I85PREP
⌘ K
CCIE Data Center study materials
CCIE Data Center study resources

CCIE Data Center practice questions

One question at a time. Build your CCIE Data Center understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Switching, routing, and multicast diagnosis

Two NX-OS switches form a healthy vPC pair. The peer-link fails but peer-keepalive still proves both peers are alive. What is the normal secondary-side protection action, absent special overrides?

  1. Forward all vPC traffic through the keepalive link
  2. Suspend its vPC member ports to avoid dual-active forwarding Correct answer.
  3. Convert every member to an independent routed port
  4. Elect a new STP root and leave both vPC sides forwarding unchanged

Explanation

The secondary suspends vPC members when the peer-link is lost and the peer remains alive. The keepalive distinguishes this condition from peer failure.

Question 2Switching, routing, and multicast diagnosis

A vPC peer-keepalive destination is reachable in the default VRF, but the keepalive is configured in management. Which check is most directly relevant?

  1. Whether the destination answers only from the default VRF
  2. Whether the peer-link alone can carry the keepalive without the configured management route
  3. Only whether a tenant SVI can reach the destination through its own VRF
  4. Reachability from the configured source within the management VRF Correct answer.

Explanation

The keepalive uses its configured source and VRF. A successful test through a different routing table does not establish that this control path works.

Question 3Switching, routing, and multicast diagnosis

On an NX-OS vPC pair, `show vpc` reports a consistency failure after a port-channel change. Which read-only command best examines the relevant compatibility parameters?

  1. show vpc consistency-parameters Correct answer.
  2. show bgp l2vpn evpn
  3. show ip route
  4. show flogi database

Explanation

The vPC consistency-parameters command inspects compatibility information. Resolve the actual mismatch rather than assume that general routing or storage output identifies the vPC problem.

Question 4Switching, routing, and multicast diagnosis

A downstream LACP bundle has two links, but one member is suspended. The member is configured as an access port while the bundle is a trunk. What should be reconciled?

  1. The bundle bandwidth figure without examining individual member settings
  2. The member's compatibility with the intended port-channel configuration Correct answer.
  3. Only the downstream port-channel load-balancing algorithm
  4. Only whether both members have the same description

Explanation

A port-channel requires compatible member parameters. The supplied access-versus-trunk discrepancy is directly relevant before investigating unrelated routing attributes or role preferences.

Question 5Switching, routing, and multicast diagnosis

A trunk permits VLANs 10 and 20 on one end but only VLAN 10 on the other. VLAN 10 works; VLAN 20 does not. What best explains the selective failure?

  1. The allowed-VLAN mismatch prevents the intended VLAN 20 path Correct answer.
  2. An IP address is required on every physical trunk
  3. The native VLAN automatically admits VLAN 20 even when its tag is excluded
  4. LACP must reject both VLANs whenever any list differs

Explanation

VLAN forwarding requires the intended VLAN to be allowed along the path. A link can remain up while a particular tagged VLAN lacks continuity.

Question 6Switching, routing, and multicast diagnosis

A host transmits untagged frames, but its switch port and upstream trunk interpret untagged traffic as different VLANs. What should be checked first?

  1. The host's TCP window scale
  2. Fibre Channel zone membership
  3. BGP local preference for every tenant route
  4. Access/native VLAN expectations at the relevant boundaries Correct answer.

Explanation

Untagged traffic classification is a Layer 2 VLAN issue. Reconcile each boundary's intended access or native behavior before changing unrelated transport or routing policies.

Question 7Switching, routing, and multicast diagnosis

An engineer enables an edge-port treatment on a link that actually connects another switch. What is the key design concern?

  1. It replaces the need for VLAN configuration
  2. It automatically increases the trunk's MTU
  3. It makes the link an OSPF point-to-point adjacency
  4. The link role may bypass appropriate spanning-tree transition and protection assumptions Correct answer.

Explanation

Edge treatment assumes an endpoint-facing role. A switch-facing connection needs an appropriate topology and protection design rather than an assumption that fast transition is always suitable.

Question 8Switching, routing, and multicast diagnosis

With BPDU guard intentionally enabled on a true server-facing edge port, a BPDU arrives. What behavior is the protection intended to trigger?

  1. Convert the BPDU into a BGP update
  2. Disable or error-disable the unexpected bridging connection under the platform policy Correct answer.
  3. Learn the BPDU source as the new anycast gateway
  4. Promote the server to spanning-tree root

Explanation

BPDU guard protects the edge assumption by taking protective action when a BPDU is received. Recovery and exact status follow the configured platform behavior.

