Computer Networks: 5-Chapter Comprehensive Practice & 75 Verified MCQs

Question 1

At which layer of the OSI 7-Layer Reference Model do routing algorithms (such as OSPF and BGP) and logical IP addressing operate?

Accepted Answer

B. Network Layer (Layer 3). The Network Layer (Layer 3) is responsible for packet forwarding, logical addressing (IPv4, IPv6), and path determination across heterogeneous networks using routing protocols like OSPF, BGP, and RIP.

Question 2

What packets are exchanged during a standard TCP Three-Way Handshake to establish a reliable connection between a client and server?

Accepted Answer

A. SYN -> SYN-ACK -> ACK. 1) The client sends a SYN packet with an Initial Sequence Number (ISN). 2) The server responds with SYN-ACK, acknowledging the client's ISN and supplying its own ISN. 3) The client sends an ACK packet acknowledging the server's ISN, transitioning the socket into the ESTABLISHED state.

Question 3

Why does a TCP connection enter the 'TIME_WAIT' state on the endpoint that initiates the active close (typically lasting 2MSL)?

Accepted Answer

B. To ensure the final ACK is received by the remote peer and to allow lingering duplicate packets from the connection to dissipate from the network.. TIME_WAIT lasts 2 * Maximum Segment Lifetime (2MSL, typically 1–2 minutes). It ensures that if the final ACK is lost, the closing peer can re-transmit it when the remote peer re-sends its FIN. It also prevents delayed packets from the old connection being misdelivered to a newly opened connection reusing the same 4-tuple (source IP, source port, dest IP, dest port).

Question 4

What is the difference between TCP Flow Control and TCP Congestion Control?

Accepted Answer

A. Flow control protects the receiving host from being overwhelmed; Congestion control protects the intermediate network routers and bandwidth.. Flow Control is an end-to-end mechanism where the receiver advertises its available buffer space (Receive Window 'rwnd') so the sender does not overflow the receiver's memory. Congestion Control manages the sender's rate based on network capacity (Congestion Window 'cwnd') to prevent packet drops at intermediate bottleneck routers.

Question 5

Which protocol is responsible for resolving a known logical IPv4 address into a physical hardware MAC address on a local area network?

Accepted Answer

B. ARP (Address Resolution Protocol). Address Resolution Protocol (ARP) broadcasts an ARP request packet ('Who has IP x.x.x.x? Tell MAC y:y:y:y') on the local Layer 2 broadcast domain (FF:FF:FF:FF:FF:FF). The target host replies with its physical MAC address, which is cached in the sender's ARP table.

Question 6

Given the IPv4 address 192.168.10.50/27, what is the subnet mask and the maximum number of usable host addresses on this subnet?

Accepted Answer

B. Subnet Mask: 255.255.255.224; 30 usable hosts. A /27 prefix has 27 network bits and 5 host bits (32 - 27 = 5). 27 bits = 255.255.255.224 (128 + 64 + 32 = 224). The total number of addresses is 2^5 = 32. Subtracting 2 (network address and broadcast address) leaves exactly 30 usable host IP addresses.

Question 7

Which transport layer protocol header is smaller, connectionless, and lacks native packet retransmission or sequencing mechanisms?

Accepted Answer

B. UDP (User Datagram Protocol). UDP has a compact 8-byte header (Source Port, Destination Port, Length, Checksum) and provides connectionless, best-effort datagram delivery with zero handshake latency, making it ideal for real-time audio/video streaming, DNS queries, and gaming.

Question 8

What is the primary difference between an iterative DNS query and a recursive DNS query?

Accepted Answer

B. In a recursive query, the contacted server assumes responsibility to return the final answer; in an iterative query, the contacted server returns the best referral address it knows.. In a recursive query, the DNS client asks a resolver (e.g. ISP or 8.8.8.8) to perform all the lookups on its behalf until finding the answer. In an iterative query, the DNS server responds with a referral to the next authoritative server (e.g., Root -> TLD -> Authoritative) without resolving the full chain itself.

Question 9

What major optimization was introduced in HTTP/2 compared to HTTP/1.1 to eliminate head-of-line blocking at the application layer?

