350-401 ENCOR — Cisco Certified Specialist - Enterprise Core
Test your knowledge with official exam-style questions
Questions and options are shuffled each attempt
Email me this practice set
Get a link sent to your inbox so you can pick this back up anytime.
▶Cisco Certified Specialist - Enterprise Core — Practice Set 1: All Questions & Explanations
Full question text, answer options, and explanations for this practice set — a spoiler-free alternative is the interactive quiz above for scored, shuffled practice.
1. A company is growing from a single office building into a multi-building campus. Network engineers now expect at least three separate distribution blocks that must all interconnect. According to Cisco's campus design guidance, what is the main reason to introduce a dedicated core layer at this point?
- A. The core layer gives endpoints and users direct wired and wireless access to the network
- B. The core layer aggregates access layers and connects them to services such as WAN routers and WLAN controllers
- C. The core layer replaces the need for any redundant uplinks because it is inherently resilient
- D. The core layer provides connectivity between distribution layers in larger LAN environments and avoids a full mesh of links between every pair of distribution blocks(correct)
Explanation: Cisco's campus guide says a core layer 'provides connectivity between distribution layers for large LAN environments' and suggests considering one once at least three distributions must connect together. It also reduces link counts compared with a full mesh between distributions. Why the other options are wrong: A: That is the role of the access layer, not the core layer. B: That describes the distribution layer. The core layer's role is connectivity between distribution layers. C: A core layer does not remove the need for redundancy; the guide sizes the core with redundant links (N x 2 for N distributions).
2. In a Cisco Catalyst SD-WAN deployment, a new branch WAN Edge router must be authenticated and told how to reach the controllers, including when it sits behind a NAT device. Which SD-WAN component performs this authentication and connectivity-coordination role?
- A. SD-WAN Manager
- B. SD-WAN Controller
- C. SD-WAN Validator(correct)
- D. WAN Edge router
Explanation: The SD-WAN Validator authenticates devices and coordinates connectivity between edge routers and controllers, and it aids NAT traversal. Why the other options are wrong: A: The Manager is the centralized management system that stores certificates and configurations and pushes them to devices; it is not the orchestration and NAT-traversal component. B: The Controller is the centralized brain that handles control-plane routing and key reflection, not device authentication and NAT-traversal coordination. D: WAN Edge routers handle the transmission of data traffic; they are the devices being authenticated, not the authenticator.
3. In a Cisco SD-Access fabric, endpoints attach to a switch that performs endpoint registration, provides the Layer 3 anycast gateway and applies VXLAN encapsulation. Which fabric role is this switch playing?
- A. Fabric edge node(correct)
- B. Fabric control plane node
- C. Fabric intermediate node
- D. Fabric border node
Explanation: Edge nodes are the fabric equivalent of access-layer switches: they connect endpoints, perform endpoint registration, provide the anycast gateway and perform VXLAN encapsulation. Why the other options are wrong: B: The control plane node is built on LISP (map-server and map-resolver) and maintains the host tracking database; it does not attach endpoints or encapsulate their traffic. C: Intermediate nodes only provide IP routing and transport between fabric-role devices and do not need VXLAN or LISP support. D: Border nodes are the gateway between the fabric site and external networks; they do not provide the endpoint-facing anycast gateway.
4. A network engineer reviews a QoS policy and sees 'police' configured on one class and 'shape average' on another. Which statement correctly distinguishes the two behaviors on a Catalyst 9300 switch?
- A. Policing buffers excess packets until bandwidth is available, whereas shaping immediately discards them
- B. Both policing and shaping can only be applied in the output direction
- C. Shaping buffers excess packets rather than dropping them immediately, whereas policing applies configured conform, exceed and violate actions to traffic above the rate(correct)
- D. Shaping marks packets with a new DSCP value, while policing assigns packets to queues
Explanation: The Catalyst 9300 QoS guide states shaping (shape average) buffers excess packets rather than dropping them, while policing imposes a maximum rate and takes conform, exceed and violate actions. Policing applies to input and output, and shaping to output only. Why the other options are wrong: A: This reverses the two behaviors. Shaping is the feature that buffers excess traffic. B: The guide's comparison shows policing works for input and output, while shaping is implemented for output traffic only. D: Marking is done with set commands (for example set dscp), and queueing uses bandwidth/priority commands. Neither is what shaping or policing does.