Question 9Switching, routing, and multicast diagnosis

Two switches use MST, but their region names or VLAN-to-instance mappings differ. What must the engineer consider?

  1. An identical instance number alone guarantees one region
  2. They may be in different MST regions despite physically adjacent links Correct answer.
  3. Equal VLAN numbers guarantee equal MST region membership despite different mappings
  4. The region mismatch automatically removes all spanning-tree boundaries

Explanation

MST region identity includes relevant configuration parameters, not only an instance number. A mismatch changes region interpretation and should be reconciled with the design.

Question 10Switching, routing, and multicast diagnosis

A port-channel has four 25-Gb/s members. Under an explicitly ideal aggregate-capacity assumption, what is total capacity in one direction?

  1. 200 Gb/s
  2. 50 Gb/s
  3. 100 Gb/s Correct answer.
  4. 25 Gb/s

Explanation

Adding four equal 25-Gb/s links gives 100 Gb/s in one direction. This aggregate figure does not promise that one hashed flow can consume all members.

Question 11Switching, routing, and multicast diagnosis

A single long-lived flow uses one 25-Gb/s member of a four-link bundle while other members are lightly loaded. What is a reasonable explanation?

  1. LACP guarantees byte-perfect load balance each second
  2. Per-flow hashing can preserve flow order without striping that flow across all links Correct answer.
  3. The bundle is necessarily operating at 100 Gb/s for that flow
  4. All idle links must have failed physically

Explanation

Link aggregation commonly distributes flows by a hash. Uneven use by one flow is not sufficient evidence of failed members or a promise of per-packet striping.

Question 12Switching, routing, and multicast diagnosis

An OSPF neighbor remains in EXSTART after an interface replacement. The two interface MTUs differ. What should be investigated before applying an ignore option?

  1. The actual MTU mismatch and end-to-end packet-size requirements Correct answer.
  2. Only that the hello packets are small enough to cross the link
  3. Only that both routers advertise a default route
  4. Whether lowering the dead timer automatically fixes packet-size requirements

Explanation

An MTU mismatch can disrupt OSPF database exchange and can also reflect a real data-plane issue. An ignore option is not evidence that the path supports required packets.

Question 13Switching, routing, and multicast diagnosis

OSPF peers have matching addressing but different hello/dead timers on the common network. What is the most direct correction to evaluate?

  1. Change only the process names while retaining the mismatched link settings
  2. Raise OSPF interface cost while leaving the incompatible timers unchanged
  3. Align required adjacency parameters with the intended network design Correct answer.
  4. Advertise a default route before fixing adjacency

Explanation

OSPF adjacency depends on compatible neighbor parameters. Reconcile timers and other required settings rather than hide the missing adjacency behind unrelated route advertisements.

Question 14Switching, routing, and multicast diagnosis

Two OSPF routers share an Ethernet broadcast network and show 2-WAY with each other; each is fully adjacent with the DR and BDR. What is the best interpretation?

  1. The DR must be the device with the highest IP prefix length
  2. 2-WAY between DROTHERs can be normal on this network type Correct answer.
  3. A 2-WAY state proves mismatched IP MTUs
  4. Every pair must be FULL on every OSPF network type

Explanation

Broadcast OSPF networks do not require full adjacency between every pair of DROTHER routers. Interpret neighbor state with the elected roles and network type.

Question 15Switching, routing, and multicast diagnosis

An OSPF design uses a stub area, but a router inside it is expected to redistribute external routes as an ASBR. What should be reconsidered?

  1. Whether every route should receive a lower administrative distance
  2. The area type and external-route requirements, including whether an appropriate NSSA design is intended Correct answer.
  3. Whether a longer hold timer permits external LSAs in any area
  4. Only lowering the external route metric

Explanation

Area types determine what external routing information and roles are appropriate. Match the intended redistribution requirement to a supported area design rather than adjust unrelated timers.

Question 16Switching, routing, and multicast diagnosis

A forwarding table has 10.0.0.0/8 via A and 10.20.30.0/24 via B. Which route matches destination 10.20.30.7 most specifically?

  1. Both must be load balanced regardless of prefix length
  2. 10.20.30.0/24 via B Correct answer.
  3. 10.0.0.0/8 via A because it was learned first
  4. Neither because the destination is not a network address

Explanation

Longest-prefix matching selects the /24 for this destination. Comparing route preferences for different prefix lengths does not replace the forwarding lookup's specificity rule.

