shut down a gRPC server gracefully during a rolling deploy?

43. How do you handle gRPC message size limits and RESOURCE_EXHAUSTED on large payloads?

44. When should you enable gRPC compression, and what are the operational risks?

45. How do you propagate tracing context and collect useful metrics across gRPC services?

46. How do you write deterministic tests for gRPC handlers, streaming and error paths?

47. How do you find and fix leaked gRPC streams, contexts and worker goroutines?

48. How do you protect a gRPC server from overload with concurrency limits and load shedding?

49. How do you debug a failing gRPC call in production when the error is just UNAVAILABLE?

50. How do you expose an existing gRPC service to REST clients without forking the implementation?

51. What goes wrong when gRPC clients and servers are written in different languages?

52. How do you manage .proto files and generated code across many teams and repositories?

53. How do you add values to a protobuf enum without breaking existing clients?

54. What happens on the server when a gRPC client cancels a call, and how should your handler react?

55. How do you make mutating gRPC methods idempotent so that retries are safe?

56. How do you choose a gRPC name resolver and client-side load balancing policy?

57. How do you design a long-lived server-streaming RPC that survives disconnects and deploys?

58. How do you enforce per-method authorization in a gRPC service?

59. What are the gRPC channel connectivity states, and how do you use them in production?

60. How do you model field presence in proto3 using optional, oneof and wrappers?

61. How do you version a gRPC API when a change cannot be made backward compatible?