5. A server in a data center boots directly into KVM with no general-purpose operating system installed underneath it, and KVM then runs several guest operating systems. How is this hypervisor classified?
- A. Type 1 (native or bare-metal) hypervisor(correct)
- B. A virtual switch
- C. Type 2 (hosted) hypervisor
- D. A VRF instance
Explanation: A Type 1 hypervisor runs directly on the host's hardware to manage guest operating systems (KVM, Hyper-V and VMware vSphere are examples), while a Type 2 hypervisor runs on top of a conventional operating system. Why the other options are wrong: B: A virtual switch is a software layer-2 switch that connects virtual machines to a network; it is not the layer that runs guest operating systems. C: A Type 2 hypervisor runs as software on top of a conventional host operating system, such as VirtualBox or VMware Workstation on a desktop. D: A VRF is a separate routing table on a router or switch and has nothing to do with running guest operating systems.
6. An engineer is following the Catalyst 9300 Multi-VRF CE chapter and has created VRF 'CUST-A' with 'ip vrf CUST-A'. Which interface-level command places a routed port into that VRF?
- A. rd CUST-A
- B. address-family ipv4 vrf CUST-A
- C. ip vrf forwarding CUST-A(correct)
- D. tunnel vrf CUST-A
Explanation: The guide states 'ip vrf forwarding vrf-name' associates a routed port or SVI with a VRF; the VRF itself is created with 'ip vrf vrf-name', the RD is set with 'rd', and the BGP address family is 'address-family ipv4 vrf vrf-name'. Why the other options are wrong: A: 'rd' is a VRF-configuration-mode command that sets the route distinguisher (an AS or IP address plus an arbitrary number); it does not bind an interface. B: This enters a per-VRF address family under BGP for PE-to-CE sessions; it does not assign the interface to the VRF. D: 'tunnel vrf' associates a VRF with a specific GRE tunnel destination, not a routed port.
7. A data center team is moving from VLAN-based segmentation to VXLAN because they will soon need far more than 4,000 Layer 2 segments. What property of VXLAN makes this possible?
- A. VXLAN uses the same 12-bit VLAN ID but encrypts each frame so segments cannot overlap
- B. VXLAN tags frames with a 16-bit security group value that defines the segment
- C. VXLAN removes the need for an IP underlay because it extends Layer 2 natively over any medium
- D. The VXLAN network identifier (VNI) is a 24-bit segment ID, allowing up to 16 million segments in one administrative domain(correct)
Explanation: Cisco describes VXLAN as a MAC-in-UDP encapsulation protocol running as a virtual overlay on an existing Layer 3 network. Each segment is identified by a 24-bit VNI, so up to 16 million segments are possible, compared with the limit of 4094 VLANs. Why the other options are wrong: A: VXLAN does not reuse the 12-bit VLAN ID and is not an encryption technology; it is a MAC-in-UDP encapsulation using a 24-bit VNI. B: A 16-bit value is the SGT used in TrustSec policy. The VXLAN segment identifier is the 24-bit VNI. C: VXLAN runs as a virtual overlay on an existing Layer 3 network, so an IP underlay is required.
8. Two Catalyst switches are connected with a single Ethernet cable. Neither port has been configured with a switchport mode command, and the ports are in the same VTP domain. What is the resulting link state?
