Visit here for our full Juniper JN0-664 exam dumps and practice test questions.
Question 106
A service provider needs to implement MPLS Layer 3 VPN with route target filtering to reduce BGP routing overhead. Which BGP address family should be configured on the route reflector?
A) inet-vpn
B) inet-unicast
C) route-target
D) inet-mvpn
Answer: C
Explanation:
The route-target address family must be configured on route reflectors to enable route target constrained route distribution in MPLS Layer 3 VPN environments. This feature significantly reduces BGP routing overhead by ensuring that PE routers only receive VPN routes that match their locally configured route targets. Route target filtering prevents unnecessary route advertisements across the MPLS network.
When route-target address family is enabled, PE routers advertise their interested route targets to the route reflector using BGP. The route reflector maintains this information and uses it to filter VPN-IPv4 routes before reflecting them to other PE routers. This constraint-based route distribution is defined in RFC 4684 and is particularly beneficial in large-scale service provider networks with thousands of VPN instances.
Without route target filtering, route reflectors would reflect all VPN routes to all PE routers regardless of whether those routes are needed. In large deployments with hundreds of VPNs and thousands of routes, this creates significant memory consumption and processing overhead on PE routers. Route target filtering ensures that each PE router maintains only the routes relevant to its locally configured VPN instances.
The inet-vpn address family carries the actual VPN routes but does not provide the filtering mechanism itself. The inet-unicast address family is used for standard IPv4 routing and is not applicable to VPN scenarios. The inet-mvpn address family supports multicast VPN implementations and serves a different purpose than route target filtering for unicast VPN routes.
Implementing route-target address family requires configuration on both route reflectors and PE routers. The route reflector must support this capability and PE routers must advertise their route target membership information. This creates an efficient routing infrastructure that scales well as the number of VPNs and routes increases.
Question 107
What is the primary purpose of the BGP Graceful Restart helper mode in Junos OS?
A) To restart the local BGP process without affecting peers
B) To maintain forwarding during a peer’s BGP restart
C) To synchronize routing tables after link failure
D) To prevent routing loops during convergence
Answer: B
Explanation:
BGP Graceful Restart helper mode enables a router to maintain forwarding based on stale routes while a BGP peer undergoes a planned or unplanned restart of its control plane. The helper router preserves the forwarding state for routes learned from the restarting peer, preventing traffic disruption during the restart period. This capability is essential for maintaining high availability in service provider networks.
When a BGP peer signals its Graceful Restart capability and then experiences a control plane restart, the helper router marks the routes from that peer as stale but continues using them for forwarding. The helper maintains these routes for a configurable restart time period, typically 120 seconds. During this window, the restarting router can re-establish the BGP session and re-advertise its routes without causing a complete withdrawal and re-advertisement cycle.
The helper mode operation is particularly important in scenarios where the data plane remains operational even though the control plane has restarted. This separation of control and forwarding planes is common in modern router architectures with redundant routing engines or with processes running on separate control cards. Without Graceful Restart helper support, all routes from the restarting peer would be immediately withdrawn, causing traffic loss.
Local BGP process restart is handled by the Graceful Restart capability on the local router itself, not by helper mode. Routing table synchronization and loop prevention are handled through different BGP mechanisms such as BGP synchronization rules and AS-path loop detection. Helper mode specifically addresses the scenario where a peer is restarting and the local router needs to assist in maintaining service continuity.
Configuration of helper mode involves enabling graceful-restart on BGP sessions and tuning parameters such as restart time and stale routes time to match network requirements and convergence expectations.
Question 108
A network engineer is troubleshooting OSPF adjacency issues between two routers. The interface MTU is 1500 on one router and 9000 on the other. What will be the result?
A) Adjacency will form normally
B) Adjacency will remain in EXSTART state
C) Adjacency will remain in INIT state
D) Adjacency will remain in LOADING state
Answer: B
Explanation:
When OSPF routers have mismatched MTU values on their interfaces, the adjacency will remain stuck in the EXSTART state. This occurs because OSPF includes the interface MTU value in Database Description packets during the adjacency formation process. Routers compare the received MTU with their local interface MTU, and if there is a mismatch, they cannot proceed beyond the EXSTART state to exchange link-state databases.
The EXSTART state is where OSPF routers establish a master-slave relationship and prepare to exchange Database Description packets. During this phase, routers send DBD packets containing their MTU value. When a router receives a DBD packet with an MTU that exceeds its own interface MTU, it recognizes that full database exchange may not be possible due to packet fragmentation issues and refuses to proceed.
