McAfee-Secured Website

Exam Bundle

Exam Code: 350-501

Exam Name Implementing and Operating Cisco Service Provider Network Core Technologies (SPCOR)

Certification Provider: Cisco

Corresponding Certifications: CCIE Service Provider , CCNP Service Provider

Cisco 350-501 Bundle $28.79

100% Cisco 350-501 Guaranteed Success With Testking Training Materials

Get Real 350-501 Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

  • Questions & Answers

    350-501 Questions & Answers

    528 Questions & Answers

    The ultimate exam preparation tool, Questions & Answers mirrors the actual 350-501 exam while allowing you to hone and measure every aspect of your test-taking ability.

  • 350-501 Video Course

    350-501 Video Course

    253 Video Lectures

    350-501 Video Course is developed by Cisco Professionals to help you pass the 350-501 exam.

    Description

    This course will improve your knowledge and skills required to pass Implementing and Operating Cisco Service Provider Network Core Technologies (SPCOR) exam.
  • Study Guide

    350-501 Study Guide

    1160 PDF Pages

    Developed by industry experts, this 1160-page guide spells out in painstaking detail all of the information you need to ace 350-501 exam.

Implementing and Operating Cisco Service Provider Network Core Technologies (SPCOR) (350-501) Practice Tests

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
Total Cost: $189.97
Bundle Price: $161.18

Purchase Individually

  • Questions & Answers

    Questions & Answers

    528 Questions

    $124.99
  • 350-501 Video Course

    Video Course

    253 Video Lectures

    $39.99
  • Study Guide

    Study Guide

    1160 PDF Pages

    $24.99

Cisco 350-501 Practice Test 53

Question 520

For a metro Ethernet service provider, the change record describes: An IS-IS neighbor is Up but the remote IPv6 prefix is missing because the address-family topology was disabled on one transit segment. What preventive validation should be standard?

  1. Verify IPv4/IPv6 route consistency after partial protocol changes.
  2. Trend prefix counts and alert before capacity thresholds are reached.
  3. Tune BFD to observed transport behavior and test at peak utilization.
  4. Review candidate redistribution tables under a default-deny policy.

Explanation

Correct Answer

A:

Verify IPv4/IPv6 route consistency after partial protocol changes. This helps prevent recurrence of the issue: An IS-IS neighbor is Up but the remote IPv6 prefix is missing because the address-family topology was disabled on one transit segment. Acceptance evidence should be: The IPv6 route appears with a valid next hop while all existing IPv4 paths are unchanged.

Incorrect Answers

B:

Trend prefix counts and alert before capacity thresholds are reached. That is a sensible control for ANOTHER scenario: An eBGP exchange reports Established but a contracted prefix is missing because the advertiser rejects it under a maximum-prefix protection threshold. It is less targeted to this issue than: Verify IPv4/IPv6 route consistency after partial protocol changes.

C:

Tune BFD to observed transport behavior and test at peak utilization. That is a sensible control for ANOTHER scenario: A router's OSPF adjacency flaps after an optical circuit fails over because differing BFD timer expectations cause repeated session-down detection. It is less targeted to this issue than: Verify IPv4/IPv6 route consistency after partial protocol changes.

D:

Review candidate redistribution tables under a default-deny policy. That is a sensible control for ANOTHER scenario: An edge router injects all OSPF internals into BGP without tagging or policy, exposing infrastructure loopbacks to a customer upstream. It is less targeted to this issue than: Verify IPv4/IPv6 route consistency after partial protocol changes.

Question 521

For a global tier-one IP transit carrier, the failure is: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname. Which authorized correction best addresses it?

  1. Use nonoverlapping deterministic port-set allocations or stateful translation keyed by subscriber identity and proper port tracking.
  2. Use an appropriate NAT64/DNS64 service design or approved application gateway for IPv6-to-IPv4 reachability.
  3. Provide supported DNS64 synthesis or an equivalent discovery mechanism consistent with NAT64 deployment.
  4. Validate PIC support, alternate paths and recursion so the forwarding plane can preinstall backup resolution where supported.