- A. A trunk forms using native VLAN 1 because dynamic auto defaults to trunk when the peer is also dynamic auto
- B. The link does not become a trunk, because the default mode (dynamic auto) only trunks if the neighbor is set to trunk or dynamic desirable(correct)
- C. A trunk forms because DTP negotiates trunking automatically on default ports
- D. Both ports are error-disabled because the mode is not explicitly configured
Explanation: The VLAN trunk chapter says the default switchport mode for all Ethernet interfaces is dynamic auto, and dynamic auto can convert to a trunk only if the neighbor is set to trunk or dynamic desirable mode. Why the other options are wrong: A: The documented rule is that dynamic auto converts to a trunk only if the neighbor is trunk or dynamic desirable. C: Dynamic auto is passive. Two dynamic auto ports wait for the other side to initiate, so no trunk forms. D: Default dynamic auto ports do not err-disable; they simply remain nontrunking.
9. Two switches must form a negotiated LACP EtherChannel between their uplinks. Which combination of channel-group modes on the two ends will successfully form the channel?
- A. One side PAgP desirable and the other side LACP active
- B. Both sides PAgP auto
- C. One side active and the other side passive(correct)
- D. Both sides passive
Explanation: For LACP, active starts negotiation and passive only responds, so active with active or active with passive works, but two passive ports cannot form a channel. PAgP and LACP cannot be mixed in a group. Why the other options are wrong: A: PAgP and LACP cannot be mixed in the same channel; groups running different protocols cannot interoperate. B: Two auto ports cannot form a channel for the same reason as two passive ports; this is also a PAgP mode, not LACP. D: Two passive ports cannot form a channel because neither side initiates LACP negotiation.
10. BPDU guard is enabled globally on PortFast access ports of a Catalyst switch. A user plugs a small unmanaged switch that sends BPDUs into one of those ports. What happens to the port?
- A. The port transitions from forwarding back through listening and learning and then forwards again
- B. The port is placed in the error-disabled state and shuts down until an administrator restores it(correct)
- C. The port silently ignores the BPDUs and keeps forwarding
- D. The port is placed in the root-inconsistent state and continues to forward user traffic
Explanation: With BPDU guard on PortFast ports, if any BPDU arrives the port is error-disabled and the entire port shuts down; the guide says you must manually put the interface back in service. Why the other options are wrong: A: BPDU guard does not simply reconverge the port; it error-disables it. C: Ignoring BPDUs would be BPDU filter behavior. BPDU guard exists precisely to react to them. D: Root-inconsistent is the behavior of root guard, not BPDU guard, and a root-inconsistent port is blocked.
11. A router learns the same prefix 10.50.0.0/16 from an internal EIGRP neighbor and from an OSPF neighbor, with default administrative distances on both. Which route is installed in the routing table?
- A. The route with the lowest metric is installed regardless of protocol
- B. The EIGRP route, because internal EIGRP has an administrative distance of 90 versus 110 for OSPF(correct)
- C. The OSPF route, because link-state protocols are always preferred over advanced distance vector protocols
- D. Both routes are installed and load-balanced because the prefix lengths match
Explanation: The IP Routing guide lists default administrative distances of 90 for internal EIGRP and 110 for OSPF. The lower AD is more trusted, so the EIGRP route wins. Why the other options are wrong: A: Metrics are only compared between routes of the same protocol; the administrative distance decides between different protocols. C: Route selection between protocols is by administrative distance, not by protocol type, and OSPF's AD of 110 is higher than internal EIGRP's 90. D: Load balancing applies to equal-cost paths from the same routing source. Different routing sources are compared by administrative distance first.
12. On a link between R1 and R2, 'show ip ospf neighbor' shows the neighbor stuck in EXSTART. R1's interface MTU is 1500 and R2's interface MTU is 9000. Which statement best describes the cause?