Accepted Answer

B. Binary framing and true multiplexing of multiple independent request/response streams over a single TCP connection.. HTTP/1.1 suffered from application-level head-of-line blocking because requests over a single TCP connection had to be answered in strict sequential order. HTTP/2 splits messages into binary frames tagged with stream IDs, allowing interleaved, concurrent requests and responses over a single shared connection.

Question 10

What is the maximum transmission unit (MTU) size of standard Ethernet frames at the Data Link layer?

Accepted Answer

B. 1,500 bytes. Standard Ethernet specifies an MTU of 1,500 bytes (the maximum size of the IP packet payload that can be encapsulated inside an Ethernet frame without fragmentation). Frames larger than 1,500 bytes are known as Jumbo Frames (typically 9,000 bytes) used in data centers.

Question 11

What is the function of the Time-to-Live (TTL) field in an IPv4 header (or Hop Limit in IPv6)?

Accepted Answer

B. It prevents infinite routing loops by decrementing by 1 at every router hop; when TTL reaches 0, the packet is discarded and an ICMP Time Exceeded message is returned.. The TTL field is an 8-bit counter decremented by each router that processes the packet. If routing misconfigurations cause a packet to loop, TTL eventually decrements to 0, whereupon the router discards the packet and sends an ICMP Type 11 (Time Exceeded) message to the source, preventing network saturation.

Question 12

How does Network Address Translation (NAT) / Port Address Translation (PAT) allow hundreds of internal private LAN devices to access the public Internet simultaneously?

Accepted Answer

B. It maps multiple private IP addresses and ephemeral source ports to a single public IP address using unique translated Layer 4 port numbers.. Port Address Translation (PAT / NAT Overload) maintains a translation state table. When an internal host (e.g. 192.168.1.10:54321) initiates an outbound connection, the NAT router rewrites the source IP to its public WAN IP and assigns a unique external port (e.g. 203.0.113.1:61000), routing inbound replies back to the correct internal client.

Question 13

What cryptographic protocol operates at Layer 3 (Network Layer) to provide end-to-end data authentication, integrity, and encryption for VPN tunnels?

Accepted Answer

B. IPsec (Internet Protocol Security). IPsec functions at the Network Layer (Layer 3) using protocols like Authentication Header (AH) and Encapsulating Security Payload (ESP) to secure all IP traffic transparently to upper-layer applications, making it standard for site-to-site VPNs.

Question 14

Which diagnostic utility uses ICMP Echo Request and Echo Reply messages to verify network layer reachability and calculate round-trip time (RTT)?

Accepted Answer

B. ping. The 'ping' utility transmits ICMP Type 8 (Echo Request) packets to a target IP address and listens for ICMP Type 0 (Echo Reply) packets, reporting packet loss percentage and round-trip latency in milliseconds.

Question 15

What is the primary role of the Border Gateway Protocol (BGP) on the global Internet?

Accepted Answer

B. Distributing routing reachability information between distinct Autonomous Systems (AS) across the global Internet backbone.. BGP is the exterior gateway routing protocol (Path-Vector protocol) that binds the Internet together. It exchanges network prefix reachability between independently administered Autonomous Systems (ISPs, large tech enterprises, telecoms) using policy-based routing metrics.

Question 16

What is the primary role of the Data Link Layer (Layer 2) in the OSI Model?

Accepted Answer

B. Node-to-node (hop-to-hop) frame transmission, physical MAC addressing, error detection, and link flow control. The Data Link layer takes network layer datagrams, encapsulates them into frames, and manages error-free hop-to-hop delivery across a single physical link using hardware MAC addresses.

Question 17

What framing technique uses a dedicated flag byte (such as 01111110 in HDLC) and inserts a '0' bit after five consecutive '1' bits in the payload?

Accepted Answer

B. Bit stuffing. Bit stuffing prevents user data containing 01111110 from being misinterpreted as the end-of-frame flag. The sender inserts a '0' whenever five consecutive '1's appear; receiver automatically strips the stuffed '0'.

Question 18

In Ethernet CSMA/CD (Carrier Sense Multiple Access with Collision Detection), what does a station do when a collision is detected during transmission?

Accepted Answer

B. Aborts transmission immediately, transmits a 32-to-48 bit jam signal, and enters Binary Exponential Backoff. Upon detecting collision, the transceiver immediately aborts frame transmission and emits a jam signal to ensure all network nodes detect the collision. It then waits a randomized backoff interval: K * slot_time.

