McAfee-Secured Website

Exam Code: 4A0-D01

Exam Name: Nokia Data Center Fabric Fundamentals

Certification Provider: Nokia

100% Nokia 4A0-D01 Guaranteed Success With Testking Training Materials

Get Real 4A0-D01 Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

61 Questions & Answers with Testing Engine

"Nokia Data Center Fabric Fundamentals Exam", also known as 4A0-D01 exam, is a Nokia certification exam.

Pass your tests with the always up-to-date 4A0-D01 Exam Engine. Your 4A0-D01 training materials keep you at the head of the pack!

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Was: $137.49
Now: $124.99

Nokia 4A0-D01 Practice Test 2

Question 10

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, the data center fabric team observes: Two tenant environments use identical private IP prefixes and operators assume they must be renumbered. What observation should be required before this case is closed?

  1. A path review confirms expected hop count and multipath reachability between tested leaf pairs
  2. Each proposed device meets its required fabric port and routing capabilities in the design inventory
  3. The routing table exposes multiple equal-cost next hops and a controlled link failure preserves reachability
  4. Both tenants reach their own gateways without inter-tenant route leakage

Explanation

Correct Answer

D:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, both tenants reach their own gateways without inter-tenant route leakage. This is linked directly to the observed symptom (Two tenant environments use identical private IP prefixes and operators assume they must be renumbered) and cannot be replaced by an unrelated healthy dashboard.

Incorrect Answers

A:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: A path review confirms expected hop count and multipath reachability between tested leaf pairs. Using that approach would investigate A proposed three-tier design forces east-west traffic through an unnecessary aggregation hop instead. No evidence for that separate failure is supplied here; the reported case is consistent with: Tenant addressing was not separated with routing instances and overlay service boundaries.

B:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: Each proposed device meets its required fabric port and routing capabilities in the design inventory. The alternative focuses on The device-role assignment was made without an approved platform and interface capability matrix, not on the stated defect. For this scenario the expected acceptance result would be: Both tenants reach their own gateways without inter-tenant route leakage.

C:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: The routing table exposes multiple equal-cost next hops and a controlled link failure preserves reachability. This option assumes the other problem (An engineer proposes spanning-tree-blocked links as the principal load-balancing method for a new routed fabric) is present. The case instead describes Two tenant environments use identical private IP prefixes and operators assume they must be renumbered, for which the causal issue is Tenant addressing was not separated with routing instances and overlay service boundaries.

Question 11

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, the data center fabric team observes: Two tenant environments use identical private IP prefixes and operators assume they must be renumbered. What causal finding should the investigation prioritize?

  1. The architecture copies a legacy hierarchical topology instead of the intended Clos fabric
  2. The design confuses legacy Layer 2 loop prevention with Layer 3 equal-cost forwarding
  3. Tenant addressing was not separated with routing instances and overlay service boundaries
  4. The design overlooks the EVPN control plane needed to distribute overlay reachability information

Explanation

Correct Answer

C:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, the evidence points to: Tenant addressing was not separated with routing instances and overlay service boundaries. The follow-up test (Both tenants reach their own gateways without inter-tenant route leakage) distinguishes this mechanism from a separate fabric fault.

Incorrect Answers

A:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: The architecture copies a legacy hierarchical topology instead of the intended Clos fabric. The alternative focuses on The architecture copies a legacy hierarchical topology instead of the intended Clos fabric, not on the stated defect. For this scenario the expected acceptance result would be: Both tenants reach their own gateways without inter-tenant route leakage.

B:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: The design confuses legacy Layer 2 loop prevention with Layer 3 equal-cost forwarding. This option assumes the other problem (An engineer proposes spanning-tree-blocked links as the principal load-balancing method for a new routed fabric) is present. The case instead describes Two tenant environments use identical private IP prefixes and operators assume they must be renumbered, for which the causal issue is Tenant addressing was not separated with routing instances and overlay service boundaries.

D:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: The design overlooks the EVPN control plane needed to distribute overlay reachability information. The option describes a different failure mode: A change plan assumes that VXLAN alone advertises remote endpoint locations. It would address The design overlooks the EVPN control plane needed to distribute overlay reachability information, but here the case reports Two tenant environments use identical private IP prefixes and operators assume they must be renumbered. The current required evidence is: Both tenants reach their own gateways without inter-tenant route leakage.

