Nokia BGP for Internet Routing study plan
Nokia BGP for Internet Routing · Give each week a clear focus.
8 weeks · About 7 hours per week
BGP Fundamentals and Protocol Operation & BGP Session Establishment and Neighbor Configuration
Explain from memory each BGP message type (OPEN, UPDATE, NOTIFICATION, KEEPALIVE, ROUTE-REFRESH) and the exact state machine transitions FSM passes through from Idle to Established
Describe how NLRI, path attributes, and the fixed 19-byte header with its length/type fields are encoded in an UPDATE, and why multiple prefixes can share one UPDATE
Compare iBGP and eBGP behavior: next-hop handling, AS_PATH prepending, TTL, loop prevention, and the requirement for a full mesh inside an AS
Configure two eBGP neighbors and one iBGP neighbor on Nokia SR OS with explicit peer-as, local-address, and import/export policies, then verify sessions reach Established with show commands
Diagnose a session stuck in Active or Connect by reading SR OS neighbor state output and identifying the missing or mismatched parameter
Your checkpoint
A one-page protocol cheat sheet (message types, FSM states, UPDATE anatomy) plus a working 3-router SR OS lab with two eBGP and one iBGP session all Established, with command output pasted as proof
Route Selection and Best-Path Decision Process
Rank the BGP best-path steps in exact order, including weight, local preference, locally originated, AS_PATH length, origin, MED, eBGP over iBGP, IGP metric to next hop, and tie-breakers
Predict the winner for at least six constructed prefix scenarios where each scenario differs in only one attribute
Explain why an SR OS implementation may evaluate weight or local preference differently from other vendors, and where the decision boundary sits
Distinguish deterministic MED, always-compare-MED, and default MED comparison, and state when each changes the outcome
Trace a full decision for one prefix using SR OS show router bgp routes detail output, naming each step that was evaluated
Your checkpoint
Six scenario flashcards with predicted winners and a written walkthrough of one real prefix through every best-path step, validated against SR OS output
BGP Attributes and Policy Control
Build import and export policies in SR OS that set local preference, prepend AS_PATH, and modify MED, and confirm the effect on the Loc-RIB and Adj-RIB-Out
Explain the difference between well-known mandatory, well-known discretionary, optional transitive, and optional non-transitive attributes and give an example of each
Construct a community scheme (for example no-export, no-advertise, and site-specific tags) and apply it with policy match and action entries
Describe how NEXT_HOP is rewritten on eBGP export and why an iBGP router may need a next-hop-self policy
Use route reflection or explicit policy to control which prefixes leave an AS, and verify with show router bgp routes advertised
Your checkpoint
A documented policy set (three policies minimum) with before/after route table excerpts showing local preference, prepending, and community tagging taking effect
Scaling and Convergence (Route Reflectors, Confederations, Timers)
Explain route reflector cluster design, ORIGINATOR_ID and CLUSTER_LIST loop prevention, and when a client-to-client reflection is allowed
Compare route reflectors with confederations on configuration complexity, policy visibility, and scaling limits
Describe BGP timers (Keepalive, Hold, ConnectRetry, MinRouteAdvertisementInterval) and the effect of each on convergence and session stability
Configure a route reflector with two clients and one non-client in SR OS and verify reflected routes carry ORIGINATOR_ID
Tune advertisement interval and hold timers in the lab and measure the change in convergence time for a withdrawn prefix
Your checkpoint
A convergence measurement report: baseline vs tuned timers for one prefix withdrawal, plus a topology diagram of the RR lab showing cluster and client roles
Internet Multihoming and Traffic Engineering & BGP Security and Route Filtering
Design a multihoming scheme that prefers one ISP for inbound and outbound traffic using local preference, AS_PATH prepending, and MED, and justify each knob
Explain the limits of MED across AS boundaries and why it is not a reliable inbound traffic-engineering tool between different providers
Build a prefix filter and an AS_PATH filter (including bogon and own-prefix filters) and apply them inbound on every eBGP session
Describe RPKI origin validation states (valid, invalid, not found) and how a router uses them in policy
Explain TCP MD5, TTL security (GTSM), and maximum-prefix limits as protective measures, and state what each does and does not stop
Your checkpoint
A hardened eBGP template with inbound prefix, AS_PATH, and RPKI-based policy plus maximum-prefix and GTSM, applied in the lab and shown blocking a deliberately invalid announcement
IPv6 and Address-Family Support in BGP & BGP Fundamentals and Protocol Operation
Explain multiprotocol BGP extensions: MP_REACH_NLRI, MP_UNREACH_NLRI, and how AFI/SAFI identify IPv6 unicast versus other families
Configure an IPv6 eBGP session in SR OS using the ipv6 address family and confirm IPv6 prefixes appear in the BGP table
Describe how next-hop handling and link-local addressing work for IPv6 BGP sessions
Review the BGP Fundamentals domain at a deeper level: route refresh capability, graceful restart, and their negotiation in OPEN
Identify three common causes of an IPv6 session failing to establish that do not occur in IPv4
Your checkpoint
A dual-stack lab where one router carries both IPv4 and IPv6 prefixes over separate sessions, with a table showing AFI/SAFI, next hop, and session state for each
Nokia SR OS BGP Implementation and Troubleshooting
Map SR OS BGP configuration hierarchy (router bgp, group, neighbor, afi-safi blocks) and explain inheritance from group to neighbor
Use at least eight SR OS show and debug commands to inspect neighbor state, received and advertised routes, policy results, and update statistics
Interpret common SR OS log and trap events for BGP session flaps and identify the root cause from the event text
Perform a policy troubleshooting exercise: given a prefix that is not being advertised, isolate whether the cause is policy, next-hop reachability, or session state
Explain SR OS commit, candidate configuration, and rollback behavior as they apply to BGP changes in production
Your checkpoint
A troubleshooting runbook of eight SR OS commands with sample output and a decision tree for isolating a missing-prefix problem
Route Selection and Best-Path Decision Process & BGP Attributes and Policy Control & Scaling and Convergence (Route Reflectors, Confederations, Timers) & Internet Multihoming and Traffic Engineering & BGP Security and Route Filtering & IPv6 and Address-Family Support in BGP & Nokia SR OS BGP Implementation and Troubleshooting
Complete a timed 60-question mixed-domain practice set and score at least 80 percent
Re-derive the full best-path order and the attribute type table from memory without notes
Re-run the two weakest lab exercises from weeks 1 through 7 and confirm correct output
Write a one-page summary of the five concepts you got wrong most often, with the corrected reasoning for each
Your checkpoint
A scored practice exam with an error log listing every missed question, its domain, the correct reasoning, and the specific study action taken to close the gap
Milestones
Week 2: Correctly predict the best path for at least five of six single-attribute scenarios without consulting notes, and explain each choice aloud in under one minute
Week 4: Route reflector lab shows reflected routes with ORIGINATOR_ID and CLUSTER_LIST present, and the convergence report contains measured before/after numbers
Week 6: Dual-stack lab passes: IPv4 and IPv6 sessions both Established and both prefix families visible in the BGP table with correct next hops
Week 8: Practice exam score of 80 percent or higher with no domain below 60 percent, and error log complete for every missed question