This MTU check is a safety mechanism to prevent database exchange problems. If routers with different MTUs were allowed to form full adjacencies, they might encounter issues when exchanging larger LSAs that exceed the smaller MTU value. These LSAs could be fragmented or dropped, leading to incomplete database synchronization and potential routing issues.
The INIT state occurs earlier in the adjacency process when a router sees its own Router ID in the neighbor’s Hello packet. The LOADING state comes after successful DBD exchange when routers are requesting specific LSAs. Neither of these states is affected by MTU mismatches. The adjacency will not form normally because OSPF specifically checks for MTU consistency before allowing full adjacency.
To resolve MTU mismatch issues, administrators should either configure consistent MTU values on both interfaces or disable the MTU check using the mtu-ignore command in Junos OS, though the latter approach should be used cautiously as it bypasses this important safety check.
Question 109
Which BGP attribute is used to prevent routing loops in an AS confederation?
A) AS_PATH
B) AS_CONFED_SEQUENCE
C) NEXT_HOP
D) LOCAL_PREF
Answer: B
Explanation:
The AS_CONFED_SEQUENCE attribute is specifically designed to prevent routing loops within BGP confederations. When BGP routes traverse confederation sub-autonomous systems, each sub-AS prepends its confederation identifier to the AS_CONFED_SEQUENCE attribute. This creates a record of the confederation path that routers use to detect and prevent loops within the confederation structure.
BGP confederations divide a large autonomous system into multiple sub-autonomous systems to improve scalability and manageability. Within the confederation, these sub-AS numbers are used for internal routing, but to the external BGP world, the entire confederation appears as a single AS. The AS_CONFED_SEQUENCE attribute tracks the internal confederation path without exposing this internal structure to external peers.
When a BGP speaker receives a route with an AS_CONFED_SEQUENCE that contains its own confederation sub-AS number, it recognizes this as a potential loop and rejects the route. This mechanism works similarly to how the standard AS_PATH attribute prevents loops between different autonomous systems, but it operates specifically within the confederation boundary.
The regular AS_PATH attribute tracks the path through standard autonomous systems and does provide loop prevention for external BGP, but it does not handle the internal confederation structure. The NEXT_HOP attribute indicates the next hop IP address for reaching advertised routes but does not participate in loop prevention. LOCAL_PREF influences route selection within an AS but has no loop prevention functionality.
In addition to AS_CONFED_SEQUENCE, confederations also use the AS_CONFED_SET attribute when route aggregation occurs within the confederation. Together, these attributes ensure proper loop prevention while maintaining the benefits of confederation-based scaling for large autonomous systems. Understanding these attributes is crucial for designing and troubleshooting confederation deployments.
Question 110
A service provider is implementing multicast using PIM Sparse Mode. What is the function of the Rendezvous Point in this deployment?
A) To forward all multicast traffic in the network
B) To act as a meeting point for sources and receivers
C) To maintain the multicast routing table for all routers
D) To provide backup paths for multicast traffic
Answer: B
Explanation:
The Rendezvous Point acts as a meeting point where multicast sources and receivers can find each other in PIM Sparse Mode deployments. When a receiver wants to join a multicast group, it sends PIM Join messages toward the RP for that group. Similarly, when a source begins sending multicast traffic, the first-hop router encapsulates the traffic and sends it to the RP. The RP then forwards the traffic down the shared tree to all receivers.
In PIM Sparse Mode, the RP eliminates the need for multicast sources and receivers to know about each other directly. This shared tree approach is particularly efficient when sources and receivers are distributed across the network. The RP maintains state information about active sources and interested receivers for each multicast group it serves, enabling it to properly route multicast traffic.
The RP’s role can be optimized through a process called SPT switchover. After receivers begin receiving traffic through the RP via the shared tree, the last-hop routers can choose to switch to a source-specific shortest path tree. This optimization reduces latency and bandwidth consumption by creating a more direct path from source to receivers, though the RP still plays a role in the initial discovery process.
The RP does not forward all multicast traffic permanently, especially after SPT switchover occurs. It does not maintain routing tables for all routers but rather maintains group-specific state information. While the RP is critical, backup mechanisms like Anycast RP or Auto-RP provide redundancy rather than the RP itself providing backup paths.
Multiple RPs can be deployed in a network for redundancy and load distribution. Mechanisms such as Bootstrap Router protocol or static RP configuration ensure that all routers in the PIM domain know which RP to use for each multicast group range.