Explanation

Correct Answer

C:

Provide supported DNS64 synthesis or an equivalent discovery mechanism consistent with NAT64 deployment. This directly addresses the fictional service-provider network case: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname. The technical mechanism is: NAT64 forwarding requires an IPv6 destination representation of the IPv4 server, frequently supplied by DNS64. Verify success by: A known IPv4-only test name yields a reachable synthesized AAAA result and end-to-end connectivity.

Incorrect Answers

A:

Use nonoverlapping deterministic port-set allocations or stateful translation keyed by subscriber identity and proper port tracking. This would respond to a DIFFERENT fictional service-provider network issue: An ISP assigns customer IPv4 address blocks through CGNAT but configures identical port-set mappings on two subscribers behind the same public address, causing connection collisions. It does not directly address the current situation: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname.

B:

Use an appropriate NAT64/DNS64 service design or approved application gateway for IPv6-to-IPv4 reachability. This would respond to a DIFFERENT fictional service-provider network issue: An IPv6-only enterprise application must reach an IPv4-only legacy service, but the provider installs NAT44 and expects it to translate the address families. It does not directly address the current situation: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname.

D:

Validate PIC support, alternate paths and recursion so the forwarding plane can preinstall backup resolution where supported. This would respond to a DIFFERENT fictional service-provider network issue: A provider enables BGP PIC in its core but expects instant convergence even though recursive alternate next hops were not installed for the affected VPN prefixes. It does not directly address the current situation: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname.

Question 522

After fixing the following network problem, what should the team verify: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname.

  1. An IPv6-only test client resolves and reaches the target IPv4-only service through the transition gateway.
  2. A lab tuple resolves to the correct subscriber session without conflating two simultaneous clients.
  3. Both routers report the intended adjacency level and the local area route becomes reachable.
  4. A known IPv4-only test name yields a reachable synthesized AAAA result and end-to-end connectivity.

Explanation

Correct Answer

D:

A known IPv4-only test name yields a reachable synthesized AAAA result and end-to-end connectivity. This confirms the approved response: Provide supported DNS64 synthesis or an equivalent discovery mechanism consistent with NAT64 deployment. It is relevant because: NAT64 forwarding requires an IPv6 destination representation of the IPv4 server, frequently supplied by DNS64.

Incorrect Answers

A:

An IPv6-only test client resolves and reaches the target IPv4-only service through the transition gateway. This would validate ANOTHER scenario: An IPv6-only enterprise application must reach an IPv4-only legacy service, but the provider installs NAT44 and expects it to translate the address families. The specific correction here is: Provide supported DNS64 synthesis or an equivalent discovery mechanism consistent with NAT64 deployment.

B:

A lab tuple resolves to the correct subscriber session without conflating two simultaneous clients. This would validate ANOTHER scenario: A CGNAT operator has millions of sessions but translation logs record only subscriber private addresses, making legal attribution impossible for shared public IPv4 connections. The specific correction here is: Provide supported DNS64 synthesis or an equivalent discovery mechanism consistent with NAT64 deployment.

C:

Both routers report the intended adjacency level and the local area route becomes reachable. This would validate ANOTHER scenario: A core interface comes up but its IS-IS hellos never result in adjacency because peer system IDs use incompatible area addressing on L1-only neighbors. The specific correction here is: Provide supported DNS64 synthesis or an equivalent discovery mechanism consistent with NAT64 deployment.

Question 523

A carrier root-cause review considers: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname. What underlying mechanism explains this?