Question 19

What is the minimum frame size in standard 10 Mbps Ethernet (802.3), and what determines this physical limit?

Accepted Answer

B. 64 bytes (512 bits), determined by the round-trip propagation delay across the maximum network diameter (slot time). To guarantee that a sender detects a collision before it finishes transmitting the frame, the transmission time must be >= round-trip propagation time (2 * Tprop). At 10 Mbps over 2.5 km with repeaters, this requires at least 64 bytes.

Question 20

How does a Layer 2 Ethernet Switch forward frames when receiving a frame with an unknown destination MAC address?

Accepted Answer

B. Floods (broadcasts) the frame out of all ports except the port on which it was received. If the destination MAC is not found in the Switch Forwarding Database (CAM table), the switch broadcasts (floods) the frame out of all active ports on that VLAN except the ingress port.

Question 21

What protocol prevents Layer 2 switching loops and broadcast radiation storms in redundant bridged networks?

Accepted Answer

B. Spanning Tree Protocol (STP / IEEE 802.1D). STP elects a Root Bridge and disables redundant physical switch ports (putting them into blocking state) to create a loop-free logical tree topology. If an active link fails, blocked ports are unblocked.

Question 22

What IEEE 802.1Q tag header field is used by network switches to identify Virtual LAN (VLAN) membership on trunk links?

Accepted Answer

A. 12-bit VLAN ID (VID) allowing up to 4094 distinct VLANs. The 802.1Q header inserts a 4-byte tag into the Ethernet frame containing a 12-bit VLAN Identifier (VID), providing 4,096 theoretical VLANs (0 and 4095 are reserved, allowing 1–4094).

Question 23

How does the Address Resolution Protocol (ARP) map a known IPv4 address to a physical MAC address on a local subnet?

Accepted Answer

B. Broadcasts an ARP Request frame (dest MAC FF:FF:FF:FF:FF:FF); the host owning the requested IP replies with a unicast ARP Reply containing its MAC address. ARP Request is broadcast to all hosts on the Layer 2 broadcast domain asking 'Who has IP X? Tell Y'. Only the device holding IP X replies via unicast frame containing its hardware MAC.

Question 24

What is the maximum throughput efficiency of the Pure ALOHA random access protocol?

Accepted Answer

B. 1 / (2e) ≈ 18.4%. In Pure ALOHA, frames can be sent anytime. Vulnerable period is 2 * Tframe. Maximum throughput S = G * e^(-2G) occurs at offered load G = 0.5, yielding S = 1/(2e) ≈ 18.4%.

Question 25

In Sliding Window Flow Control, what is the maximum sender window size for Go-Back-N with N-bit sequence numbers?

Accepted Answer

B. 2^n - 1. In Go-Back-N, the sender window size cannot exceed 2^n - 1. If window size were 2^n, the receiver would be unable to distinguish whether an ACK represents a new frame or a duplicate retransmission.

Question 26

What is the purpose of the 32-bit Cyclic Redundancy Check (CRC-32) field in the Ethernet frame trailer?

Accepted Answer

B. Detect bit errors introduced during transmission over the physical channel via polynomial division. CRC is an error-detecting code based on cyclic polynomials. The sender calculates CRC remainder and appends it to frame. Receiver recalculates; if remainder != 0, the frame is corrupted and discarded.

Question 27

How does CSMA/CA (Carrier Sense Multiple Access with Collision Avoidance) in Wi-Fi (IEEE 802.11) differ from Ethernet's CSMA/CD?

Accepted Answer

A. Wi-Fi devices cannot detect collisions while transmitting because their own signal drowns out incoming signals; therefore, they avoid collisions using DIFS intervals, random backoff, and optional RTS/CTS handshakes. Wireless radios cannot receive while transmitting on the same frequency (half-duplex RF dynamic range issue). Thus, 802.11 avoids collisions proactively using DIFS, exponential backoff, and RTS/CTS (Request to Send / Clear to Send).

Question 28

What is the difference between a Collision Domain and a Broadcast Domain?

Accepted Answer

B. Each port on a Switch is an isolated Collision Domain; an entire Switch (or VLAN) is a single Broadcast Domain, whereas Routers break up Broadcast Domains. Switches eliminate collisions by providing dedicated bandwidth per port (each switch port is its own collision domain). However, broadcast frames (FF:FF:FF:FF:FF:FF) reach all switch ports. Routers block broadcasts, isolating broadcast domains.