Question 111
What is the default BGP keepalive interval in Junos OS?
A) 30 seconds
B) 60 seconds
C) 90 seconds
D) 120 seconds
Answer: B
Explanation:
The default BGP keepalive interval in Junos OS is 60 seconds. This timer determines how frequently a BGP router sends keepalive messages to its peers to maintain the BGP session. Regular keepalive messages inform the peer that the BGP session is still active and the sender is still operational. This mechanism is essential for detecting failed peers and maintaining reliable BGP peering relationships.
The keepalive interval works in conjunction with the hold timer, which has a default value of 180 seconds in Junos OS. The hold timer must be at least three times the keepalive interval to allow for potential packet loss or temporary network issues without prematurely tearing down the BGP session. If a router does not receive any BGP messages from its peer within the hold time period, it declares the peer down and removes all routes learned from that peer.
BGP negotiates both keepalive and hold timers during session establishment. Each router proposes its configured hold time in the BGP OPEN message, and both peers use the smaller of the two proposed values. The keepalive interval is then automatically set to one-third of the negotiated hold time. This negotiation ensures compatibility between routers with different timer configurations.
The 30-second value represents the keepalive interval when the hold time is reduced to 90 seconds. While 90 and 120 seconds can be configured as hold times, they do not represent the default keepalive interval. Administrators can adjust these timers to match specific network requirements, with lower values providing faster failure detection but generating more control plane traffic.
Understanding BGP timers is important for troubleshooting session stability issues and optimizing convergence times in service provider networks. In environments requiring faster convergence, administrators might configure more aggressive timers, while stable environments might use longer intervals to reduce control plane overhead.
Question 112
A network administrator needs to implement IS-IS route leaking between Level 1 and Level 2 areas. Which configuration statement enables this functionality?
A) wide-metrics-only
B) level 2 disable
C) no-attached-bit
D) interface-level 1-2
Answer: C
Explanation:
The no-attached-bit configuration statement is used in conjunction with route leaking configurations to enable the advertisement of specific Level 2 routes into Level 1 areas in IS-IS. By default, Level 1-2 routers set the attached bit in their Level 1 LSPs to indicate they have connectivity to the Level 2 backbone. Level 1 routers use this information to forward traffic destined for other areas toward the nearest Level 1-2 router. However, this creates a default route behavior rather than specific route visibility.
Route leaking allows specific Level 2 routes to be advertised into Level 1 areas, providing more granular routing information and enabling better path selection. The no-attached-bit statement can be configured on Level 1-2 routers to suppress the attached bit, which is often done when implementing controlled route leaking to prevent default routing behavior that might conflict with the specific leaked routes.
The actual route leaking is configured using export policies that redistribute routes between levels. These policies specify which Level 2 routes should be advertised into Level 1 areas, allowing administrators to selectively control information flow. This is particularly useful in large IS-IS deployments where complete route visibility between levels is needed for specific prefixes.
The wide-metrics-only statement forces the use of wide metric TLVs and does not directly enable route leaking. The level 2 disable command would prevent Level 2 operation entirely, which contradicts the goal of inter-level communication. The interface-level 1-2 statement configures an interface to participate in both Level 1 and Level 2, but it does not control route leaking behavior.
Route leaking must be carefully planned to avoid routing loops and suboptimal routing. Proper policy configuration with appropriate route filtering and metric manipulation ensures that leaked routes provide the intended benefits without introducing problems.
Question 113
Which MPLS traffic engineering constraint can be configured to ensure bandwidth availability along the LSP path?
A) Administrative groups
B) Bandwidth reservation
C) Hop limit
D) Priority levels
Answer: B
Explanation:
Bandwidth reservation is the MPLS traffic engineering constraint that ensures bandwidth availability along the Label Switched Path. When configuring an LSP, administrators can specify the required bandwidth, and the Constrained Shortest Path First algorithm calculates a path that has sufficient available bandwidth on all links along the route. This reservation mechanism guarantees that the LSP has the necessary resources to meet its traffic requirements.
The bandwidth reservation process involves the Resource Reservation Protocol with Traffic Engineering extensions. When establishing an LSP, RSVP-TE signals the bandwidth requirement to each router along the path. Each router checks if the requested bandwidth is available on the outgoing interface and, if so, reserves that bandwidth for the LSP. This prevents over-subscription and ensures quality of service for traffic using the LSP.