  1. NAT64 forwarding requires an IPv6 destination representation of the IPv4 server, frequently supplied by DNS64.
  2. Shared-address NAT needs distinct translated transport identifiers to demultiplex flows accurately.
  3. Dual-Stack Lite transports IPv4 traffic over IPv6 access to a carrier AFTR performing shared IPv4 translation.
  4. OSPF neighbors require compatible hello/dead settings and authentication before upper-layer reachability can rely on them.

Explanation

Correct Answer

A:

NAT64 forwarding requires an IPv6 destination representation of the IPv4 server, frequently supplied by DNS64. This explains the investigated issue: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname. The approved corrective response is: Provide supported DNS64 synthesis or an equivalent discovery mechanism consistent with NAT64 deployment.

Incorrect Answers

B:

Shared-address NAT needs distinct translated transport identifiers to demultiplex flows accurately. That explains a DIFFERENT scenario: An ISP assigns customer IPv4 address blocks through CGNAT but configures identical port-set mappings on two subscribers behind the same public address, causing connection collisions. The mechanism to prioritize for this question is: NAT64 forwarding requires an IPv6 destination representation of the IPv4 server, frequently supplied by DNS64.

C:

Dual-Stack Lite transports IPv4 traffic over IPv6 access to a carrier AFTR performing shared IPv4 translation. That explains a DIFFERENT scenario: A customer uses DS-Lite but expects its home CPE to perform normal IPv4 public-NAT termination locally, and the B4 tunnel is never provisioned. The mechanism to prioritize for this question is: NAT64 forwarding requires an IPv6 destination representation of the IPv4 server, frequently supplied by DNS64.

D:

OSPF neighbors require compatible hello/dead settings and authentication before upper-layer reachability can rely on them. That explains a DIFFERENT scenario: Two OSPF routers connected through an L2VPN are configured with incompatible hello/dead timers and a monitoring dashboard calls the missing adjacency a BGP failure. The mechanism to prioritize for this question is: NAT64 forwarding requires an IPv6 destination representation of the IPv4 server, frequently supplied by DNS64.

Question 524

After addressing: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname. Which recurring service assurance check is most useful?

  1. Validate B4 reachability and AFTR capacity at subscriber activation.
  2. Include DNS synthesis and DNSSEC compatibility cases in NAT64 acceptance.
  3. Validate dual-family translation behavior before onboarding IPv6-only customers.
  4. Avoid using accidental router IDs as a business traffic-engineering policy.

Explanation

Correct Answer

B:

Include DNS synthesis and DNSSEC compatibility cases in NAT64 acceptance. This helps prevent recurrence of the issue: An IPv6-only access service installs NAT64 but does not provide a method to obtain suitable IPv6 destinations for IPv4-only DNS names, so subscribers fail by hostname. Acceptance evidence should be: A known IPv4-only test name yields a reachable synthesized AAAA result and end-to-end connectivity.

Incorrect Answers

A:

Validate B4 reachability and AFTR capacity at subscriber activation. That is a sensible control for ANOTHER scenario: A customer uses DS-Lite but expects its home CPE to perform normal IPv4 public-NAT termination locally, and the B4 tunnel is never provisioned. It is less targeted to this issue than: Include DNS synthesis and DNSSEC compatibility cases in NAT64 acceptance.

C:

Validate dual-family translation behavior before onboarding IPv6-only customers. That is a sensible control for ANOTHER scenario: An IPv6-only enterprise application must reach an IPv4-only legacy service, but the provider installs NAT44 and expects it to translate the address families. It is less targeted to this issue than: Include DNS synthesis and DNSSEC compatibility cases in NAT64 acceptance.

D:

Avoid using accidental router IDs as a business traffic-engineering policy. That is a sensible control for ANOTHER scenario: BGP flips between two equal candidates because a team ignores the router-id and neighbor-address tie-breakers after all higher-ranked attributes match. It is less targeted to this issue than: Include DNS synthesis and DNSSEC compatibility cases in NAT64 acceptance.

Question 525

A provider backbone escalation identifies: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss. What should the network team implement?