Question 17Switching, routing, and multicast diagnosis

Which single aligned prefix summarizes 10.40.8.0/24 through 10.40.11.0/24 without adding another /24?

  1. 10.40.8.0/23
  2. 10.40.8.0/22 Correct answer.
  3. 10.40.0.0/22
  4. 10.40.8.0/21

Explanation

A /22 covers four consecutive /24 networks on a four-block boundary. Starting at eight includes third-octet values eight through eleven exactly.

Question 18Switching, routing, and multicast diagnosis

Mutual redistribution between two routing domains repeatedly reintroduces routes. What policy mechanism should be evaluated?

  1. Only increasing interface speed
  2. Deleting all connected routes
  3. Tagging and filtering reintroduced routes with explicit redistribution boundaries Correct answer.
  4. Using identical router IDs to merge the protocols

Explanation

Redistribution can create feedback when learned routes return through another boundary. Explicit tagging and filtering help prevent unintended re-entry while preserving required reachability.

Question 19Switching, routing, and multicast diagnosis

A BGP path has an unreachable next hop even though the neighbor session is established. What does session state fail to prove?

  1. That the TCP session itself is functioning
  2. That every received route has a resolvable forwarding next hop Correct answer.
  3. That the configured peer address exists
  4. That BGP messages can be exchanged at all

Explanation

BGP control-session establishment and route usability are distinct. A received path still needs valid next-hop resolution and the relevant policy and installation conditions.

Question 20Switching, routing, and multicast diagnosis

Two otherwise comparable BGP paths differ in local preference: 200 versus 100. Under ordinary Cisco best-path handling, which is favored at that comparison?

  1. Both because local preference is never compared
  2. The path with the longer AS path regardless of local preference
  3. The path with local preference 100
  4. The path with local preference 200 Correct answer.

Explanation

Higher local preference is preferred when this attribute is the deciding comparison. Earlier best-path criteria and policy context still matter in an actual device.

Question 21Switching, routing, and multicast diagnosis

An operator raises Cisco BGP weight on one router and expects all iBGP routers to inherit it. What is the issue?

  1. Weight controls only OSPF costs
  2. Weight is an Ethernet VLAN field
  3. Weight is automatically carried as an EVPN route target
  4. Weight is local to the router, unlike an AS-wide propagated local-preference policy Correct answer.

Explanation

Cisco weight is a local best-path attribute. Use an appropriate propagated policy when the intended outcome concerns other routers in the autonomous system.

Question 22Switching, routing, and multicast diagnosis

Two iBGP clients rely on a route reflector. The reflector learns a client route but outbound policy rejects it toward the other client. What must be examined?

  1. Reflection policy and permitted route advertisement, not just TCP session state Correct answer.
  2. Whether raising the Ethernet MTU forces advertisement
  3. Only whether the reflector's hostname includes RR
  4. Whether every client uses the same MAC address

Explanation

A route reflector changes iBGP advertisement behavior but does not bypass explicit policy. Established sessions do not guarantee that each required route is reflected and accepted.

Question 23Switching, routing, and multicast diagnosis

An eBGP session is intentionally sourced from loopbacks several routed hops apart. What must the design address?

  1. Only a common access VLAN on both remote routers
  2. A static ARP entry for every remote tenant endpoint
  3. Peer-source reachability and appropriate multihop/session settings Correct answer.
  4. Identical autonomous-system numbers as an eBGP requirement

Explanation

Loopback-sourced multihop eBGP requires reachability and compatible session configuration beyond a directly connected peering assumption. Verify both the source and remote endpoint paths.

Question 24Switching, routing, and multicast diagnosis

A BGP filter is intended to allow only 192.0.2.0/24. A policy also accepts its /25 more-specifics. What should be reviewed?

  1. Prefix-length matching conditions in the filter Correct answer.
  2. The time synchronization source
  3. The switch's power supply redundancy
  4. Only the route's next-hop MAC address

Explanation

Prefix filters can distinguish an exact prefix from a range of longer prefixes. Review the matching length conditions rather than assume the network address alone defines the rule.

Question 25Switching, routing, and multicast diagnosis

An IS-IS Level-1 adjacency is intended within one area, but the peers have incompatible area identifiers. What is the relevant check?