Reserved bandwidth is tracked separately from best-effort traffic bandwidth. Routers maintain information about total interface bandwidth, reserved bandwidth for all LSPs, and available bandwidth for new reservations. This accounting allows CSPF to calculate paths that meet bandwidth requirements while ensuring that existing LSPs continue to have their guaranteed resources.
Administrative groups (also called affinity or link coloring) provide a way to classify links and constrain LSP paths based on link attributes, but they do not directly ensure bandwidth availability. Hop limit restricts the maximum number of hops an LSP can traverse but does not guarantee bandwidth. Priority levels determine which LSPs can preempt others during resource contention but are not the primary mechanism for ensuring bandwidth availability.
Proper bandwidth reservation configuration requires understanding traffic patterns, over-subscription ratios, and failure scenarios. Administrators must balance between reserving sufficient bandwidth for guaranteed service and maintaining flexibility for best-effort traffic and future LSP establishments.
Question 114
A service provider needs to implement seamless MPLS to interconnect multiple IGP domains. Which protocol is used for inter-domain label distribution?
A) LDP
B) RSVP-TE
C) BGP with labeled-unicast
D) IS-IS
Answer: C
Explanation:
BGP with labeled-unicast address family is the protocol used for inter-domain label distribution in seamless MPLS architectures. This approach enables MPLS connectivity across multiple IGP domains without requiring a single unified IGP across the entire network. BGP labeled-unicast carries both IP prefix information and associated MPLS labels, allowing for end-to-end label switched paths spanning multiple administrative or IGP boundaries.
In seamless MPLS deployments, each IGP domain operates independently with its own label distribution protocol such as LDP or RSVP-TE for intra-domain connectivity. At the domain boundaries, Autonomous System Boundary Routers or Area Border Routers use BGP labeled-unicast to advertise loopback addresses of PE routers along with MPLS labels. This creates a hierarchical MPLS architecture where intra-domain LSPs are stitched together with inter-domain labeled paths.
The seamless MPLS architecture is particularly valuable for large service provider networks that need to maintain multiple IGP domains for scalability or administrative reasons. BGP labeled-unicast provides the scalability benefits of BGP while enabling MPLS forwarding across domain boundaries. This eliminates the need for complex label distribution schemes or MPLS-in-IP tunneling between domains.
LDP and RSVP-TE are typically used within IGP domains but do not extend across domain boundaries in seamless MPLS designs. IS-IS is an interior gateway protocol that provides routing information but does not distribute MPLS labels between domains. BGP labeled-unicast specifically addresses the inter-domain requirement by combining BGP’s ability to scale across domains with MPLS label distribution.
Implementation requires careful planning of BGP route reflector hierarchies, label allocation strategies, and next-hop self policies to ensure proper label stack construction for end-to-end connectivity across all domains.
Question 115
What is the purpose of the BGP ADD-PATH capability?
A) To advertise multiple AS paths for redundancy
B) To allow multiple paths for the same prefix to be advertised
C) To enable load balancing across BGP peers
D) To reduce BGP convergence time during failures
Answer: B
Explanation:
The BGP ADD-PATH capability allows a BGP speaker to advertise multiple paths for the same prefix to its peers, rather than advertising only the single best path as traditional BGP does. This capability significantly improves path diversity in BGP networks by enabling route reflectors and other BGP speakers to share multiple viable paths with their peers, allowing those peers to make better local routing decisions.
In traditional BGP operation, a router selects a single best path using the BGP decision process and advertises only that path to its peers. This behavior can hide alternate paths that might be useful for redundancy or load balancing. With ADD-PATH enabled, a router can advertise multiple paths distinguished by unique path identifiers, allowing receiving routers to maintain and potentially use these alternate paths.
The ADD-PATH capability is negotiated during BGP session establishment. Each path advertisement includes a path identifier that distinguishes it from other paths for the same prefix. Receiving routers can then maintain multiple paths in their BGP table and use them according to configured policies. This is particularly beneficial in route reflector designs where path diversity can be significantly reduced.
ADD-PATH does not advertise multiple AS paths for the same route, as each path still contains its own AS-PATH attribute. While it can enable better load balancing by making multiple paths available, the actual load balancing decisions are made by separate mechanisms. Although ADD-PATH can improve convergence by maintaining pre-computed backup paths, reducing convergence time is not its primary purpose.
Implementation considerations include memory consumption for maintaining additional paths, BGP update message overhead for advertising multiple paths, and policy configuration to control which additional paths are advertised and accepted. Proper deployment requires understanding these tradeoffs and the specific path diversity needs.