- A. Both routers are configured as passive on this interface, so they never send hellos
- B. The hello and dead intervals differ between the routers, so hellos are being discarded
- C. The MTU mismatch prevents the database description (DBD) exchange from completing, so the adjacency cannot move past EXSTART/EXCHANGE(correct)
- D. Both routers have OSPF priority 0, so they stay in the 2-WAY state
Explanation: Cisco's troubleshooting note identifies an MTU mismatch as the most common cause of neighbors stuck in EXSTART/EXCHANGE. The router with the higher MTU keeps sending DBD packets that are too large, and the lower-MTU router rejects them. Why the other options are wrong: A: Passive interfaces suppress hellos entirely, so no neighbor entry would exist at all, rather than a neighbor stuck in EXSTART. B: Cisco's note requires hello and dead timers to match for an adjacency; a timer mismatch stops the neighbor relationship forming at all, so it would not show a neighbor that has progressed as far as EXSTART. D: Two priority-0 neighbors stay in 2-WAY and never attempt a full adjacency, which is a different state from EXSTART.
13. A router has two valid BGP paths to the same prefix. Among the attributes WEIGHT, LOCAL_PREF, AS_PATH length and MED, which one does the BGP best-path algorithm evaluate first?
- A. AS_PATH length
- B. WEIGHT, which is Cisco-specific and local to the router(correct)
- C. LOCAL_PREF
- D. MED
Explanation: Cisco's best-path document lists the order as WEIGHT (highest wins, local to the router), then LOCAL_PREF, then local origination, AS_PATH length, origin, MED, eBGP over iBGP, IGP metric to the next hop, and so on. Why the other options are wrong: A: AS_PATH length is evaluated after WEIGHT, LOCAL_PREF and local origination. C: LOCAL_PREF is evaluated second, after WEIGHT. D: MED is evaluated after AS_PATH and origin, and only compared when the neighboring AS is the same.
14. A Catalyst 9300 interface is configured for HSRP with only a group number and a virtual IP address; no timers or priority are set. Which THREE values apply by default?
- A. Hold time of 10 seconds(correct)
- B. Hello interval of 3 seconds(correct)
- C. HSRP version 2 is used by default
- D. Priority 100(correct)
- E. The virtual MAC address is 0000.5E00.01XX
Explanation: The HSRP chapter documents a default priority of 100, a default hello interval of 3 seconds and a default hold time of 10 seconds. The default version is HSRPv1, whose virtual MAC is 0000.0c07.acXX. Why the other options are wrong: C: The default HSRP version is HSRPv1 (version 2 must be enabled explicitly). E: That is the VRRP virtual MAC format. HSRPv1 uses 0000.0c07.acXX where XX is the group number.
15. An engineer configures 'logging host 10.20.30.40' on a Catalyst switch and does not change any other logging setting. Which messages does the syslog server receive?
- A. No messages until 'logging trap' is explicitly configured
- B. Debugging (level 7) messages and all lower levels
- C. Only emergency (level 0) messages
- D. Informational (level 6) messages and numerically lower levels(correct)
Explanation: The System Message Logs chapter states that by default syslog servers receive informational messages and numerically lower levels (the default server severity is Informational), whereas the console receives debugging messages and lower levels. Why the other options are wrong: A: Syslog servers receive informational messages and numerically lower levels by default once a host is configured. B: Debugging is the default for the console (and terminal) but not for syslog servers. C: The default server severity is informational, so much more than level 0 is sent.
16. An engineer wants to measure jitter between two Cisco routers with a 'udp-jitter' IP SLA operation. What must be configured on the destination router for the operation to work?
- A. The IP SLA responder, enabled with 'ip sla responder'(correct)
- B. Nothing, because UDP jitter is measured passively from normal traffic
- C. A SPAN destination session, so the operation can copy packets
- D. An SNMP RW community, because IP SLA statistics are written by SNMP
Explanation: The IP SLA responder is required for UDP jitter operations and is configured on the target device with 'ip sla responder'. By contrast, ICMP echo operations do not need a responder. Why the other options are wrong: B: UDP jitter is an active operation that requires the responder on the target. C: SPAN copies traffic for monitoring and is unrelated to IP SLA responders. D: SNMP is not required for the responder; the responder is enabled with 'ip sla responder'.
