JNCIA MPLS and VPNs 2 — Questions and Answers

Question 1: In Junos L3VPN, what is the function of a Route Target (RT)?

Controls which VRFs import which VPNv4 routes, implementing the VPN topology (full-mesh, hub-spoke, extranet) (Correct answer)
Uniquely identifies each VPN route in the global BGP table
Specifies the MPLS label value for the VPN
Sets the routing metric within the VPN

Correct answer: Controls which VRFs import which VPNv4 routes, implementing the VPN topology (full-mesh, hub-spoke, extranet)

Route Targets are BGP extended communities that control VRF route import and export. A VRF exports routes with certain RT values; other VRFs import routes matching specific RT values, defining VPN connectivity topology.

RT implementation: PE1's VRF exports customer routes with RT 65000:100; PE2's VRF has import RT 65000:100, so it imports these routes into the customer's VRF. Full-mesh topology: all VRFs use the same export and import RT. Hub-spoke: hub VRF has different import/export RTs from spoke VRFs to enforce traffic flows through hub. Extranet: shared service VRF exports with RT that multiple customer VRFs import. In Junos: 'set routing-instances VRF vrf-import IMPORT-POLICY' (using policy with community match) or 'set routing-instances VRF vrf-target target:65000:100'.

Question 2: What is VPLS (Virtual Private LAN Service) in Junos?

A Layer 2 VPN service that provides multipoint Ethernet connectivity across an MPLS network, emulating a LAN switch for geographically dispersed sites (Correct answer)
A Layer 3 routing-based VPN similar to L3VPN
A point-to-point Layer 2 circuit replacement for leased lines
A Juniper-proprietary VPN protocol

Correct answer: A Layer 2 VPN service that provides multipoint Ethernet connectivity across an MPLS network, emulating a LAN switch for geographically dispersed sites

VPLS emulates a LAN (with MAC learning, flooding, and bridging) across an MPLS backbone, allowing multiple customer sites to appear as if they are connected to the same Ethernet switch.

VPLS (RFC 4762 for LDP, RFC 4761 for BGP) creates a logical bridge domain across the MPLS network. Each VPLS instance on a PE has a virtual switch instance (VSI) that learns customer MAC addresses. Traffic is forwarded by MAC lookup; unknown unicast and broadcast are flooded to all PE pseudowires. Two signaling methods: LDP-based (RFC 4762) — uses LDP targeted sessions between PEs; BGP-based (RFC 4761) — uses MP-BGP with L2VPN NLRI. In Junos: routing-instance type 'vpls' under [edit routing-instances]. EVPN is the modern replacement.

Question 3: How does Junos implement MPLS traffic engineering (TE)?

Using RSVP-TE to signal LSPs with explicit routes and bandwidth reservations, with OSPF or IS-IS TE extensions advertising link bandwidth (Correct answer)
Using BGP to distribute MPLS labels with bandwidth constraints
Through static MPLS label assignments with manual path configuration
Traffic engineering is handled automatically by LDP

Correct answer: Using RSVP-TE to signal LSPs with explicit routes and bandwidth reservations, with OSPF or IS-IS TE extensions advertising link bandwidth

Junos MPLS TE uses RSVP-TE for LSP signaling with constraint-based routing. OSPF-TE or IS-IS-TE extensions flood link attributes (bandwidth, TE metric, SRLG) that CSPF uses to compute explicit paths.

MPLS TE in Junos: (1) Enable TE extensions: 'set protocols ospf traffic-engineering' or 'set protocols isis traffic-engineering'; (2) Enable RSVP: 'set protocols rsvp interface ge-x/x/x'; (3) Enable MPLS: 'set protocols mpls interface ge-x/x/x'; (4) Configure LSP: 'set protocols mpls label-switched-path LSP-NAME to [destination] bandwidth [bps] primary PATH-NAME'. CSPF (Constrained Shortest Path First) computes the path respecting bandwidth and administrative constraints. 'show rsvp session' and 'show mpls lsp' verify LSP status.

Question 4: What is 'fast reroute' (FRR) in Junos MPLS?

A protection mechanism that pre-computes backup paths for LSPs, enabling sub-50ms traffic rerouting upon link or node failure without waiting for IGP reconvergence (Correct answer)
A method to quickly re-establish failed BGP sessions
A feature that automatically reroutes traffic based on load
A rapid OSPF reconvergence technique

Correct answer: A protection mechanism that pre-computes backup paths for LSPs, enabling sub-50ms traffic rerouting upon link or node failure without waiting for IGP reconvergence

MPLS FRR pre-establishes detour or bypass LSPs before failures occur. When a protected link or node fails, traffic is immediately switched to the pre-computed backup path, typically within 50ms.

MPLS FRR (RFC 4090) provides two protection models: Facility Backup (bypass LSP) — a single bypass LSP protects a link or node for multiple primary LSPs passing through; One-to-One Backup (detour LSP) — each primary LSP has its own dedicated detour LSP. In Junos, FRR is enabled on LSPs with 'set protocols mpls label-switched-path LSP-NAME fast-reroute'. The PLR (Point of Local Repair) detects failure (via RSVP Hello or IGP) and immediately switches traffic to the pre-established bypass. MPLS FRR is the gold standard for carrier-grade protection switching.

Question 5: In Junos, what command shows the MPLS label forwarding table?

show route table mpls.0 (Correct answer)
show mpls labels
show forwarding-table mpls
show label-table

Correct answer: show route table mpls.0

'show route table mpls.0' displays the MPLS label table (mpls.0 routing table) showing incoming labels, operations (pop/swap/push), and next-hops for each MPLS label binding.

'show route table mpls.0' displays MPLS label bindings: incoming label → operation (Pop, Swap [new-label], Push) → next-hop → outgoing interface. 'show mpls lsp' shows RSVP-TE LSP status (up/down, bandwidth, path). 'show ldp session' shows LDP sessions. 'show ldp database' shows LDP label bindings. 'show rsvp session' shows RSVP state. For L3VPN, 'show route table [VRF-name].inet.0' shows the per-VRF routing table. 'show bgp summary' with VPNv4 family shows MP-BGP VPN route distribution.

Question 6: What is Segment Routing (SR-MPLS) and how does it differ from LDP-based MPLS?

SR-MPLS encodes the entire path in the packet header as a stack of node/adjacency SIDs, eliminating per-flow state in the network core and simplifying operations (Correct answer)
SR-MPLS uses BGP instead of OSPF for path computation
SR-MPLS requires dedicated hardware not used by traditional MPLS
SR-MPLS only works with IPv6 traffic

Correct answer: SR-MPLS encodes the entire path in the packet header as a stack of node/adjacency SIDs, eliminating per-flow state in the network core and simplifying operations

Segment Routing places the path information (as a stack of Segment IDs) in the packet at the ingress, eliminating the need for RSVP signaling and per-LSP state in core routers, while retaining MPLS forwarding.

Segment Routing (RFC 8660 for SR-MPLS) uses: Node SIDs — globally unique MPLS labels representing a router (distributed via IS-IS/OSPF SR extensions); Adjacency SIDs — labels for specific links; Prefix SIDs — for specific prefixes. The ingress router imposes a label stack encoding the path (or uses a single SID for shortest-path). Core routers pop/swap labels without per-LSP state. Benefits: eliminates RSVP, simplifies operations, enables source-based traffic engineering. Junos SR-MPLS: [edit protocols isis source-packet-routing node-segment ipv4-index N].

JNCIA MPLS and VPNs 2

In Junos L3VPN, what is the function of a Route Target (RT)?