questions.md

1. How does the remote client determine when a command's output is fully received from the server, and what techniques can be used to handle partial reads or ensure complete message transmission?

answer here

Loop Until EOF Character is Received

• The client continuously reads from the socket using recv(), checking if the last byte received is 0x04 (EOF).
• If the EOF character is detected, the client knows the full response has been received.

Buffering and Aggregation Since recv() may return partial data, the client should append data to a buffer until the EOF character is found.

2. This week's lecture on TCP explains that it is a reliable stream protocol rather than a message-oriented one. Since TCP does not preserve message boundaries, how should a networked shell protocol define and detect the beginning and end of a command sent over a TCP connection? What challenges arise if this is not handled correctly?

answer here

Use a Terminator Character

• The protocol can use a special delimiter, such as \0(null byte) for commands and 0x04 (EOF) for responses. Include Message Length in the Header
• A fixed-size header can precede each message, specifying the number of bytes in the payload. Use a Structured Format (e.g., JSON, XML)
• While not ideal for simple shell protocols, structured formats like JSON allow parsing with defined message start and end markers. Chanllenges:
• TCP might split a message across multiple recv() calls, requiring the client to reconstruct it.
• TCP might merge multiple commands into one recv() call, requiring the client to parse multiple messages properly.
• If no end delimiter is used, the client may read incomplete commands, leading to unexpected behavior.

3. Describe the general differences between stateful and stateless protocols.

answer here

A stateful protocol maintains session state between client and server across multiple interactions, whereas a stateless protocol treats each request as independent with no retained memory of past interactions.

Stateful Protocols: The server remembers previous interactions. Requires more server-side resources (e.g., memory, session tracking). Examples:

• TCP (tracks sequence numbers, retransmissions).
• SSH (maintains authentication state).
• HTTP with Sessions (stores login cookies).

Stateless Protocols: Each request is independent; no session state is stored. Easier to scale since no session tracking is needed. Examples:

• UDP (each packet is independent).
• HTTP (by default, unless cookies or tokens are used).
• DNS (each lookup request is processed independently).

4. Our lecture this week stated that UDP is "unreliable". If that is the case, why would we ever use it?

answer here

Although UDP is unreliable, which does not guarantee packet delivery, order, or integrity, it is still useful for applications where speed and low latency are more important than reliability. Since there is no connection setup or retransmission, UDP is much faster than TCP, which lower the latency. UDP allows sending packets to multiple recipients simultaneously, which is crucial for Streaming video/audio, Online gaming (real-time updates), IoT devices (sensor updates) DNS uses UDP because queries are small and fast (retrying is faster than using TCP).

5. What interface/abstraction is provided by the operating system to enable applications to use network communications?

answer here

The operating system provides socket APIs that allow applications to communicate over a network using protocols like TCP and UDP.

Example: Berkeley Sockets API (socket.h)

Standard API used in C, C++, Python, etc. for network communication.

Functions:

• socket() → Creates a socket.
• bind() → Binds a socket to an IP/port.
• listen() → Marks a socket for incoming connections.
• accept() → Accepts a new client connection.
• connect() → Connects to a remote server.
• send()/recv() → Sends & receives data.