Question 116
Which command displays the active and inactive BGP routes in Junos OS?
A) show route protocol bgp
B) show bgp summary
C) show route receive-protocol bgp
D) show bgp neighbor
Answer: A
Explanation:
The show route protocol bgp command displays both active and inactive BGP routes in the Junos routing table. This command shows all routes learned via BGP, with active routes marked with an asterisk and inactive routes shown without the asterisk. This comprehensive view helps administrators understand which BGP routes are being used for forwarding and which routes are available but not selected.
Active routes are those selected by the BGP decision process and installed in the forwarding table for packet forwarding. Inactive routes are valid BGP routes that were not selected as the best path due to BGP path selection criteria such as local preference, AS path length, origin type, or other attributes. Understanding why routes are inactive is crucial for troubleshooting routing issues and optimizing path selection.
The command output includes detailed information such as destination prefixes, next hops, preferences, metrics, AS paths, and route attributes. The display clearly distinguishes between routes from different BGP peers and shows multiple paths when available. Additional output includes communities, local preference values, and MED values that influenced the path selection decision.
The show bgp summary command provides a high-level overview of BGP peer status including session state and route counts but does not display individual routes. The show route receive-protocol bgp command shows routes received from a specific peer before any import policy processing, which is useful for troubleshooting but does not show the final active/inactive status. The show bgp neighbor command displays detailed information about BGP peer relationships but not route details.
For more specific troubleshooting, administrators can use additional options such as show route protocol bgp detail for comprehensive route attributes or show route protocol bgp extensive for maximum detail including hidden attributes and path selection reasons.
Question 117
A network engineer is configuring VPLS and needs to ensure loop prevention in the Layer 2 network. Which mechanism is used?
A) Spanning Tree Protocol
B) Split horizon for pseudowires
C) MAC address learning limits
D) BPDU filtering
Answer: B
Explanation:
Split horizon for pseudowires is the primary loop prevention mechanism used in Virtual Private LAN Service deployments. This mechanism prevents frames received on one pseudowire from being forwarded out another pseudowire on the same PE router. Split horizon ensures that traffic entering the VPLS instance through a pseudowire can only be forwarded to local attachment circuits, not to other pseudowires, effectively preventing loops in the VPLS full mesh topology.
VPLS creates a multipoint Layer 2 VPN service over an MPLS network by establishing a full mesh of pseudowires between all PE routers participating in the VPLS instance. Without split horizon, a frame could potentially loop indefinitely through the pseudowire mesh. When a PE router receives a frame on a pseudowire, split horizon rules prevent it from forwarding that frame to any other pseudowire, ensuring loop-free operation.
The split horizon mechanism is built into the VPLS forwarding logic and operates automatically without requiring explicit configuration. It works in conjunction with MAC address learning, where PE routers learn which MAC addresses are reachable through which pseudowires or attachment circuits. This combination of split horizon and MAC learning creates a loop-free Layer 2 forwarding environment across the MPLS core.
Spanning Tree Protocol can be used on the attachment circuits connecting customer equipment to the PE routers, but it is not used for loop prevention within the VPLS pseudowire mesh. MAC address learning limits provide protection against MAC table overflow attacks but do not prevent forwarding loops. BPDU filtering prevents processing of spanning tree BPDUs and is not a loop prevention mechanism for VPLS pseudowires.
Understanding split horizon operation is essential for troubleshooting VPLS connectivity issues, as misconfigurations or equipment failures that bypass split horizon rules can lead to broadcast storms and network instability.
Question 118
What is the function of the BGP route reflector cluster ID?
A) To identify the AS number of the route reflector
B) To prevent routing loops within the cluster
C) To establish priority among multiple route reflectors
D) To encrypt BGP updates within the cluster
Answer: B
Explanation:
The BGP route reflector cluster ID is used to prevent routing loops when BGP routes are reflected within and between route reflector clusters. When a route reflector reflects a route to its clients, it adds the cluster ID to the CLUSTER_LIST attribute in the BGP update. If a route reflector receives an update containing its own cluster ID in the CLUSTER_LIST, it recognizes this as a potential loop and discards the update.
Route reflectors provide a scalable alternative to full-mesh iBGP by allowing certain routers to reflect routes between BGP peers without requiring every router to peer with every other router. The cluster ID identifies a group of route reflectors and their clients as a single cluster. By default, the cluster ID is set to the router ID of the route reflector, but it can be manually configured when multiple route reflectors serve the same set of clients.