  1. Configure detection and multiplier values consistent with platform and WAN path characteristics.
  2. Validate supported ISSU procedure, image compatibility and redundancy state before proceeding with a staged maintenance plan.
  3. Use nonoverlapping deterministic port-set allocations or stateful translation keyed by subscriber identity and proper port tracking.
  4. Trace shared-risk link groups and relocate one service path to independent optical facilities where required.

Explanation

Correct Answer

D:

Trace shared-risk link groups and relocate one service path to independent optical facilities where required. This directly addresses the fictional service-provider network case: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss. The technical mechanism is: Logical ECMP diversity does not ensure physical failure independence when transport infrastructure shares common risk. Verify success by: A failure drill on one shelf leaves at least one approved end-to-end route available.

Incorrect Answers

A:

Configure detection and multiplier values consistent with platform and WAN path characteristics. This would respond to a DIFFERENT fictional service-provider network issue: A BFD session on a long-haul link uses submillisecond detection and repeatedly declares faults during normal microbursts despite healthy optics. It does not directly address the current situation: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss.

B:

Validate supported ISSU procedure, image compatibility and redundancy state before proceeding with a staged maintenance plan. This would respond to a DIFFERENT fictional service-provider network issue: A dual-supervisor router reports standby Hot but an engineer schedules an in-service upgrade without checking platform release interoperability and ISSU prerequisites. It does not directly address the current situation: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss.

C:

Use nonoverlapping deterministic port-set allocations or stateful translation keyed by subscriber identity and proper port tracking. This would respond to a DIFFERENT fictional service-provider network issue: An ISP assigns customer IPv4 address blocks through CGNAT but configures identical port-set mappings on two subscribers behind the same public address, causing connection collisions. It does not directly address the current situation: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss.

Question 526

A service-provider operations test targets: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss. Which result is required?

  1. A failure drill on one shelf leaves at least one approved end-to-end route available.
  2. Disconnecting the access link leaves a genuinely diverse next hop forwarding test traffic.
  3. The laboratory upgrade keeps the intended traffic classes forwarding while both supervisors reach the approved release.
  4. Route calculation shifts the test prefix to the expected lower-cost path without interrupting control adjacencies.

Explanation

Correct Answer

A:

A failure drill on one shelf leaves at least one approved end-to-end route available. This confirms the approved response: Trace shared-risk link groups and relocate one service path to independent optical facilities where required. It is relevant because: Logical ECMP diversity does not ensure physical failure independence when transport infrastructure shares common risk.

Incorrect Answers

B:

Disconnecting the access link leaves a genuinely diverse next hop forwarding test traffic. This would validate ANOTHER scenario: A network uses two next hops with ECMP but one backup route resolves through the same failed access link and the team reports full path redundancy. The specific correction here is: Trace shared-risk link groups and relocate one service path to independent optical facilities where required.

C:

The laboratory upgrade keeps the intended traffic classes forwarding while both supervisors reach the approved release. This would validate ANOTHER scenario: A dual-supervisor router reports standby Hot but an engineer schedules an in-service upgrade without checking platform release interoperability and ISSU prerequisites. The specific correction here is: Trace shared-risk link groups and relocate one service path to independent optical facilities where required.

D:

Route calculation shifts the test prefix to the expected lower-cost path without interrupting control adjacencies. This would validate ANOTHER scenario: An IS-IS L2 router forms adjacency but traffic chooses a high-latency optical link because the interface metric was left at a low default. The specific correction here is: Trace shared-risk link groups and relocate one service path to independent optical facilities where required.

Question 527

A provider troubleshooting note states: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss. Which mechanism is central?

  1. Equal-cost routes can conceal a common downstream failure when their recursive forwarding dependencies overlap.
  2. Logical ECMP diversity does not ensure physical failure independence when transport infrastructure shares common risk.
  3. Link aggregation requires compatible control negotiation and consistent member roles to share forwarding correctly.
  4. OSPF area-type agreement is required for adjacency and stub routing behavior.