  1. The intended level and NSAP/area addressing Correct answer.
  2. Whether changing only the hostname makes the area identifiers match
  3. Only whether the two interfaces have equal metric values
  4. Only the administrative distance of installed IS-IS routes

Explanation

IS-IS level and area relationships affect adjacency. Validate the addressing and intended topology instead of changing unrelated overlay or storage identifiers.

Question 26Switching, routing, and multicast diagnosis

A BFD session is bound to a routing neighbor and fails. What capability is BFD intended to provide?

  1. Encryption of every data packet
  2. Automatic computation of the best BGP policy
  3. Replacement of every routing table
  4. Rapid detection of forwarding-path liveliness for a client protocol Correct answer.

Explanation

BFD provides a liveliness signal that routing or other client protocols can use. It does not independently compute routes, encrypt traffic, or replace policy decisions.

Question 27Switching, routing, and multicast diagnosis

Two first-hop routers form an FHRP group. A host should use which address as its default gateway for the intended shared service?

  1. The OSPF router ID irrespective of subnet
  2. The group's virtual gateway address Correct answer.
  3. The multicast RP address
  4. A different physical address chosen for every packet

Explanation

An FHRP presents a shared virtual gateway that can survive a member failure. The host should use that intended gateway rather than protocol identifiers or per-packet address selection.

Question 28Switching, routing, and multicast diagnosis

An active gateway remains healthy but loses its upstream route. The design requires failover when that dependency fails. What feature should be evaluated?

  1. Appropriate tracking tied to FHRP priority or availability Correct answer.
  2. Increasing the FHRP hello interval while leaving upstream failure untracked
  3. Only lowering the host's ARP timeout
  4. Changing the virtual IP on every host manually

Explanation

Physical router health alone does not prove useful upstream forwarding. Tracking can connect a required dependency to the first-hop redundancy decision under the chosen design.

Question 29Switching, routing, and multicast diagnosis

A PIM sparse-mode receiver joins a group, but there is no usable RP mapping for an ASM group requiring one. What should be investigated?

  1. Only whether IGMP snooping has learned a receiver-facing port
  2. Whether reducing the IGMP query interval automatically creates an RP mapping
  3. Only whether the source appears in an ordinary Ethernet MAC table
  4. The intended RP mapping and rendezvous reachability Correct answer.

Explanation

An ASM sparse-mode design needs the appropriate rendezvous information and paths. Receiver membership alone does not establish the shared-tree control state needed for the flow.

Question 30Switching, routing, and multicast diagnosis

A multicast packet arrives on an interface that fails the reverse-path check toward its source. What is a likely result?

  1. It creates a new shortest unicast route
  2. It disables IGMP on every access port
  3. It is rejected by the multicast RPF check under the relevant forwarding state Correct answer.
  4. It is automatically converted to unicast

Explanation

Multicast RPF checks whether arrival is consistent with the expected path toward the source or tree root. A failed check can prevent forwarding despite receiver interest.

Question 31Switching, routing, and multicast diagnosis

IGMP snooping is enabled on a VLAN with receiver memberships, but there is no active querier. What longer-term issue should be examined?

  1. Membership maintenance and aging without periodic queries Correct answer.
  2. Whether every receiver should be configured as a static PIM neighbor
  3. Whether increasing the snooping table capacity alone prevents membership aging
  4. A mandatory conversion to PIM bidirectional mode

Explanation

A querier supports ongoing membership discovery and refresh. Snooping behavior and forwarding after aging should be verified rather than assuming learned memberships persist indefinitely.

Question 32Switching, routing, and multicast diagnosis

A receiver needs source-specific membership for group G from source S. Which membership capability is relevant?

  1. A group-only static MAC entry as complete proof of source-specific membership
  2. An ASM RP mapping alone without a source-aware receiver request
  3. IGMPv3 source filtering in an appropriate SSM design Correct answer.
  4. Only an IGMPv2 group-only report

Explanation

Source-specific multicast needs a source-aware membership design. IGMPv3 supports source filtering, unlike treating a group-only report as a complete source-specific request.

Question 33Switching, routing, and multicast diagnosis

A routed link supports a 9,000-byte packet, but another segment on the path permits only 1,500. Large DF packets fail while small pings work. What is the key inference?

  1. BGP local preference guarantees fragmentation support
  2. The endpoint's MAC table must be empty
  3. Small-packet reachability does not verify the required end-to-end MTU Correct answer.
  4. Every application packet must fit because ICMP worked once

Explanation

The smallest relevant path MTU can limit large packets. Test the required size and handling, including DF and feedback, rather than infer it from small probes.

