350-601

Free 350-601 practice questions

10 questions from our 350-601 practice tests, each with the correct answer and a written explanation. Read them with the answers shown, or switch to quiz mode and answer first.

Back to the 350-601 practice exams

Question 1 of 10, Network

An organization interconnects three VXLAN BGP EVPN fabrics that are built on Cisco Nexus 9000 switches. Each fabric must remain a separate failure domain, with BUM flooding contained inside each site where possible. Which component should handle the traffic between the fabrics?

  1. AOTV edge devices that extend the Layer 2 VLANs between the three data centers
  2. BBorder gateways that terminate and re-originate the VXLAN tunnels between sites Correct answer
  3. CAn ACI Multi-Pod inter-pod network carrying PIM Bidir traffic between the sites
  4. DSpine switches extended across sites as one shared underlay with a single IGP

Answer and explanation: In VXLAN EVPN Multi-Site, border gateways terminate the site-internal tunnels and re-originate them toward other sites, which keeps each fabric a separate control plane and failure domain and controls BUM forwarding. ACI Multi-Pod applies to ACI rather than NX-OS fabrics, a single shared underlay makes the sites one failure domain, and OTV is an older Layer 2 extension technology that does not integrate with EVPN fabrics in this way.

Question 2 of 10, Network

Two Cisco Nexus switches form a vPC domain. The vPC peer link fails while the peer-keepalive link remains up and both switches stay healthy. What is the default behavior?

  1. AThe operational secondary suspends its vPC member ports and vPC VLAN SVIs while the primary keeps forwarding Correct answer
  2. BBoth peers suspend every vPC member port until the peer link is restored, to prevent a split-brain condition
  3. CBoth peers keep forwarding on all vPC member ports because the keepalive proves neither peer has failed
  4. DThe operational primary suspends its vPC member ports so that the secondary can take over the primary role

Answer and explanation: When the peer link fails but keepalives still arrive, the secondary knows the primary is alive, so it suspends its vPC member ports and the SVIs of vPC VLANs to avoid a dual-active state, and the primary keeps forwarding. Leaving both peers forwarding would create inconsistent MAC and ARP state, the primary does not hand over its role in this case, and suspending all ports on both peers would cause an unnecessary full outage.

Question 3 of 10, Network

A VXLAN BGP EVPN fabric uses an eBGP design in which each leaf has its own AS number. The leaves use route-target both auto under their VRFs and VNIs, and routes from other leaves are received but not imported. Which change resolves this with the least configuration?

  1. AConfigure rewrite-evpn-rt-asn so that received route targets use the local AS Correct answer
  2. BChange the route distinguisher to a manual value that is the same on all leaves
  3. CConfigure allowas-in on all leaves so that their own AS is accepted in the path
  4. DChange every leaf to the same AS number and keep the auto route targets

Answer and explanation: Auto route targets are derived from the local AS and VNI, so leaves in different AS numbers generate different route targets; rewrite-evpn-rt-asn rewrites the AS part of received route targets to the local AS so they match. The route distinguisher does not control import, allowas-in addresses AS path loops rather than route target mismatch, and moving all leaves into one AS is a redesign rather than a minimal fix.

Question 4 of 10, Compute

A UCS Manager administrator creates two vNIC templates, one for fabric A and one for fabric B, which must always have identical VLANs and MTU. When VLANs are added to the fabric A template, the change must be copied to the fabric B template automatically. What should be configured?

  1. AAn updating service profile template that contains vNICs from both of the vNIC templates
  2. BFabric failover on the fabric A template so that it mirrors its settings to fabric B
  3. CA template redundancy pair with fabric A as the primary and fabric B as the secondary Correct answer
  4. DA LAN connectivity policy that references both templates in the same service profile

Answer and explanation: A vNIC template redundancy pair links a primary and a secondary template, and shared settings such as VLANs and MTU configured on the primary are synchronized to the secondary. A LAN connectivity policy groups vNICs but does not synchronize templates, fabric failover moves traffic rather than copying settings, and an updating service profile template propagates profile changes rather than vNIC template settings.

Question 5 of 10, Compute

A pair of Cisco UCS fabric interconnects uplinks to the data center LAN. The design requires that the fabric interconnects do not participate in spanning tree with the upstream switches and that server vNICs are pinned to uplink ports. Which Ethernet switching mode meets this requirement?

  1. ATransparent bridging mode, in which all VLANs are bridged between uplinks without MAC learning
  2. BN-Port Virtualization mode, in which the FIs proxy server logins to the upstream core switches
  3. CSwitch mode, in which the fabric interconnects run spanning tree with the upstream LAN switches
  4. DEnd-host mode, in which server vNICs are pinned to uplinks and the FIs do not run STP upstream Correct answer

Answer and explanation: End-host mode is the default Ethernet mode: the fabric interconnects appear as a host to the LAN, pin server vNICs to uplinks, and do not run spanning tree toward the upstream switches. Switch mode makes them participate in STP, NPV is a Fibre Channel mode that applies to vHBA logins rather than Ethernet, and there is no transparent bridging mode on UCS fabric interconnects.

