Dynamic protection against failure of a head-end node of one or more TE-LSPs

• US 20070165515A1
• Filed: 01/18/2006
• Published: 07/19/2007
• Est. Priority Date: 01/18/2006
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Vasseur, Jean-Philippe
• Granted Patent: US 8,441,919 B2
• Time in Patent Office: Days
• US Class Current: 370/216
• CPC Class Codes: H04L 12/66 Arrangements for connecting between networks having differing types of switching systems, e.g. gateways
• CPC Class Codes: H04L 41/0659 by isolating or reconfiguring faulty entities
• CPC Class Codes: H04L 41/0668 by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure
• CPC Class Codes: H04L 45/00 Routing or path finding of packets in data switching networks routing or path finding in wireless networks H04W40/00
• CPC Class Codes: H04L 45/12 Shortest path evaluation
• CPC Class Codes: H04L 45/22 Alternate routing
• CPC Class Codes: H04L 45/247 using M:N active or standby paths
• CPC Class Codes: H04L 45/28 using route fault recovery
• CPC Class Codes: H04L 45/502 Frame based
• CPC Class Codes: H04L 45/851 Dynamic network selection or re-selection, e.g. after degradation of quality

Abstract

A technique dynamically protects against failure of a head-end node of one or more primary Traffic Engineering Label Switched Paths (TE-LSPs) in a computer network. According to the novel technique, a neighboring upstream node (“protecting node”) of the head-end node learns of the primary TE-LSPs (and their respective state control blocks) extending from the head-end node to one or more address prefixes. The protecting node establishes a repair TE-LSP for each of the primary TE-LSPs to a corresponding downstream neighboring node of the head-end node (a “next-next-hop”). In response to detecting a failure of the head-end node, the protecting node locally reroutes traffic destined for the address prefixes to an appropriate repair TE-LSP. Due to the failure of the head-end node, the protecting node then refreshes the states of the primary TE-LSPs using replicated state control blocks accordingly, until the repair TE-LSPs are no longer needed.

Claims

1. A method for dynamically protecting against failure of a head-end node of one or more primary Traffic Engineering Label Switched Paths (TE-LSPs) in a computer network, the method comprising:

learning the one or more primary TE-LSPs extending from the head-end node to one or more address prefixes at a neighboring upstream node (protecting node) of the head-end node;

establishing a repair TE-LSP for each of the one or more primary TE-LSPs from the protecting node to a corresponding downstream neighboring node of the head-end node (next-next-hop);

detecting a failure of the head-end node at the protecting node; and

, in response

locally rerouting traffic at the protecting node destined for the one or more address prefixes via the head-end node to an appropriate repair TE-LSP.

2. The method as in claim 1, further comprising:

requesting, from the head-end node, that the protecting node protect the head-end node from failure.

3. The method as in claim 2, further comprising:

requesting protection using one or more flags in an Interior Gateway Protocol (IGP) message.

4. The method as in claim 2, further comprising:

requesting protection for a selected subset of address prefixes reachable by the head-end node.

5. The method as in claim 1, further comprising:

learning the one or more primary TE-LSPs to the one or more address prefixes at the protecting node through Interior Gateway Protocol (IGP) messages.

6. The method as in claim 1, further comprising:

learning the one or more primary TE-LSPs to the one or more address prefixes at the protecting node through tunnel protocol signaling messages exchanged through a tunnel between the protecting node and the head-end node.

7. The method as in claim 6, wherein the tunnel between the protecting node and the head-end node is an unconstrained one-hop tunnel used only for the learning of the primary TE-LSPs.

8. The method as in claim 6, wherein the tunnel protocol is a Resource ReSerVation Protocol (RSVP).

9. The method as in claim 1, wherein the protecting node is configured as a provider edge (PE) node, and the head-end node is configured as a provider (P) node.

10. The method as in claim 1, further comprising:

associating at the protecting node each of the address prefixes with one of the repair TE-LSPs.

11. The method as in claim 1, further comprising:

preventing refresh state timeout of the primary TE-LSPs in the event of a head-end node failure.