Question 29

What is Point-to-Point Protocol over Ethernet (PPPoE) primarily used for?

Accepted Answer

B. Encapsulating PPP frames within Ethernet frames to enable user authentication, encryption, and billing across DSL and broadband ISP connections. PPPoE combines Ethernet's multi-host capability with PPP's point-to-point authentication (PAP/CHAP) and accounting features, widely used by residential broadband and fiber providers.

Question 30

What is ARP Spoofing (ARP Poisoning)?

Accepted Answer

B. An attack where an attacker sends falsified gratuitous ARP messages over a LAN to associate the attacker's MAC address with the IP of the default gateway, intercepting network traffic (Man-in-the-Middle). Because ARP is stateless and unauthenticated, devices accept unsolicited ARP responses. An attacker poisons target ARP caches, causing all traffic destined for the gateway to be routed through the attacker.

Question 31

What is the total number of usable host addresses available in a /26 IPv4 subnet?

Accepted Answer

B. 62 (2^(32 - 26) - 2 = 64 - 2). A /26 mask leaves 32 - 26 = 6 host bits. Total addresses = 2^6 = 64. Subtracting 2 (all-0s network address and all-1s broadcast address) leaves 62 assignable host addresses.

Question 32

What is the purpose of the Time-to-Live (TTL) field in an IPv4 header?

Accepted Answer

B. Prevents packets from circulating endlessly in routing loops; each router decrements TTL by 1 and drops the packet with an ICMP Time Exceeded message when TTL hits 0. TTL (Hop Limit in IPv6) guarantees that packets caught in routing loops are terminated. When TTL reaches 0, the router discards the packet and sends an ICMP Type 11 (Time Exceeded) message to the sender.

Question 33

What are the three RFC 1918 Private IPv4 Address ranges reserved for internal networks?

Accepted Answer

A. 10.0.0.0/8, 172.16.0.0/12, and 192.168.0.0/16. RFC 1918 defines private non-routable IP blocks: Class A: 10.0.0.0 - 10.255.255.255 (/8); Class B: 172.16.0.0 - 172.31.255.255 (/12); Class C: 192.168.0.0 - 192.168.255.255 (/16).

Question 34

How does Network Address Port Translation (NAPT / PAT) allow thousands of private hosts to share a single public IP address?

Accepted Answer

B. By translating private IP + source port combinations into a single public IP with unique dynamically mapped source ports. NAPT (NAT Overload) tracks concurrent connections using 16-bit TCP/UDP port numbers. Up to ~65,000 simultaneous connections can theoretically be multiplexed over a single external public IP address.

Question 35

What fundamental routing algorithm is used by RIP (Routing Information Protocol)?

Accepted Answer

B. Distance-Vector (Bellman-Ford Algorithm) using hop count as metric (max 15 hops). RIP is a Distance-Vector protocol based on Bellman-Ford. Routers broadcast entire routing tables every 30 seconds. A hop count of 16 is treated as infinity (unreachable), preventing infinite loops.

Question 36

Why does Open Shortest Path First (OSPF) converge significantly faster than Distance-Vector protocols like RIP?

Accepted Answer

B. OSPF is a Link-State protocol where routers flood Link-State Advertisements (LSAs), build an identical topological graph of the network, and run Dijkstra's algorithm locally. In OSPF, every router has a complete map of the network topology stored in its Link-State Database (LSDB). When a link status changes, only localized LSAs are flooded, and routers recalculate paths instantaneously via Dijkstra.

Question 37

What protocol is the de facto standard exterior routing protocol that powers inter-domain routing between Autonomous Systems (AS) across the global Internet?

Accepted Answer

B. Border Gateway Protocol (BGP-4). BGP is a Path-Vector protocol that operates across Autonomous Systems. It advertises path vectors (AS-PATH attribute) to prevent inter-AS routing loops and enforces complex corporate peering policies.

Question 38

How does IPv6 eliminate the need for IP packet fragmentation by intermediate routers?

