System and method for handling link loss in a network

1. A system for managing a network, comprising:

a plurality of network nodes interconnected via a plurality of primary links;

a controller in communication with the plurality of network nodes, the controller including one or more processors configured to:

provide an event profile to the plurality of network nodes, the event profile indicating possible link loss scenarios, including potential combinations of failed links, and routing changes to be implemented by each of the plurality of nodes in the possible link loss scenarios; and

wherein each of the plurality of nodes is configured to:

store the event profile;

determine that a particular link loss event occurred;

determine, based on the event profile, routing changes for the particular link loss event; and

implement the determined routing changes; and

broadcast the particular link loss event on all available links.

2. The system of claim 1 , wherein the controller is further configured to:

receive information from the plurality of nodes; and

update the event profiles based on the received information.

3. The system of claim 2 , wherein:

the received information identifies links having a high probability of link breaks; and

the updating of the profiles accounts for scenarios where the identified links are lost.

4. The system of claim 1 , wherein the plurality of network nodes communicate link loss events to one another through at least one of: applications on the nodes or L2 separated control paths.

5. The system of claim 4 , wherein communicating the link loss events through the applications comprises:

broadcasting, by a first node of the plurality of nodes, a frame with a special ether type; and

detecting, by a second node of the plurality of nodes, the ether type and forwarding the frame to an application running on the second node.

6. The system of claim 5 , wherein the second node is configured to determine whether to re-broadcast the frame, the determining based on whether the frame has not yet been transmitted over one or more interfaces.

7. The system of claim 6 , wherein determining whether to re-broadcast the frame comprises looking, by the application on the second node, at a cookie in the frame.

8. A method of updating routing information in a network node, comprising:

receiving, from a controller, an event profile, the event profile indicating possible link loss scenarios, including potential combinations of failed links, and routing changes to be implemented by each of a plurality of nodes in the possible link loss scenarios;

storing, by the network node, the event profile; and

determining, with one or more processors, that a particular link loss event occurred;

determining, with the one or more processors based on the event profile, routing changes for the particular link loss event;

implementing, with the one or more processors, the determined routing changes; and

broadcasting the particular link loss event on all available links.

9. The method of claim 8 , further comprising:

providing information to the controller, the information indicating a status of one or more links in the network.

10. The method of claim 9 , wherein:

the information identifies links having a high probability of link breaks.

11. The method of claim 9 , further comprising receiving an updated event profile from the controller in response to providing the information.

12. The method of claim 8 , wherein determining that a particular link loss event occurred comprises detecting failure of a link directly coupled to the node.

13. The method of claim 8 , wherein determining that a particular link loss event occurred comprises receiving a notification of the link loss event from a neighboring node.

14. The method of claim 13 , further comprising:

determining whether the node has previously transmitted the notification to other nodes; and

re-transmitting the notification if it has not yet been transmitted.

15. The system of claim 14 , wherein determining whether the node has previously transmitted the notification comprises looking, by an application running on the node, at a cookie in the frame.

16. The method of claim 8 , further comprising communicating link loss events to other nodes in the network through an application running on the node.

17. The method of claim 16 , wherein communicating the link loss events through the application comprises broadcasting a frame with a special ether type detectable by the other nodes.

18. A network node, comprising:

a memory storing one or more routing tables; and

one or more processors in communication with the memory, the one or more processors programmed to:

receive, from a controller, an event profile, the event profile indicating possible link loss scenarios, including potential combinations of failed links, and routing changes to be implemented by each of a plurality of nodes in of the possible link loss scenarios;

store the event profile in the memory;

determine that a particular link loss event occurred;

determine, based on the stored event profile, routing changes for the particular link loss event;

implement the determined routing changes; and

broadcast the particular link loss event on all available links.

19. The node of claim 18 , wherein determining that a particular link loss event occurred comprises receiving a notification of the link loss event from a neighboring node; and wherein the node is further programmed to:

determine whether it has previously transmitted the notification to other nodes; and

re-transmit the notification if it has not yet been transmitted.

20. The node of claim 19 , the node executing one or more applications, wherein the one or more applications determine whether the notification was previously transmitted and re-transmit the notification.