CCNA Multiple Select — 10 Questions

Question 1

Which TWO statements correctly describe the causes or implications of CRC errors, runts, giants, or output errors as seen in the output of 'show interface' or 'show interface status'?

Accepted Answer

A high number of runts on an interface typically indicates excessive collisions or a faulty NIC.. Option B is correct because runts—frames smaller than 64 bytes—often result from collisions truncating frames on half-duplex links or a faulty NIC. Option D is correct because duplex mismatch can cause late collisions, which appear as output errors in 'show interface'; a device on one side full-duplex and the other half-duplex leads to collisions and framing errors. Option A is wrong because CRC errors can stem from faulty cabling, interference, or a mismatched NIC, not exclusively a bad switch port. Option C is wrong because giants (frames over maximum MTU) may be forwarded if the interface is configured with jumbo frames or the switch is set to accept oversize frames. Option E is wrong because 'show controllers' displays frame-size errors like runts and giants, including details beyond CRC errors.

Answer

CRC errors are always caused by a faulty switch port and require port replacement.

Answer

Giants are frames that exceed the maximum transmission unit (MTU) and are always discarded by the switch.

Answer

The 'show controllers' command provides a detailed view of CRC errors but does not show runts or giants.

Question 2

Which TWO statements accurately describe how AI/ML concepts are applied to network operations in modern enterprise networks?

Accepted Answer

Supervised machine learning models can be used to classify network traffic into predefined categories, such as identifying whether traffic is voice, video, or data.. Option A is correct because supervised learning uses labeled data to classify traffic (e.g., voice, video, data). Option B is correct because anomaly detection often uses unsupervised learning to identify deviations from normal behavior. Option C is wrong because reinforcement learning is not used for spam classification; that task uses supervised learning. Option D is wrong because clustering groups data but cannot predict exact bandwidth usage; prediction requires regression models. Option E is wrong because predictive analytics in network operations leverages machine learning models, not solely static thresholds defined by administrators.

Answer

Reinforcement learning is primarily used to automatically classify email traffic as spam or not spam based on a labeled dataset.

Answer

Clustering algorithms, a type of unsupervised learning, are used to predict the exact bandwidth usage of a specific application over the next hour.

Answer

Predictive analytics in network operations relies solely on static thresholds defined by network administrators to forecast potential failures.

Question 3

Which TWO switch port configurations are required when connecting a Cisco IP phone and a desktop PC to a single access port?

Accepted Answer

Use the 'switchport voice vlan' command to assign a dedicated voice VLAN.. Option B is correct because the 'switchport voice vlan' command assigns a dedicated VLAN for voice traffic, allowing the IP phone to tag its packets with the voice VLAN ID while the PC remains in the native (data) VLAN. Option D is correct because 'mls qos trust cos' preserves the Layer 2 Class of Service (CoS) markings from the IP phone, ensuring voice packets receive appropriate QoS treatment across the network. Option A is incorrect because a trunk port is not required—the access port with the voice VLAN command handles both VLANs without trunking. Option C is incorrect because disabling Spanning Tree Protocol (STP) is not a recommended practice and does not prevent voice delays; STP is essential for loop prevention and can be tuned with PortFast instead. Option E is incorrect because the port must remain a Layer 2 access port, not a routed port, to support both the PC and IP phone.

Answer

Configure the port as a trunk and allow both VLANs.

Answer

Disable spanning tree on the port to prevent voice delays.

Answer

Configure the port as a routed port with an IP address for management.

Question 4

Which TWO statements about IPv4/IPv6 static routing are true?

Accepted Answer

An IPv6 default static route can be configured using the destination prefix ::/0.. Option B is correct because the IPv6 default static route uses the destination prefix ::/0, which matches all IPv6 addresses, similar to 0.0.0.0/0 in IPv4. Option D is correct because a floating static route is configured with a higher administrative distance and only becomes active when the primary route (with a lower AD) is removed or fails. Option A is wrong: a floating static route is configured with a higher administrative distance, not lower. Option C is wrong: a directly connected route has an administrative distance of 0, which is always preferred over a static route (even with AD 1). Option E is wrong: IPv4 static routes use the 'ip route' command, while IPv6 static routes use the 'ipv6 route' command; the syntax is different.

Answer

A floating static route is configured with a lower administrative distance than the primary route.

Answer

A static route with an administrative distance of 1 is preferred over a directly connected route.

Answer

IPv4 and IPv6 static routes are configured using the same command syntax.

Question 5

Which four of the following correctly describe how AI/ML techniques can improve network operations in a modern enterprise? (Choose all that apply.)

Accepted Answer

AI models can analyze historical traffic data to predict future bandwidth congestion. Options A, B, C, and D are correct. AI models can predict bandwidth congestion by analyzing historical traffic data, enabling proactive capacity planning. Natural language processing (NLP) automates helpdesk ticket responses by interpreting user intent, reducing manual effort. Reinforcement learning can dynamically adjust firewall rules in response to evolving attack patterns, improving threat response without human intervention. Unsupervised learning can cluster behavior patterns to identify unknown device types on the network. Option E ("AI eliminates the need for baseline performance metrics because it learns in real-time") is incorrect because even AI/ML models require baseline metrics to establish normal behavior and detect anomalies; real-time learning does not remove the need for baselines. Option F ("ML models always require labeled training data to be effective in network operations") is incorrect because many ML techniques, such as unsupervised learning (as shown in option D), operate effectively on unlabeled data by discovering patterns and clusters without predefined labels.

Answer

AI eliminates the need for baseline performance metrics because it learns in real-time

Answer

ML models always require labeled training data to be effective in network operations

Question 6

Which three of the following statements about VLAN Trunking Protocol (VTP) are correct? (Choose three.)