Accepted Answer

B. Intermediate routers drop packets exceeding MTU and return ICMPv6 Packet Too Big, forcing the sending host to perform Path MTU Discovery (PMTUD) and fragment at source. In IPv4, routers fragmented oversized packets in transit, taxing CPU. In IPv6, intermediate routers NEVER fragment packets. If a packet exceeds link MTU, it is dropped and ICMPv6 Packet Too Big instructs the sender to adjust.

Question 39

What is the primary function of the Internet Control Message Protocol (ICMP)?

Accepted Answer

B. Reporting network-layer diagnostics, errors, and operational information (e.g. Destination Unreachable, Time Exceeded, Echo Request/Reply). ICMP (encapsulated directly in IP datagrams, protocol 1) is used by network devices to send error messages and operational queries (such as 'ping' using Echo Request Type 8 and Echo Reply Type 0).

Question 40

What is the purpose of CIDR (Classless Inter-Domain Routing) prefix aggregation (Supernetting)?

Accepted Answer

B. Combining multiple contiguous smaller IP network prefixes into a single consolidated routing advertisement to curtail global routing table explosion. CIDR replaced rigid Class A/B/C addressing. Route aggregation allows a router to summarize e.g. four contiguous /24 blocks into a single /22 route, drastically reducing global BGP routing table size.

Question 41

What is the Solicited-Node Multicast address in IPv6 and what replaces ARP in IPv6 networks?

Accepted Answer

A. Neighbor Discovery Protocol (NDP) using ICMPv6 messages replaces ARP; uses solicited-node multicast (FF02::1:FFxx:xxxx). IPv6 eliminates broadcast addressing entirely. Neighbor Discovery Protocol (NDP) uses ICMPv6 Neighbor Solicitation and Neighbor Advertisement over targeted Solicited-Node multicast addresses to resolve Layer 2 addresses.

Question 42

What is an Autonomous System (AS) in Internet architecture?

Accepted Answer

B. A collection of connected IP routing prefixes under the control of a single administrative entity or ISP presenting a common, unified routing policy to the Internet. An AS is an independently administered network domain (e.g., Google, Comcast, Level3) assigned a globally unique Autonomous System Number (ASN), exchanging routing information via BGP.

Question 43

In IP routing, what does the 'Longest Prefix Match' (LPM) rule dictate?

Accepted Answer

B. When a destination IP matches multiple routing table entries, the router selects the entry with the longest subnet mask (most specific prefix). If a destination matches both 10.0.0.0/8 and 10.1.2.0/24, the router forwards out the interface for 10.1.2.0/24 because /24 is more specific (24 matching leading bits vs 8).

Question 44

What is Anycast routing?

Accepted Answer

B. Assigning the same unicast IP address to multiple geographically distributed servers, where network routers route traffic to the topologically nearest instance via BGP. Anycast allows multiple servers (e.g. Cloudflare DNS 1.1.1.1, Google 8.8.8.8) to share one IP. BGP automatically directs client traffic to the closest server node, providing high availability and DDoS resilience.

Question 45

What does DHCP Option 82 (Relay Agent Information Option) do?

Accepted Answer

B. Allows a DHCP relay agent to insert network location details (such as the switch port or circuit ID where the client is attached) into DHCP requests sent to a remote DHCP server. Option 82 enables ISPs and enterprise networks to allocate IP addresses based on the physical switch port or DSLAM circuit where the subscriber is connected, preventing IP address spoofing.

Question 46

What is the purpose of the 3-Way Handshake in the Transmission Control Protocol (TCP)?

Accepted Answer

B. Establish a reliable full-duplex connection, synchronize Initial Sequence Numbers (ISNs), and agree on TCP buffer socket parameters. The 3-way handshake (SYN -> SYN-ACK -> ACK) establishes synchronized Initial Sequence Numbers (ISNs) in both directions and negotiates TCP options such as MSS (Maximum Segment Size) and Window Scaling.

Question 47

Why does the active closer in a TCP connection transition into the TIME_WAIT state for 2*MSL (Maximum Segment Lifetime)?

Accepted Answer

B. To ensure the final ACK was received by the remote peer (retransmitting ACK if remote re-sends FIN) and to allow stale in-flight duplicate segments to expire in the network. If the final ACK is lost, the other side will retransmit its FIN. The node in TIME_WAIT can resend the ACK. Furthermore, waiting 2*MSL (typically 60-120s) prevents stray packets from corrupting a new connection on the same port.