Question 34Switching, routing, and multicast diagnosis

A server has 192.0.2.130/26. What is its subnet's network address?

  1. 192.0.2.192
  2. 192.0.2.64
  3. 192.0.2.0
  4. 192.0.2.128 Correct answer.

Explanation

A /26 divides the final octet into blocks of sixty-four addresses. Address 130 belongs to the block beginning at 128 and ending at 191.

Question 35Switching, routing, and multicast diagnosis

A /27 IPv4 subnet uses conventional network and broadcast reservations. How many ordinary host addresses remain?

  1. 31
  2. 32
  3. 30 Correct answer.
  4. 62

Explanation

A /27 leaves five host bits, giving thirty-two total addresses. With the two explicitly stated conventional reservations, thirty ordinary host addresses remain.

Question 36Switching, routing, and multicast diagnosis

A route is present in the control-plane RIB, but hardware forwarding lacks the expected entry. Which distinction is essential?

  1. The route must be a Fibre Channel zone
  2. RIB presence guarantees successful ASIC programming in every case
  3. The hardware table is irrelevant to packet forwarding
  4. Route selection and hardware programming are separate validation stages Correct answer.

Explanation

The control plane can select a route before or despite a data-plane programming issue. Validate next-hop adjacency, hardware installation, and platform resources as separate evidence.

Question 37Switching, routing, and multicast diagnosis

An ACL permits the forward application flow, but return traffic follows a different device where policy blocks it. What should be checked?

  1. Whether the original packet has a lower source port
  2. Both directions, routing symmetry requirements, and the actual policy path Correct answer.
  3. Only whether DNS resolves the name
  4. Only the forward TCP SYN at its source

Explanation

Connectivity requires the relevant return path as well as forward permission. Asymmetric paths can expose different policy or stateful-service behavior even when the initial route exists.

Question 38Switching, routing, and multicast diagnosis

An underlay has equal-cost paths, but all sampled traffic uses one link. Which evidence is needed before declaring ECMP broken?

  1. Installed next hops, hashing inputs, traffic diversity, and per-path counters Correct answer.
  2. Only the sum of all interface line rates
  3. Only the total number of physical links
  4. A requirement that each single flow alternate packets across paths

Explanation

ECMP selection depends on installed equal-cost next hops and hash inputs. A limited sample or one flow can use one path without proving an implementation fault.

Question 39Switching, routing, and multicast diagnosis

Two adjacent switches negotiate LACP, but both are configured passive and no other activation is present. What should be evaluated?

  1. The absence of an active initiator for LACP negotiation Correct answer.
  2. The number of allowed VLANs without examining LACP activity mode
  3. A requirement to assign the same port-channel number on unrelated devices regardless of negotiation
  4. Whether changing only the load-balancing hash initiates LACP

Explanation

Passive LACP participants do not initiate negotiation. The intended bundle needs a compatible active/passive or active/active arrangement rather than unrelated routing changes.

Question 40Switching, routing, and multicast diagnosis

A diagram shows redundant links, but all traverse one line card and one cable tray. What must availability analysis distinguish?

  1. Logical path multiplicity from independent failure domains Correct answer.
  2. Two drawn links from guaranteed physical independence
  3. ECMP configuration from automatic site disaster recovery
  4. A higher VLAN ID from a lower failure probability

Explanation

Redundant logical paths can share physical dependencies. Evaluate line cards, power, conduits, and other common causes rather than count links alone as independent protection.

Question 41ACI objects, access relationships, and policy

In an established ACI fabric, which statement best describes the APIC's ordinary role in user-packet forwarding?

  1. It functions as the mandatory external default gateway
  2. It manages policy and fabric state but is not an inline hop for every endpoint packet Correct answer.
  3. Every endpoint packet is routed through an APIC appliance
  4. It replaces all leaf forwarding tables with controller-only lookups

Explanation

ACI distributes forwarding and policy enforcement to fabric switches. Management through APIC does not mean every application packet traverses the controller appliances.

Question 42ACI objects, access relationships, and policy

An ACI tenant object exists but no VRF or bridge domain has been associated with an endpoint EPG. What does creating the tenant alone fail to establish?

  1. A valid organizational container
  2. An entry in the controller's object model
  3. The complete forwarding and policy relationships required for the application Correct answer.
  4. A named management object

Explanation

A tenant provides organizational scope, but endpoint connectivity depends on additional objects and relationships. Existence of the container alone is not a completed application configuration.