17. A switch configuration contains both 'enable password Cisc0123' and 'enable secret 9 <hash>'. Which password does the switch require for privileged EXEC access?
- A. The enable secret, because it takes precedence over the enable password(correct)
- B. The enable password, because it was configured first
- C. Either one; the switch accepts both interchangeably
- D. Neither; privileged EXEC access requires a line password
Explanation: The Catalyst 9300 security guide states that if you configure the enable secret command, it takes precedence over the enable password command. The enable secret is also stored with a nonreversible hash, while the enable password is not encrypted in the configuration file. Why the other options are wrong: B: Configuration order does not matter; if an enable secret exists it takes precedence. C: Only the enable secret is used when both are configured. D: A line password is used for line access, not as a replacement for the enable secret.
18. A switch uses 'aaa authentication login default group tacacs+ local' and vty lines use the default list. The TACACS+ server is reachable but rejects the user's password. What happens?
- A. Authentication fails and the local database is not consulted, because the next method is tried only when the previous method does not respond(correct)
- B. The switch falls back to the local database and authenticates the user if the local username exists
- C. The switch grants access with the line password because the TACACS+ rejection is treated as an error
- D. The switch retries TACACS+ indefinitely until the user enters the correct password
Explanation: The authentication chapter explains that the software moves to the next listed method only when the previous method gives no response. If a server or the local database actively denies access, authentication stops and no other methods are tried. Why the other options are wrong: B: Fallback to the next method happens only when the previous method gives no response, not when a reachable server actively rejects the credentials. C: A rejection is a definitive denial, not an error, and the line password is not part of this method list. D: The user is denied after the rejection; the router does not loop on the TACACS+ method.
19. An inbound extended ACL on a router interface contains these entries in order: '10 permit ip 10.1.1.0 0.0.0.255 any' then '20 deny ip host 10.1.1.5 any'. What happens to traffic from host 10.1.1.5?
- A. It is denied by the implicit deny at the end of the list
- B. It is logged and then dropped because host entries are more specific
- C. It is denied, because deny entries always override permit entries
- D. It is permitted, because the first matching entry (sequence 10) determines the result(correct)
Explanation: The IPv4 ACL chapter states that the first match determines whether the device accepts or rejects the packet, and that each ACL ends with an implicit deny only for packets that match no entry. Host 10.1.1.5 matches the broader permit at sequence 10 first. Why the other options are wrong: A: The implicit deny only applies to packets that match no explicit entry; this packet matches entry 10. B: ACLs do not apply longest-match logic; processing is sequential and ends at the first match. C: ACLs are processed in order and the first match wins; deny does not automatically override.
20. A network team wants to ensure that a flood of unnecessary traffic punted to the CPU of a Catalyst switch cannot overwhelm the control plane. Which feature is designed for this?
- A. Flexible NetFlow, which exports flow records to a collector
- B. MACsec, which encrypts frames between two switches
- C. SPAN, which mirrors traffic to a monitoring port
- D. Control Plane Policing (CoPP), which uses policers on CPU queues to protect the CPU from unnecessary traffic and DoS attacks(correct)
Explanation: The CoPP chapter says CoPP improves device security by protecting the CPU from unnecessary traffic and DoS attacks, using the modular QoS CLI and CPU queues with dedicated hardware policers. Why the other options are wrong: A: Flexible NetFlow is a monitoring feature and does not limit traffic reaching the CPU. B: MACsec provides authentication and encryption between MACsec-capable devices and does not police traffic to the CPU. C: SPAN copies traffic for analysis and does not protect the control plane.
21. Which IEEE standard does Cisco identify as the one for authenticating and encrypting packets between two MACsec-capable devices?
