120 likes | 139 Views
This draft presents a solution for Shared Mesh Protection in MPLS-TP networks, focusing on coordination efficiency and fast failover mechanisms. The protocol messages and operational principles are detailed, with examples illustrating various scenarios.
E N D
MPLS-TP Shared Mesh Protectiondraft-cheung-mpls-tp-mesh-protection-05 IETF 83 – Paris, March 2012 Taesik Cheung and Jeong-dong Ryoo (ETRI) Yaacov Weingarten and Nurit Sprecher (NSN) Daniel King (Old Dog)
Main Points of Solution • Utilizes the existing linear protection as is • No modification needed in the state transition tables • Only the additional coordination actions for SMP are defined in this draft • One LSP for each protection path • Each LSP has end-to-end OAM/PSC channel • Parallel coordination of protection segments • Use SPMEs for direct communication • Minimize coordination processing time • Immediate traffic switchover • Rapid failover as fast as linear protection
Changes from 04 • Block function at the SEN was added. • To reduce possible congestion during pre-emption • SEN blocks protection LSP to be pre-empted and sends Lockout of Protection (LoP) request. • When an end node of the protection LSP completes the LoP operation, it will send an acknowledge to SEN. • SEN unblocks protection LSP when it receives the LoP Ack. • SMP protocol message format was defined. • Initial definition • The mechanism was described more in detail. • Editorial changes
SMP protocol message format (4b) (4b) (4b) (2b) (10b) (8b)
Next steps • Currently, there are multiple drafts regarding SMP. • We will continue the discussion with the authors of the other drafts to explore a possibility to develop a single solution optimized for MPLS-TP networks. • We kindly ask the WG experts to review our document and provide us feedback.
Backup slides Protocol Messages • Blue: Linear protection protocol messages • Red: PS Event (EN SEN) • Orange (solid / dotted): RA Notification / Ack (SEN SSN) • Green (solid / dotted): LoP Request / Ack (SEN/SSN EN) SF (AB) X W2 A B [Acronyms] PS: Protection Switching RA: Resource Allocation LoP: Lockout of Protection EN: End Node SEN: Shard End Node SSN: Shared Start Node P2 P Q R S T U P1 P3 P4 C D E F G H W1 W3 W4 [ Priority: P1>P2>P3>P4 ]
Backup slides Principle of Operation • End node detecting failure initiates protection • Starts linear protection and sends PS Event to SEN • SEN coordinates use of shared resource • Sends RA Notification to SSN (and waits for RA Ack from SSN) • Blocks the equal or lower priority protection LSP • Sends LoP Request to the end node of the protection LSP • Unblocks the protection LSP when it receives LoP Ack • SSN coordinates use of shared resource • Sends RA Ack to SEN • Same procedures with SEN (step 2-4)
Backup slides Example 1 (SF at high priority LSP) • Procedures • A detects SF, sends PS Event to P and starts linear protection. • P blocks P2, sends LoP Request to C and RA Notification to Q. • Q blocks P2 and sends LoP Request to D (and replies RA Ack to P). • C and D perform LoP and send LoP Ack to P and Q. • P and Q unblock P2. 1. SF (AB) X A B A B W1 W1 P1 P1 Priority P1 > P2 P Q P Q P2 P2 W2 W2 C D C D Lockout Lockout [ Normal ] [ Protection ]
Backup slides Example 2 (SF at low priority LSP) • Procedures • C detects SF, sends PS Event to P and starts linear protection. • P does nothing. A B A B W1 W1 P1 P1 P Q P Q P2 P2 W2 W2 C D C D X 1. SF (CD) [ Normal ] [ Protection ]
Backup slides Example 3 (Pre-emption) • Procedures (continued from Example 2) • A detects SF, sends PS Event to P and starts linear protection. • P blocks P2, sends LoP Request to C and RA Notification to Q. • Q blocks P2, sends LoP Request to D (and replies RA Ack to P). • C and D perform LoP and send LoP Ack to P and Q. (return to failed W2) • P and Q unblock P2. 2. SF (AB) X A B A B W1 W1 P1 P1 Priority P1 > P2 P Q P Q P2 P2 W2 W2 X C D C D X X 1. SF (CD) 1. SF (CD) Lockout Lockout [ P2 occupies P-Q ] [ P2 pre-empted ]
Backup slides Example 4 (Equal priority) • Procedures (First come, first served) • A detects SF, sends PS Event to P and starts linear protection. • P sends RA Notification to Q and waits for reply. • Q blocks P2, sends LoP Request to D and replies RA Ack to P. • P blocks P2 and sends LoP Request to C. • C and D perform LoP and send LoP Ack to P and Q. • P and Q unblock P2. 1. SF (AB) X A B A B W1 W1 P1 P1 Priority P1 = P2 P Q P Q P2 P2 W2 W2 C D C D Lockout Lockout [ Normal ] [ Protection ]
Backup slides Example 5 (Race condition) • Procedures (Equal priority resolved by Protection path ID) • B and C detect SF, send PS Event to Q and P and start linear protection. • Q and P send RA Notification to each other and wait for reply. • P blocks P2, sends LoP Request to C and replies RA Ack to Q. • Q blocks P2 and sends LO Request to D. • C and D perform LoP and send LO Ack to P and Q. (return to failed W2) • P and Q unblock P2. 1. SF (AB) X A B A B W1 W1 P1 P1 Priority P1 = P2 (ID1 > ID2) P Q P Q P2 P2 W2 W2 X C D C D X 1. SF (CD) Lockout Lockout [ Normal ] [ Protection: P2 rejected ]