Question 6 of 10, Compute

A company wants Cisco Intersight to configure its UCS fabric interconnects and blade servers directly through domain profiles and server profiles, with no UCS Manager instance. Which management mode should be used?

  1. AUCS Manager mode with a device connector claim, in which Intersight only reads the domain inventory
  2. BIntersight Standalone Mode, which manages individual rack servers through their own Cisco IMC
  3. CIntersight Assist, which proxies Intersight requests to on-premises vCenter and storage endpoints
  4. DIntersight Managed Mode, in which the FIs and servers are configured with profiles from Intersight Correct answer

Answer and explanation: Intersight Managed Mode replaces UCS Manager on the fabric interconnects, and Intersight configures the domain and servers through domain, chassis, and server profiles. Claiming a UCS Manager domain gives inventory and limited actions while UCS Manager keeps the configuration role, Standalone Mode targets rack servers without fabric interconnects, and Intersight Assist is a virtual appliance that connects Intersight to third-party on-premises endpoints.

Question 7 of 10, Storage Network

A storage team wants to measure exchange completion time and IOPS for each initiator, target, and LUN combination on its Cisco MDS fabric to find which hosts experience storage latency. Which solution provides this visibility?

  1. ASPAN sessions on each host port that mirror traffic to a local Fibre Channel analyzer
  2. BThe show fcns database command run periodically on every switch in the fabric
  3. CSAN Analytics on the MDS switches, viewed through the SAN Insights dashboards Correct answer
  4. DPort monitor policies that track TxWait and credit counters on the host ports

Answer and explanation: SAN Analytics inspects FC and SCSI or NVMe headers on MDS ports to calculate metrics such as exchange completion time and IOPS per initiator, target, and LUN, which SAN Insights in Nexus Dashboard displays. Port monitor reacts to port-level counters, the name server shows logins rather than performance, and SPAN to a local analyzer does not scale across the fabric.

Question 8 of 10, Storage Network

A storage team uses device aliases in zoning on a Cisco MDS fabric. When a host HBA is replaced, they want to change only the PWWN in the device alias, with every zone that uses the alias being updated automatically. Which device alias mode is required?

  1. AEnhanced mode with each alias also defined as a matching fcalias inside every VSAN
  2. BBasic mode with CFS distribution disabled for the device alias database
  3. CEnhanced mode, in which zones reference the alias name rather than the PWWN Correct answer
  4. DBasic mode, in which every zone stores the PWWN that the alias resolved to

Answer and explanation: In enhanced mode, zones store the device alias name, so changing the PWWN of an alias updates every zone that references it. In basic mode the alias is replaced by its PWWN when the zone is created, so each zone must be edited, disabling CFS distribution only makes the database inconsistent, and fcaliases are VSAN-specific objects that do not add this behavior.

Question 9 of 10, Automation and Artificial Intelligence

A team uses a generative AI assistant to draft NX-OS configuration changes for its VXLAN EVPN fabric. Which practice BEST reduces the risk of deploying an incorrect change that the assistant produced?

  1. ATreat the output as a draft that is peer reviewed and checked by pre-change analysis before deployment Correct answer
  2. BLimit the assistant to show commands so that it never produces any configuration for the fabric
  3. CApply the change directly during a maintenance window and roll it back with a checkpoint if anything fails
  4. DAsk the assistant to confirm that its own output is correct and complete before the change is applied

Answer and explanation: Generated configuration can contain errors, so it should go through the same peer review and automated validation, such as pre-change analysis or CI tests, as any other change before it reaches production. Rolling back after a failure still causes an outage, self-confirmation by the same model is not independent validation, and restricting the assistant to show commands removes the benefit instead of managing the risk.

Question 10 of 10, Security

DHCP snooping and Dynamic ARP Inspection are enabled on a VLAN on a Cisco Nexus switch. Hosts that use DHCP work, but two servers with static IP addresses lose connectivity because their ARP packets are dropped. Inspection must stay active on their ports. What should be configured?

  1. AAn ARP ACL that permits the IP and MAC pairs of the static servers Correct answer
  2. BThe two server ports as DHCP snooping trusted interfaces for the VLAN
  3. CA higher DAI rate limit on the two server ports for ARP packets
  4. DIP Source Guard disabled on the two server ports in the inspected VLAN

Answer and explanation: DAI validates ARP packets against the DHCP snooping binding table, which has no entries for statically addressed hosts, so an ARP ACL with their IP-to-MAC bindings lets their ARP packets pass while inspection stays active. Trusting the ports for DHCP does not create bindings, IP Source Guard does not drop ARP packets, and a higher rate limit does not change the validation result.