Question 43ACI objects, access relationships, and policy

Two bridge domains must share one IP routing table while retaining separate Layer 2 domains. Which relationship fits that intent?

  1. Remove all bridge-domain-to-VRF associations
  2. Associate both bridge domains with the intended common VRF Correct answer.
  3. Put each bridge domain in a separate isolated VRF with no leaking
  4. Replace both bridge domains with one EPG only

Explanation

A VRF supplies the routing context, while bridge domains define Layer 2 scope. Sharing a VRF does not require merging the bridge domains into one Layer 2 domain.

Question 44ACI objects, access relationships, and policy

Web and database endpoints are in different EPGs under an enforced VRF, with no preferred-group or other bypass. Routes resolve but TCP 5432 is blocked. What policy relationship should be examined?

  1. Whether disabling ARP suppression alone grants the TCP service
  2. Only whether both EPGs use the same access VLAN number
  3. The contract, filter, and provider/consumer relationships for the intended service Correct answer.
  4. Only the bridge-domain subnet because routed reachability was already verified

Explanation

In the stated enforced policy model, route reachability does not automatically permit inter-EPG traffic. Check the intended service filter and the relevant contract attachments.

Question 45ACI objects, access relationships, and policy

An ACI contract filter permits TCP destination port 443. A test uses TCP destination port 8443 and other policy is unchanged. What should be verified?

  1. Whether the bridge domain assigns transport ports
  2. Whether changing source MAC makes 8443 equal 443
  3. Whether all TCP traffic is automatically permitted by any TCP entry
  4. Whether the intended filter actually matches the tested destination port Correct answer.

Explanation

A filter matches its configured fields. A successful HTTPS-style service on one port is not permission for every other TCP destination port.

Question 46ACI objects, access relationships, and policy

Two endpoints in the same ordinary EPG communicate without a contract. No intra-EPG isolation or intra-EPG contract restriction is configured. What is the correct interpretation?

  1. This can be normal intra-EPG behavior in the stated configuration Correct answer.
  2. It proves every endpoint belongs to every EPG
  3. It requires an L3Out for the local exchange
  4. It proves that all inter-EPG contracts are disabled

Explanation

Default intra-EPG communication should not be confused with permission between different EPGs. The question excludes the settings that would impose the relevant intra-EPG restrictions.

Question 47ACI objects, access relationships, and policy

An administrator wants same-EPG endpoints isolated from one another. What should be evaluated instead of assuming separate names enforce isolation?

  1. Only assigning a different endpoint display name for each server
  2. Increasing the bridge-domain subnet mask length without redesign
  3. Only placing the servers on separate leaf ports while keeping unrestricted policy
  4. The appropriate intra-EPG isolation or policy mechanism for the actual release and requirements Correct answer.

Explanation

Policy isolation needs an actual supported enforcement mechanism. Inventory labels and descriptions do not by themselves change communication permissions between endpoints in an EPG.

Question 48ACI objects, access relationships, and policy

A static EPG path binding specifies VLAN 210, but VLAN 210 is absent from the associated domain's VLAN pool. What is the targeted check?

  1. Only the endpoint's DNS suffix
  2. The EPG-domain-access-policy chain and its permitted encapsulation Correct answer.
  3. Whether the spine has the server's local VLAN configured manually
  4. Only the APIC login timeout

Explanation

The path binding depends on a coherent domain and access-policy relationship. A requested encapsulation outside the applicable pool can prevent intended deployment even when the EPG exists.

Question 49ACI objects, access relationships, and policy

An ACI physical interface is operationally up, but its intended EPG is not deployed there. What should be traced?

  1. Only optical power because link-up proves policy deployment
  2. Interface selector, policy group, AEP, domain, and EPG binding relationships Correct answer.
  3. The server's SAN boot LUN before checking the network mapping
  4. Only the tenant and EPG object names without their attachment relationships

Explanation

Physical link state and logical policy deployment are separate. Trace the complete intended access mapping rather than interpret carrier state as proof of EPG attachment.

Question 50ACI objects, access relationships, and policy

An endpoint is learned in the wrong EPG after a host trunk change. Which evidence is most useful?

  1. Only the fabric's total health score
  2. The observed port, encapsulation, and classification bindings for that endpoint Correct answer.
  3. Only whether the endpoint can reach an out-of-band management address
  4. The number of APIC appliances without path information

Explanation

Endpoint classification depends on the applicable attachment and policy context. Compare actual observed placement and encapsulation with the intended mapping before changing unrelated application settings.