Explanation

Correct Answer

B:

Logical ECMP diversity does not ensure physical failure independence when transport infrastructure shares common risk. This explains the investigated issue: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss. The approved corrective response is: Trace shared-risk link groups and relocate one service path to independent optical facilities where required.

Incorrect Answers

A:

Equal-cost routes can conceal a common downstream failure when their recursive forwarding dependencies overlap. That explains a DIFFERENT scenario: A network uses two next hops with ECMP but one backup route resolves through the same failed access link and the team reports full path redundancy. The mechanism to prioritize for this question is: Logical ECMP diversity does not ensure physical failure independence when transport infrastructure shares common risk.

C:

Link aggregation requires compatible control negotiation and consistent member roles to share forwarding correctly. That explains a DIFFERENT scenario: A provider aggregates two PE uplinks in LACP but one port belongs to a different bundle ID, creating unpredictable blackholes for hashed subscriber flows. The mechanism to prioritize for this question is: Logical ECMP diversity does not ensure physical failure independence when transport infrastructure shares common risk.

D:

OSPF area-type agreement is required for adjacency and stub routing behavior. That explains a DIFFERENT scenario: An OSPF stub area loses its expected default route when an engineer configures stub mode on the access router but not consistently on the area border router. The mechanism to prioritize for this question is: Logical ECMP diversity does not ensure physical failure independence when transport infrastructure shares common risk.

Question 528

To prevent recurrence of this provider network issue: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss. Which control should be retained?

  1. Verify port-channel membership and individual-link counters after physical changes.
  2. Maintain release-specific ISSU qualification and rollback documentation.
  3. Map provider optical shared-risk groups in high-availability path review.
  4. Verify v4 and v6 filter family independently during peering certification.

Explanation

Correct Answer

C:

Map provider optical shared-risk groups in high-availability path review. This helps prevent recurrence of the issue: An operator assumes two physically diverse core paths are independent, yet both circuits terminate in the same optical amplifier shelf and fail together during power loss. Acceptance evidence should be: A failure drill on one shelf leaves at least one approved end-to-end route available.

Incorrect Answers

A:

Verify port-channel membership and individual-link counters after physical changes. That is a sensible control for ANOTHER scenario: A provider aggregates two PE uplinks in LACP but one port belongs to a different bundle ID, creating unpredictable blackholes for hashed subscriber flows. It is less targeted to this issue than: Map provider optical shared-risk groups in high-availability path review.

B:

Maintain release-specific ISSU qualification and rollback documentation. That is a sensible control for ANOTHER scenario: A dual-supervisor router reports standby Hot but an engineer schedules an in-service upgrade without checking platform release interoperability and ISSU prerequisites. It is less targeted to this issue than: Map provider optical shared-risk groups in high-availability path review.

D:

Verify v4 and v6 filter family independently during peering certification. That is a sensible control for ANOTHER scenario: An IPv6 border router accepts an IPv6 customer route under an IPv4 route-policy address-family clause and quietly blocks legitimate IPv6 service NLRI. It is less targeted to this issue than: Map provider optical shared-risk groups in high-availability path review.

Question 529

At a managed VPN service carrier, an engineering review finds: An MPLS pseudowire intermittently drops large customer frames after VLAN tagging increases payload size beyond a transit link's supported frame budget. Which change most directly corrects the service problem?

  1. Align MPLS payload MTU and transport frame size, or adjust effective service MTU with the customer handoff contract.
  2. Interpret implicit-null advertisement as penultimate-hop popping behavior and verify the service's remaining labels.
  3. Validate transport versus service label stacking and make sure penultimate-hop popping applies only to the intended outer transport label.
  4. Include available shared-risk link group constraints or explicit path diversity in TE computation.