Question 48

How does TCP distinguish between a segment loss caused by network congestion vs normal out-of-order packet delivery?

Accepted Answer

B. Receiving 3 duplicate ACKs (4 identical ACKs total) triggers Fast Retransmit without waiting for the RTO (Retransmission Timeout) timer to expire. When an out-of-order segment arrives, receiver immediately generates a duplicate ACK. If 3 duplicate ACKs arrive, sender assumes packet loss (Fast Retransmit), retransmitting immediately and entering Fast Recovery.

Question 49

What is the behavior of the TCP Slow Start algorithm upon establishing a new connection?

Accepted Answer

A. Congestion Window (cwnd) increases by 1 MSS for every received ACK, effectively doubling cwnd every Round-Trip Time (exponential growth). Despite its name, Slow Start increases cwnd exponentially: 1 -> 2 -> 4 -> 8 MSS per RTT until reaching the Slow Start Threshold (ssthresh), whereupon it transitions to linear Congestion Avoidance.

Question 50

What does the TCP Reset (RST) flag indicate when received by a host?

Accepted Answer

B. An abnormal, immediate teardown of the connection; sent when a packet arrives for a closed port, or an unrecoverable protocol error occurred. An RST packet is an abrupt connection abort. If a client attempts to connect (SYN) to a port where no listening service is running, the host replies with RST-ACK.

Question 51

What is the primary architectural difference between TCP and UDP?

Accepted Answer

A. TCP is connection-oriented, provides ordered byte-stream delivery, flow control, and error recovery; UDP is connectionless, lightweight, and provides best-effort message datagrams with zero delivery guarantees. TCP is a heavy, stateful protocol guaranteeing order, flow control, and reliability. UDP is a thin 8-byte wrapper providing only port multiplexing and an optional checksum, ideal for real-time VoIP, gaming, and DNS.

Question 52

What is the purpose of the TCP Selective Acknowledgment (SACK) option (RFC 2018)?

Accepted Answer

B. Enables the receiver to inform the sender about non-contiguous blocks of data successfully received, allowing sender to retransmit only missing segments rather than the entire window. Standard cumulative ACKs only report the highest contiguous byte received. SACK communicates discontinuous byte ranges (e.g. bytes 1-1000 and 2000-3000 received), preventing wasteful retransmissions.

Question 53

How does the Nagle Algorithm optimize network transmission, and why is TCP_NODELAY used to disable it?

Accepted Answer

B. Nagle buffers small outgoing packets until previous data is acknowledged or a full MSS is accumulated to prevent tinygram packet storms; TCP_NODELAY disables buffering for latency-sensitive applications (e.g. games, SSH). Nagle prevents 40 bytes of header overhead for a 1-byte keystroke. However, interacting with Delayed ACK can cause a 200ms deadlock. Setting socket option TCP_NODELAY disables Nagle, sending small packets immediately.

Question 54

What is the 'Bandwidth-Delay Product' (BDP) in high-speed networks?

Accepted Answer

B. Bandwidth (bits/sec) multiplied by Round-Trip Time (sec), representing the maximum volume of unacknowledged data 'in flight' required to fully saturate the network pipeline. BDP measures the capacity of the network pipe (BDP = Link Bandwidth * RTT). If the TCP window is smaller than BDP, the sender stalls waiting for ACKs and the link remains underutilized.

Question 55

What congestion control paradigm does Google's BBR (Bottleneck Bandwidth and RTT) algorithm utilize?

Accepted Answer

B. Model-based congestion control that measures delivery rate and round-trip time to operate at optimal bottleneck bandwidth without causing bufferbloat or relying on packet drops. Traditional Reno/Cubic fill router buffers until packets drop, causing bufferbloat and latency. BBR creates an explicit model of the network pipeline, maximizing throughput while minimizing queuing delay.

Question 56

What is a SYN Flood attack and how do SYN Cookies mitigate it?

Accepted Answer

B. An attacker sends high volumes of TCP SYN requests from spoofed IPs without completing the handshake, exhausting server connection backlogs; SYN Cookies encode connection state inside the initial sequence number, eliminating half-open state allocation. In a SYN flood, the server allocates TCB state memory for half-open connections until RAM is exhausted. SYN Cookies calculate ISN = cryptographic_hash(client_ip, port, timestamp, secret), allocating zero RAM until the client returns ACK.