Question 12

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, the data center fabric team observes: Two tenant environments use identical private IP prefixes and operators assume they must be renumbered. What ongoing operational safeguard directly addresses this root issue?

  1. Treat configuration reproducibility as a fabric architecture requirement
  2. Review VRF-to-tenant assignments before connecting new workloads
  3. Maintain a versioned switch-role and hardware support matrix
  4. Validate east-west path requirements when selecting data center topology

Explanation

Correct Answer

B:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, review VRF-to-tenant assignments before connecting new workloads. This ongoing review specifically addresses: Tenant addressing was not separated with routing instances and overlay service boundaries. It should remain part of change acceptance, not only incident closure.

Incorrect Answers

A:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: Treat configuration reproducibility as a fabric architecture requirement. This option assumes the other problem (Fabric operations require per-device CLI edits even for repeated tenant changes) is present. The case instead describes Two tenant environments use identical private IP prefixes and operators assume they must be renumbered, for which the causal issue is Tenant addressing was not separated with routing instances and overlay service boundaries.

C:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: Maintain a versioned switch-role and hardware support matrix. The option describes a different failure mode: A hardware inventory assigns 7220 IXR leaf roles to every spine without validating model capabilities. It would address The device-role assignment was made without an approved platform and interface capability matrix, but here the case reports Two tenant environments use identical private IP prefixes and operators assume they must be renumbered. The current required evidence is: Both tenants reach their own gateways without inter-tenant route leakage.

D:

At Pinebrook Municipal Networks, while expanding a dual-spine Clos topology, this option proposes: Validate east-west path requirements when selecting data center topology. That choice would be more relevant when this different symptom occurs: A proposed three-tier design forces east-west traffic through an unnecessary aggregation hop. The present scenario instead points to: Tenant addressing was not separated with routing instances and overlay service boundaries. The correct resolution has its own verification outcome: Both tenants reach their own gateways without inter-tenant route leakage.

Question 13

At Farsight Health Systems, while evaluating routed-fabric capacity, the data center fabric team observes: An operations team assumes a healthy management dashboard proves the fabric dataplane is passing traffic. Which supported remediation most directly addresses the stated failure?

  1. Reassess connectivity using a leaf-spine layout with predictable routed paths for east-west workloads
  2. Use supported routed leaf-spine adjacencies and appropriate ECMP routes for parallel fabric paths
  3. Validate the selected Nokia IXR platform, supported interfaces and software against its assigned spine or leaf role
  4. Combine management health with independent protocol state and representative traffic checks

Explanation

Correct Answer

D:

At Farsight Health Systems, while evaluating routed-fabric capacity, the targeted correction is: Combine management health with independent protocol state and representative traffic checks. The reported symptom is explained by: Central management availability was mistaken for underlay and overlay forwarding health. Confirm the change by testing this outcome: Controlled fabric traffic passes while the associated routing and service states agree with design.

Incorrect Answers

A:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: Reassess connectivity using a leaf-spine layout with predictable routed paths for east-west workloads. The option describes a different failure mode: A proposed three-tier design forces east-west traffic through an unnecessary aggregation hop. It would address The architecture copies a legacy hierarchical topology instead of the intended Clos fabric, but here the case reports An operations team assumes a healthy management dashboard proves the fabric dataplane is passing traffic. The current required evidence is: Controlled fabric traffic passes while the associated routing and service states agree with design.

B:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: Use supported routed leaf-spine adjacencies and appropriate ECMP routes for parallel fabric paths. That choice would be more relevant when this different symptom occurs: An engineer proposes spanning-tree-blocked links as the principal load-balancing method for a new routed fabric. The present scenario instead points to: Central management availability was mistaken for underlay and overlay forwarding health. The correct resolution has its own verification outcome: Controlled fabric traffic passes while the associated routing and service states agree with design.

C:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: Validate the selected Nokia IXR platform, supported interfaces and software against its assigned spine or leaf role. This is a valid category of fabric maintenance for another scenario (A hardware inventory assigns 7220 IXR leaf roles to every spine without validating model capabilities), not the targeted action for Farsight Health Systems. The distinction is the actual cause: Central management availability was mistaken for underlay and overlay forwarding health.

Question 14

At Farsight Health Systems, while evaluating routed-fabric capacity, the data center fabric team observes: An operations team assumes a healthy management dashboard proves the fabric dataplane is passing traffic. Which evidence directly validates the intended repair?