When multiple route reflectors serve the same clients for redundancy, they should be configured with the same cluster ID. This configuration tells each route reflector that they are part of the same cluster and prevents them from reflecting routes between each other’s clients unnecessarily. Without proper cluster ID configuration, routing information might not propagate correctly or could create unintended routing loops.
The cluster ID does not identify the AS number, as the AS number is a separate attribute carried in AS_PATH. It does not establish priority among route reflectors, as all route reflectors in a cluster are considered equal. BGP route encryption is handled by separate mechanisms such as TCP MD5 authentication or IPsec and is not related to cluster ID functionality.
Proper cluster ID configuration is critical in complex route reflector hierarchies where multiple levels of reflection exist. Understanding cluster ID operation helps in designing scalable BGP networks and troubleshooting route propagation issues in route reflector deployments.
Question 119
Which IS-IS LSP type is used to advertise IPv6 reachability information?
A) TLV 128
B) TLV 129
C) TLV 135
D) TLV 236
Answer: D
Explanation:
TLV 236 is the IS-IS Type-Length-Value field used to advertise IPv6 reachability information in IS-IS Link State PDUs. This TLV, formally known as IPv6 Reachability, carries IPv6 prefix information including the prefix, prefix length, metric, and various sub-TLVs for additional attributes. Support for TLV 236 enables IS-IS to function as an IPv6 routing protocol while maintaining its existing IPv4 routing capabilities.
IS-IS was originally designed as a protocol-independent routing protocol, making it well-suited for supporting multiple network layer protocols. The addition of IPv6-specific TLVs allows IS-IS to carry IPv6 routing information without fundamental changes to the protocol structure. TLV 236 can carry both internal and external IPv6 routes, along with route preferences and metrics used for path selection.
When IS-IS is configured for IPv6 routing, routers originate LSPs containing TLV 236 to advertise their connected IPv6 prefixes and learned routes. Other routers in the IS-IS domain receive these LSPs, extract the IPv6 reachability information, and use it to build their IPv6 routing tables. The SPF algorithm operates on this information to calculate shortest paths to IPv6 destinations.
TLV 128 is used for IP internal reachability information for IPv4. TLV 129 carries protocols supported information. TLV 135 is used for extended IP reachability and supports wide metrics for IPv4 routes but does not handle IPv6. These TLVs are part of the original IS-IS specification or early extensions and do not provide IPv6 support.
Modern IS-IS implementations support both IPv4 and IPv6 simultaneously, often called dual-stack IS-IS. This capability allows service providers to migrate to IPv6 while maintaining IPv4 connectivity using a single routing protocol instance, simplifying network operations and reducing protocol overhead.
Question 120
A service provider is implementing MPLS Fast Reroute using facility backup. What is protected by this mechanism?
A) Individual LSPs only
B) Links and nodes along the LSP path
C) The ingress PE router
D) The egress PE router
Answer: B
Explanation:
MPLS Fast Reroute using facility backup protects both links and nodes along the Label Switched Path by creating bypass tunnels that can quickly reroute traffic around failures. Facility backup provides scalable protection by using a single bypass LSP to protect multiple primary LSPs that traverse the same network element. This approach significantly reduces the number of backup paths needed compared to one-to-one protection schemes.
In facility backup, routers create bypass LSPs that route around potential failure points before any failures occur. These bypass LSPs can protect against link failures by routing around a specific link, or they can protect against node failures by routing around both a node and its attached links. When a failure is detected, traffic from all affected primary LSPs is immediately switched to the appropriate bypass LSP, typically achieving sub-50 millisecond convergence.
Link protection bypass LSPs terminate at the next-hop router beyond the protected link, providing an alternate path if that specific link fails. Node protection bypass LSPs go further by terminating at the next-next-hop router, protecting against failure of both the link and the downstream node. Node protection provides more comprehensive protection but requires more complex path calculation and may consume more network resources.
Facility backup does not protect only individual LSPs, as one bypass LSP protects multiple primary LSPs simultaneously. The ingress PE router is not protected by facility backup mechanisms along the LSP path, as protection begins at the first hop. Similarly, the egress PE router is the ultimate destination and is not protected by intermediate fast reroute mechanisms. Protection of ingress and egress routers requires different high availability mechanisms.
Implementing facility backup requires careful network design to ensure bypass paths have sufficient bandwidth and meet traffic engineering constraints. Administrators must balance protection coverage, resource utilization, and operational complexity when designing fast reroute protection schemes.