Accepted Answer

VTP pruning helps reduce unnecessary broadcast traffic on trunk links by pruning VLANs that are not active on any downstream switch.. All three selected statements are correct: VTP pruning reduces unnecessary broadcast traffic by pruning inactive VLANs from trunk links; a VTP server can create/modify/delete VLANs and propagates changes; transparent mode forwards advertisements but does not synchronize. The three distractors are incorrect. D: VTP version 3 is not backward compatible with version 1 and 2; it requires compatible versions. E: A VTP client cannot create local VLANs; it only synchronizes from the server. F: VTP uses multicast MAC address 0100.0CCC.CCCC for advertisements, not 0180.c200.0000 which is used by other protocols like STP.

Answer

VTP version 3 supports extended VLANs (1006-4094) and is backward compatible with VTP version 1 and 2.

Answer

A switch in VTP client mode can create local VLANs that are not advertised to the rest of the domain.

Answer

VTP advertisements are sent as multicast frames using the destination MAC address 0180.c200.0000.

Question 7

Which three statements about Power over Ethernet (PoE) and PoE+ standards are correct? (Choose three.)

Accepted Answer

PoE (IEEE 802.3af) can deliver up to 15.4 watts of power per port.. Option 1 is correct because the IEEE 802.3af PoE standard specifies a maximum power delivery of 15.4 watts per port at the PSE (Power Sourcing Equipment), with a minimum of 12.95 watts guaranteed at the PD (Powered Device) after cable losses. Option 2 is correct because the IEEE 802.3at PoE+ standard increases the maximum power per port to 30 watts at the PSE, with 25.5 watts available at the PD. Option 4 is correct because PDs can negotiate power requirements using LLDP with the IEEE 802.1AB LLDP-MED (Media Endpoint Discovery) extensions, which include PoE TLV (Type-Length-Value) fields for power negotiation beyond the simple classification method. Option 3 (both PoE and PoE+ use the same four pairs) is incorrect because both standards deliver power over only two pairs (spare or data pairs depending on mode); four-pair power delivery is introduced in IEEE 802.3bt (PoE++). Option 5 (PoE+ requires Category 3 cabling) is incorrect because PoE+ requires at least Category 5e cabling to support the higher power levels without excessive heat or signal degradation. Option 6 (PoE automatically disables power if a non-PoE device is detected) is incorrect because the PSE first performs a detection phase to identify a valid PoE signature; if none is found, power is never applied—so it is not disabled after being enabled.

Answer

Both PoE and PoE+ use the same four pairs of a twisted-pair cable for power delivery.

Answer

PoE+ requires Category 3 cabling or better.

Answer

PoE automatically disables power if a non-PoE device is detected.

Question 8

Which three of the following are valid considerations when configuring EtherChannel? (Choose three.)

Accepted Answer

EtherChannel can be configured using either PAgP (Cisco proprietary) or LACP (IEEE 802.3ad) protocols.. All three correct statements (A, B, C) are valid considerations when configuring EtherChannel. Option D is incorrect because load balancing can be based on source/destination IP, MAC, or Layer 4 port, not just MAC addresses. Option E is false because although up to 16 interfaces can be in a port channel, only 8 can be active simultaneously (the remaining are in standby). Option F is false because STP operates on the port-channel interface as a whole, not on individual links within the bundle.

Answer

EtherChannel load balancing is always based on source and destination MAC addresses only.

Answer

On a Cisco switch, you can bundle up to 16 interfaces into a single EtherChannel, all of which can be active simultaneously.

Answer

EtherChannel provides loop prevention by using Spanning Tree Protocol on each individual link within the bundle.

Question 9

Which three of the following are correct steps in the process of CDP (Cisco Discovery Protocol) neighbor discovery? (Choose three.)

Accepted Answer

CDP operates at Layer 2 and sends advertisements to the multicast address 01:00:0C:CC:CC:CC.. CDP is a Cisco proprietary Layer 2 protocol that uses multicast MAC address 01:00:0C:CC:CC:CC to send advertisements to directly connected neighbors. Advertisements include device identifier, platform, and capabilities. CDP is enabled by default on Cisco devices and runs over interfaces supporting SNAP headers. Incorrect options: CDP advertisements are sent every 60 seconds by default (not 30); CDP discovers only directly connected neighbors (not devices more than one L2 hop away); CDP does not require both devices to be in the same IP subnet because it operates at Layer 2.

Answer

CDP advertisements are sent every 30 seconds by default.

Answer

CDP can discover devices that are more than one Layer 2 hop away.

Answer

CDP requires that both devices be in the same IP subnet.

Question 10

A network engineer is implementing AIOps to improve network reliability. Which four of the following are core capabilities that AIOps platforms typically provide? (Choose four.)

Accepted Answer

Correlation of alerts from multiple network devices to reduce noise. Correlation of alerts from multiple network devices is a core AIOps capability because it uses machine learning to analyze and group related alerts, reducing noise and helping engineers focus on root causes rather than individual symptoms. Self-healing by automatically reverting problematic configuration changes is a key feature, as AIOps can detect anomalies caused by changes and roll back to a stable state to minimize downtime. Predictive maintenance identifies devices likely to fail by analyzing historical performance data and telemetry, enabling proactive replacement or repair before failures occur. Baseline learning of normal traffic patterns allows AIOps to detect deviations that indicate potential issues. The incorrect options are not core AIOps capabilities: Direct manipulation of physical layer hardware is beyond the scope of AIOps, which focuses on software-based analysis and automation; guaranteed elimination of all false positives is unrealistic because no system can achieve 100% accuracy in alarm filtering.

Answer

Direct manipulation of physical layer hardware to replace faulty optics

Answer

Guaranteed elimination of all false positive alarms in the network