  1. The leaf reaches the required peers over both paths and traffic continues after one uplink is disabled
  2. The tested broadcast domain is isolated by tenant while routed underlay links remain independently reachable
  3. Controlled fabric traffic passes while the associated routing and service states agree with design
  4. Both tenants reach their own gateways without inter-tenant route leakage

Explanation

Correct Answer

C:

At Farsight Health Systems, while evaluating routed-fabric capacity, the decisive validation is: Controlled fabric traffic passes while the associated routing and service states agree with design. It verifies that the change (Combine management health with independent protocol state and representative traffic checks) resolved the mechanism: Central management availability was mistaken for underlay and overlay forwarding health.

Incorrect Answers

A:

An additional leaf is cabled only to one spine in a two-spine fabric. Here, central management availability was mistaken for underlay and overlay forwarding health.

B:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: The tested broadcast domain is isolated by tenant while routed underlay links remain independently reachable. This is a valid category of fabric maintenance for another scenario (A team extends a single large Layer 2 broadcast domain across every rack to simplify mobility), not the targeted action for Farsight Health Systems. The distinction is the actual cause: Central management availability was mistaken for underlay and overlay forwarding health.

D:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: Both tenants reach their own gateways without inter-tenant route leakage. Using that approach would investigate Two tenant environments use identical private IP prefixes and operators assume they must be renumbered instead. No evidence for that separate failure is supplied here; the reported case is consistent with: Central management availability was mistaken for underlay and overlay forwarding health.

Question 15

At Farsight Health Systems, while evaluating routed-fabric capacity, the data center fabric team observes: An operations team assumes a healthy management dashboard proves the fabric dataplane is passing traffic. Which underlying mechanism best accounts for this symptom?

  1. Central management availability was mistaken for underlay and overlay forwarding health
  2. The leaf-spine design did not compare permitted transit behavior across redundant fabric paths
  3. The design confuses legacy Layer 2 loop prevention with Layer 3 equal-cost forwarding
  4. The capacity model ignores EVPN service instances, MAC entries, routing instances and VTEP state

Explanation

Correct Answer

A:

At Farsight Health Systems, while evaluating routed-fabric capacity, for Farsight Health Systems, investigate this cause first: Central management availability was mistaken for underlay and overlay forwarding health. It is directly linked to the observed symptom rather than another service issue.

Incorrect Answers

B:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: The leaf-spine design did not compare permitted transit behavior across redundant fabric paths. Although related to Fabric architecture and data center design, the proposed change fits another mechanism: The leaf-spine design did not compare permitted transit behavior across redundant fabric paths. The evidence in the current case requires a different result: Controlled fabric traffic passes while the associated routing and service states agree with design.

C:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: The design confuses legacy Layer 2 loop prevention with Layer 3 equal-cost forwarding. Using that approach would investigate An engineer proposes spanning-tree-blocked links as the principal load-balancing method for a new routed fabric instead. No evidence for that separate failure is supplied here; the reported case is consistent with: Central management availability was mistaken for underlay and overlay forwarding health.

D:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: The capacity model ignores EVPN service instances, MAC entries, routing instances and VTEP state. The alternative focuses on The capacity model ignores EVPN service instances, MAC entries, routing instances and VTEP state, not on the stated defect. For this scenario the expected acceptance result would be: Controlled fabric traffic passes while the associated routing and service states agree with design.

Question 16

At Farsight Health Systems, while evaluating routed-fabric capacity, the data center fabric team observes: An operations team assumes a healthy management dashboard proves the fabric dataplane is passing traffic. Which review gate should be added to prevent the same error?

  1. Validate east-west path requirements when selecting data center topology
  2. Use a leaf-to-all-spines link matrix when accepting fabric expansions
  3. Keep control-plane and data-plane evidence separate in service health reporting
  4. Reassess service-scale assumptions after tenant onboarding increases

Explanation

Correct Answer

C:

At Farsight Health Systems, while evaluating routed-fabric capacity, for Farsight Health Systems, include the safeguard: Keep control-plane and data-plane evidence separate in service health reporting. It reduces repeat risk from the original mechanism and complements a validation of: Controlled fabric traffic passes while the associated routing and service states agree with design.