12. The method as in claim 11, further comprising:

learning the one or more primary TE-LSPs and their respective state control block from the head-end node to one or more address prefixes; and

preventing refresh state timeout by sending replica state control blocks from the protecting node to the corresponding primary TE-LSPs.

13. The method as in claim 12, wherein the state control block is a label-switched packet, the method further comprising:

pushing a label onto the state control block at the protecting node that the head-end node is configured to send to the next-next-hop.

14. The method as in claim 12, further comprising:

configuring the next-next-hop to appropriately handle multiple replica state control blocks from different protecting nodes.

15. The method as in claim 14, wherein the appropriate handling further comprises:

sending returned state control blocks from the next-next-hop to each of the different protecting nodes.

16. The method as in claim 12, further comprising:

preventing refresh state timeout until the primary TE-LSPs are no longer needed.

17. The method as in claim 16, wherein the primary TE-LSPs are no longer needed in response to at least one of either network reconvergence or head-end node restoration.

18. The method as in claim 1, further comprising:

detecting failure of the head-end node based on a detection mechanism selected from the group comprising;

Interior Gateway Protocol (IGP) keepalive messages, tunnel signaling protocol hello messages on tunnels from the protecting node to the head-end node, and Bidirectional Forwarding Detection (BFD) control messages.

19. An apparatus for dynamically protecting against failure of a head-end node of one or more primary Traffic Engineering Label Switched Paths (TE-LSPs) in a computer network, the apparatus comprising:

means for learning the one or more primary TE-LSPs extending from the head-end node to one or more address prefixes at a neighboring upstream node (protecting node) of the head-end node;

means for establishing a repair TE-LSP for each of the one or more primary TE-LSPs from the protecting node to a corresponding downstream neighboring node of the head-end node (next-next-hop);

means for detecting a failure of the head-end node at the protecting node; and

, in response

means for locally rerouting traffic at the protecting node destined for the one or more address prefixes via the head-end node to an appropriate repair TE-LSP.

20. A computer readable medium containing executable program instructions for dynamically protecting against failure of a head-end node of one or more primary Traffic Engineering Label Switched Paths (TE-LSPs) in a computer network, the executable program instructions comprising program instructions for:

learning the one or more primary TE-LSPs extending from the head-end node to one or more address prefixes at a neighboring upstream node (protecting node) of the head-end node;

establishing a repair TE-LSP for each of the one or more primary TE-LSPs from the protecting node to a corresponding downstream neighboring node of the head-end node (next-next-hop);

detecting a failure of the head-end node at the protecting node; and

, in response

locally rerouting traffic at the protecting node destined for the one or more address prefixes via the head-end node to an appropriate repair TE-LSP.

21. A node for dynamically protecting against failure of a head-end node of one or more primary Traffic Engineering Label Switched Paths (TE-LSPs) in a computer network, wherein the node is a neighboring upstream node (protecting node) of the head-end node, the node comprising:

one or more network interfaces;

a processor coupled to the one or more network interfaces and adapted to execute software processes; and

a memory adapted to store a protection process executable by the processor, the protection processes configured to;

i) learn the one or more primary TE-LSPs extending from the head-end node to one or more address prefixes, ii) establish a repair TE-LSP for each of the one or more primary TE-LSPs to a corresponding downstream neighboring node of the head-end node (next-next-hop), iii) detect a failure of the head-end node, and, in response iv) locally reroute traffic destined for the one or more address prefixes via the head-end node to an appropriate repair TE-LSP.

22. A head-end node of one or more primary Traffic Engineering Label Switched Paths (TE-LSPs) in a computer network, the head-end node comprising:

one or more network interfaces;

a processor coupled to the one or more network interfaces and adapted to execute software processes; and

a memory adapted to store a protection request process executable by the processor, the protection request processes configured to;

i) request that one or more neighboring upstream nodes (protecting nodes) of the head-end node protect the head-end node against failure, and ii) inform the one or more protecting nodes of the one or more primary TE-LSPs and their respective state control block from the head-end node to one or more address prefixes.