Question 57

What is the difference between TCP Flow Control and TCP Congestion Control?

Accepted Answer

B. Flow Control protects the receiving endpoint buffer from being overwhelmed (managed by advertised Receiver Window: rwnd); Congestion Control protects intermediate network routers and links from collapse (managed by Congestion Window: cwnd). Flow Control is end-to-end (sender does not overrun receiver RAM: effective_window = min(rwnd, cwnd)). Congestion Control is network-wide, preventing intermediate links from saturating.

Question 58

What is the purpose of Explicit Congestion Notification (ECN)?

Accepted Answer

A. Allows intermediate routers to mark bits (CE = Congestion Experienced) in the IP header when queues build up, signaling endpoints to slow down before packets are dropped. ECN uses 2 bits in the IP header (ECT and CE) and 2 flags in the TCP header (ECE and CWR). Routers with RED (Random Early Detection) mark packets rather than dropping them, preventing retransmissions.

Question 59

In TCP, what is the Maximum Segment Size (MSS)?

Accepted Answer

B. The maximum amount of data (payload only, excluding IP and TCP headers) that a host can receive in a single TCP segment, negotiated in the SYN packet. MSS = MTU - (IP header + TCP header). On standard Ethernet with 1500-byte MTU, IPv4 (20 bytes) and TCP (20 bytes) yield an MSS of 1460 bytes.

Question 60

What happens when a TCP sender receives a Zero Window Advertisement (rwnd = 0) from the receiver?

Accepted Answer

B. Sender pauses data transmission and starts a Persist Timer, periodically transmitting 1-byte Zero-Window Probes to trigger window updates from the receiver. When the receiver's buffer fills, it advertises rwnd = 0. If the subsequent window update ACK is lost, both sides would deadlock. The Persist Timer sends 1-byte probe packets to poll for window reopening.

Question 61

What transport protocol does DNS primarily use, and when does it switch to TCP?

Accepted Answer

B. Primarily uses UDP port 53 for fast query/response; falls back to TCP port 53 for zone transfers (AXFR) or when responses exceed 512 bytes (truncated TC bit set). Standard DNS lookups use UDP for speed (no 3-way handshake). If response data exceeds 512 bytes (or EDNS0 buffer size), server sets the TC (Truncation) bit, instructing the client to retry over TCP port 53.

Question 62

What is the primary improvement of HTTP/2 over HTTP/1.1?

Accepted Answer

B. Binary framing, multiplexing multiple concurrent bidirectional streams over a single TCP connection, HPACK header compression, and Server Push. HTTP/1.1 suffered from Head-of-Line (HoL) blocking at the application layer, requiring multiple parallel TCP connections. HTTP/2 multiplexes independent streams over one TCP pipe using binary frames.

Question 63

Why does HTTP/3 abandon TCP in favor of QUIC running over UDP?

Accepted Answer

B. To solve TCP Head-of-Line blocking (where a single dropped packet stalls all multiplexed streams), achieve 0-RTT connection resumption, and provide seamless connection migration across network changes. In HTTP/2, a lost packet stalls all streams in the TCP window. QUIC implements independent per-stream reliability over UDP: dropping a packet on stream A never stalls stream B.

Question 64

What happens during the TLS 1.3 handshake to establish an encrypted HTTPS connection?

Accepted Answer

B. Achieves full cryptographic handshake in 1 RTT (or 0-RTT resumption) using Ephemeral Diffie-Hellman key exchange (ECDHE) for Forward Secrecy. TLS 1.3 dropped legacy ciphers (RSA key exchange, RC4, CBC). Client sends supported key shares in ClientHello; Server replies with chosen share, certificate, and Finished message in 1 RTT.

Computer Networks: 5-Chapter Comprehensive Practice & 75 Verified MCQs

Computer Networks: 5-Chapter Comprehensive Practice & 75 Verified MCQs

Explore Other Comprehensive 75-Question Subject Tracks

Queryiest

Queryiest – Technology Writer | Software Developer | Digital Learning Enthusiast

What is a database?

How to rotate an array in-place with O(1) space and ...

What steps can businesses take to identify the most valuable ...

Queryiest

Anonymous

paperubofficial