Incorrect Answers

A:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: Validate east-west path requirements when selecting data center topology. Although related to Fabric architecture and data center design, the proposed change fits another mechanism: The architecture copies a legacy hierarchical topology instead of the intended Clos fabric. The evidence in the current case requires a different result: Controlled fabric traffic passes while the associated routing and service states agree with design.

B:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: Use a leaf-to-all-spines link matrix when accepting fabric expansions. Using that approach would investigate An additional leaf is cabled only to one spine in a two-spine fabric instead. No evidence for that separate failure is supplied here; the reported case is consistent with: Central management availability was mistaken for underlay and overlay forwarding health.

D:

At Farsight Health Systems, while evaluating routed-fabric capacity, this option proposes: Reassess service-scale assumptions after tenant onboarding increases. This option assumes the other problem (A design review counts physical server VLANs as the only measure of tenant scale) is present. The case instead describes An operations team assumes a healthy management dashboard proves the fabric dataplane is passing traffic, for which the causal issue is Central management availability was mistaken for underlay and overlay forwarding health.

Question 17

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, the data center fabric team observes: A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks. What configuration or operational change should the engineer prioritize?

  1. Establish and verify routed IP reachability between the planned VTEP addresses before enabling overlays
  2. Align approved routing and filtering intent across equal-cost paths while preserving deliberate exceptions
  3. Use supported routed leaf-spine adjacencies and appropriate ECMP routes for parallel fabric paths
  4. Validate the selected Nokia IXR platform, supported interfaces and software against its assigned spine or leaf role

Explanation

Correct Answer

A:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, establish and verify routed IP reachability between the planned VTEP addresses before enabling overlays. This addresses the specific failure mechanism (The overlay was designed without a working routed underlay between tunnel endpoints) rather than treating only the visible alarm. The required acceptance evidence is: Each pair of intended VTEP source addresses reaches its peer over the underlay.

Incorrect Answers

B:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: Align approved routing and filtering intent across equal-cost paths while preserving deliberate exceptions. This option assumes the other problem (Two spines have asymmetric network policy and traffic takes a different security path on failover) is present. The case instead describes A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks, for which the causal issue is The overlay was designed without a working routed underlay between tunnel endpoints.

C:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: Use supported routed leaf-spine adjacencies and appropriate ECMP routes for parallel fabric paths. It tests a different symptom: An engineer proposes spanning-tree-blocked links as the principal load-balancing method for a new routed fabric. For the observed case, the relevant distinction is The overlay was designed without a working routed underlay between tunnel endpoints, with verification: Each pair of intended VTEP source addresses reaches its peer over the underlay.

D:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: Validate the selected Nokia IXR platform, supported interfaces and software against its assigned spine or leaf role. The option describes a different failure mode: A hardware inventory assigns 7220 IXR leaf roles to every spine without validating model capabilities. It would address The device-role assignment was made without an approved platform and interface capability matrix, but here the case reports A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks. The current required evidence is: Each pair of intended VTEP source addresses reaches its peer over the underlay.

Question 18

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, the data center fabric team observes: A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks. What test result proves the affected fabric function now works?

  1. Each pair of intended VTEP source addresses reaches its peer over the underlay
  2. The routing table exposes multiple equal-cost next hops and a controlled link failure preserves reachability
  3. A failover test shows equivalent permitted forwarding behavior through either spine
  4. A path review confirms expected hop count and multipath reachability between tested leaf pairs

Explanation

Correct Answer

A:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this result is the relevant closure evidence: Each pair of intended VTEP source addresses reaches its peer over the underlay. It checks the effect of: Establish and verify routed IP reachability between the planned VTEP addresses before enabling overlays. The underlying issue was: The overlay was designed without a working routed underlay between tunnel endpoints.

Incorrect Answers

B:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: The routing table exposes multiple equal-cost next hops and a controlled link failure preserves reachability. It tests a different symptom: An engineer proposes spanning-tree-blocked links as the principal load-balancing method for a new routed fabric. For the observed case, the relevant distinction is The overlay was designed without a working routed underlay between tunnel endpoints, with verification: Each pair of intended VTEP source addresses reaches its peer over the underlay.

C:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: A failover test shows equivalent permitted forwarding behavior through either spine. The option describes a different failure mode: Two spines have asymmetric network policy and traffic takes a different security path on failover. It would address The leaf-spine design did not compare permitted transit behavior across redundant fabric paths, but here the case reports A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks. The current required evidence is: Each pair of intended VTEP source addresses reaches its peer over the underlay.

D:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: A path review confirms expected hop count and multipath reachability between tested leaf pairs. That choice would be more relevant when this different symptom occurs: A proposed three-tier design forces east-west traffic through an unnecessary aggregation hop. The present scenario instead points to: The overlay was designed without a working routed underlay between tunnel endpoints. The correct resolution has its own verification outcome: Each pair of intended VTEP source addresses reaches its peer over the underlay.

Question 19

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, the data center fabric team observes: A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks. What specific misconfiguration or design error is most likely?

  1. The design confuses legacy Layer 2 loop prevention with Layer 3 equal-cost forwarding
  2. The architecture copies a legacy hierarchical topology instead of the intended Clos fabric
  3. The overlay was designed without a working routed underlay between tunnel endpoints
  4. The device-role assignment was made without an approved platform and interface capability matrix

Explanation

Correct Answer

C:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, the evidence points to: The overlay was designed without a working routed underlay between tunnel endpoints. The follow-up test (Each pair of intended VTEP source addresses reaches its peer over the underlay) distinguishes this mechanism from a separate fabric fault.

Incorrect Answers

A:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: The design confuses legacy Layer 2 loop prevention with Layer 3 equal-cost forwarding. It tests a different symptom: An engineer proposes spanning-tree-blocked links as the principal load-balancing method for a new routed fabric. For the observed case, the relevant distinction is The overlay was designed without a working routed underlay between tunnel endpoints, with verification: Each pair of intended VTEP source addresses reaches its peer over the underlay.

B:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: The architecture copies a legacy hierarchical topology instead of the intended Clos fabric. The option describes a different failure mode: A proposed three-tier design forces east-west traffic through an unnecessary aggregation hop. It would address The architecture copies a legacy hierarchical topology instead of the intended Clos fabric, but here the case reports A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks. The current required evidence is: Each pair of intended VTEP source addresses reaches its peer over the underlay.

D:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: The device-role assignment was made without an approved platform and interface capability matrix. This is a valid category of fabric maintenance for another scenario (A hardware inventory assigns 7220 IXR leaf roles to every spine without validating model capabilities), not the targeted action for Stonegate Rail Logistics. The distinction is the actual cause: The overlay was designed without a working routed underlay between tunnel endpoints.

Question 20

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, the data center fabric team observes: A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks. What repeatable acceptance check would reduce recurrence?

  1. Use a leaf-to-all-spines link matrix when accepting fabric expansions
  2. Require underlay transport tests as a prerequisite for overlay service acceptance
  3. Test redundant-path policy parity before production handoff
  4. Validate east-west path requirements when selecting data center topology

Explanation

Correct Answer

B:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, require underlay transport tests as a prerequisite for overlay service acceptance. This ongoing review specifically addresses: The overlay was designed without a working routed underlay between tunnel endpoints. It should remain part of change acceptance, not only incident closure.

Incorrect Answers

A:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: Use a leaf-to-all-spines link matrix when accepting fabric expansions. The option describes a different failure mode: An additional leaf is cabled only to one spine in a two-spine fabric. It would address The leaf-spine cabling plan omitted its second equal-cost uplink, but here the case reports A diagram shows tenant VXLAN traffic but omits IP reachability between VTEP loopbacks. The current required evidence is: Each pair of intended VTEP source addresses reaches its peer over the underlay.

C:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: Test redundant-path policy parity before production handoff. This is a valid category of fabric maintenance for another scenario (Two spines have asymmetric network policy and traffic takes a different security path on failover), not the targeted action for Stonegate Rail Logistics. The distinction is the actual cause: The overlay was designed without a working routed underlay between tunnel endpoints.

D:

At Stonegate Rail Logistics, while reviewing fabric architecture tradeoffs, this option proposes: Validate east-west path requirements when selecting data center topology. Although related to Fabric architecture and data center design, the proposed change fits another mechanism: The architecture copies a legacy hierarchical topology instead of the intended Clos fabric. The evidence in the current case requires a different result: Each pair of intended VTEP source addresses reaches its peer over the underlay.

Nokia Data Center Fabric Fundamentals (4A0-D01) Practice Tests