- A. IEEE 802.3ad
- B. IEEE 802.1AE(correct)
- C. IEEE 802.1w
- D. IEEE 802.1Q
Explanation: The MACsec chapter describes MACsec as the IEEE 802.1AE standard for authenticating and encrypting packets between two MACsec-capable devices, with MACsec Key Agreement (MKA) managing the keys. Why the other options are wrong: A: 802.3ad is the link aggregation (LACP) standard. C: 802.1w is Rapid Spanning Tree Protocol. D: 802.1Q is the VLAN tagging and trunking standard; it does not encrypt frames.
22. What is the output of this Python script? interfaces = ['Gi1/0/1', 'Gi1/0/2', 'Gi1/0/3'] interfaces.append('Gi1/0/4') print(interfaces.pop(0), len(interfaces))
- A. Gi1/0/1 3(correct)
- B. Gi1/0/1 4
- C. Gi1/0/4 4
- D. Gi1/0/4 3
Explanation: append adds 'Gi1/0/4' to the end (four items). pop(0) removes and returns the item at index 0, 'Gi1/0/1', leaving three items, so print outputs 'Gi1/0/1 3'. Why the other options are wrong: B: After append there are four items, and pop(0) removes one, so len() is 3. C: pop(0) returns the first element, and the list has three items after the pop. D: pop(0) removes the item at index 0, not the last item; pop() with no index removes the last item.
23. A developer must send a JSON body to a REST API. Which of the following is a valid JSON document?
- A. {"vlan": 10, "name": "Users", "active": true}(correct)
- B. {'vlan': 10, 'name': 'Users', 'active': true}
- C. {"vlan": 10, "name": "Users", "active": True}
- D. {"vlan": 10, "name": "Users",}
Explanation: RFC 8259 defines strings as beginning and ending with quotation marks, lowercase literal names (false, null, true), and commas only between members or elements, so only the first document is valid. Why the other options are wrong: B: JSON strings, including names, must use double quotation marks; single quotes are not valid JSON. C: JSON literal names must be lowercase (true, false, null); True is Python syntax. D: A comma may appear only between members; a trailing comma after the last member is not valid in the JSON grammar.
24. A script sends a POST request to the Cisco Catalyst Center discovery API to create a new discovery job and receives HTTP 202 Accepted with a taskId and url in the body. What should the script do next?
- A. Request a new token with basic authentication, because 202 means the token expired
- B. Send the POST again until the response code becomes 200
- C. Treat the 202 as proof that the discovery was created successfully and finish
- D. Issue a GET on the task URL (/dna/intent/api/v1/task/{taskId}) with a valid X-Auth-Token and check isError, because 202 does not guarantee the task succeeded(correct)
Explanation: Catalyst Center returns a taskId and URL with 202 Accepted. The script retrieves the task with GET /task/{taskId} (with X-Auth-Token) and inspects isError and related fields. The documentation shows a discovery with a duplicate name returning 202 even though the task then fails. Why the other options are wrong: A: An expired token returns 401 Unauthorized, not 202 Accepted. B: Repeating the POST can create duplicates or fail with a duplicate-name error; the task ID is how progress is tracked. C: The documentation warns that 202 Accepted does not necessarily imply success of the task that the request created.
25. An EEM applet has action labels 10.0, 2.0 and 5.0, each running a CLI command. In what order do the actions execute?
- A. 5.0, then 2.0, then 10.0
- B. In the order the action lines were typed into the configuration
- C. 10.0, then 2.0, then 5.0(correct)
- D. 2.0, then 5.0, then 10.0
Explanation: The EEM guide states that actions run in ascending alphanumeric (lexicographical) order based on the label, and notes that because '10.0' sorts before '2.0' you should use labels such as 01.0, 02.0 and so on. Why the other options are wrong: A: No documented rule produces this order; labels run in ascending alphanumeric order. B: Execution order is determined by the label sort order, not the order the lines were entered. D: That is numeric order, but EEM sorts action labels as strings in ascending alphanumeric order, and '10.0' sorts before '2.0'.