Explanation

Correct Answer

A:

Align MPLS payload MTU and transport frame size, or adjust effective service MTU with the customer handoff contract. This directly addresses the fictional service-provider network case: An MPLS pseudowire intermittently drops large customer frames after VLAN tagging increases payload size beyond a transit link's supported frame budget. The technical mechanism is: MPLS labels and service tags add encapsulation overhead that must fit within the backbone's effective MTU. Verify success by: A test at the required maximum service size passes without fragmentation-dependent loss or interface MTU errors.

Incorrect Answers

B:

Interpret implicit-null advertisement as penultimate-hop popping behavior and verify the service's remaining labels. This would respond to a DIFFERENT fictional service-provider network issue: The forwarding plane sees an implicit-null label for a PE loopback and a troubleshooter treats it as an invalid unassigned label rather than checking the upstream label disposition. It does not directly address the current situation: An MPLS pseudowire intermittently drops large customer frames after VLAN tagging increases payload size beyond a transit link's supported frame budget.

C:

Validate transport versus service label stacking and make sure penultimate-hop popping applies only to the intended outer transport label. This would respond to a DIFFERENT fictional service-provider network issue: A transit router keeps unexpectedly popping a provider's VPN label one hop before the target PE, sending customer traffic toward the global routing table. It does not directly address the current situation: An MPLS pseudowire intermittently drops large customer frames after VLAN tagging increases payload size beyond a transit link's supported frame budget.

D:

Include available shared-risk link group constraints or explicit path diversity in TE computation. This would respond to a DIFFERENT fictional service-provider network issue: A traffic engineering policy considers only link bandwidth when selecting primary and backup paths, but both use one optical shared-risk group. It does not directly address the current situation: An MPLS pseudowire intermittently drops large customer frames after VLAN tagging increases payload size beyond a transit link's supported frame budget.

Question 530

For a managed VPN service carrier, operators addressed: An MPLS pseudowire intermittently drops large customer frames after VLAN tagging increases payload size beyond a transit link's supported frame budget. What is the strongest acceptance evidence?

  1. A service prefix traverses the core layers with valid outer transport labels and correct VPN label disposition at the egress PE.
  2. A test at the required maximum service size passes without fragmentation-dependent loss or interface MTU errors.
  3. A maximal permitted label-stack test forwards across the planned platform while an unsupported stack is rejected during provisioning.
  4. A stress test admits or rejects reservations according to the per-class budget without starving the protected class.

Explanation

Correct Answer

B:

A test at the required maximum service size passes without fragmentation-dependent loss or interface MTU errors. This confirms the approved response: Align MPLS payload MTU and transport frame size, or adjust effective service MTU with the customer handoff contract. It is relevant because: MPLS labels and service tags add encapsulation overhead that must fit within the backbone's effective MTU.

Incorrect Answers

A:

A service prefix traverses the core layers with valid outer transport labels and correct VPN label disposition at the egress PE. This would validate ANOTHER scenario: A service provider exports VPNv4 labels across multiple autonomous regions but assumes LDP sessions must directly span every access network and interregion core. The specific correction here is: Align MPLS payload MTU and transport frame size, or adjust effective service MTU with the customer handoff contract.

C:

A maximal permitted label-stack test forwards across the planned platform while an unsupported stack is rejected during provisioning. This would validate ANOTHER scenario: A core node imposes two service labels on an access interface with a hardware label-stack depth limit below the engineered VPN service, causing selective drops. The specific correction here is: Align MPLS payload MTU and transport frame size, or adjust effective service MTU with the customer handoff contract.

D:

A stress test admits or rejects reservations according to the per-class budget without starving the protected class. This would validate ANOTHER scenario: Two DS-TE class types share one link, but one tunnel consumes the other class's reserved budget due to a misconfigured bandwidth constraint allocation. The specific correction here is: Align MPLS payload MTU and transport frame size, or adjust effective service MTU with the customer handoff contract.