Managing TCP anycast requests

1. A method for managing traffic, comprising:

receiving a request at a first node of a content delivery network;

determining at the first node that the request should be serviced by a second node of the content delivery network; and

redirecting the request to the second node;

wherein the first node and the second node of the content delivery network comprise endpoints for serving content and wherein the content delivery network is at least in part an anycast network anycast network, and

wherein the request is serviceable by any of a plurality of nodes including the first node and the second node that share a same anycast IP address to which the request is directed.

2. A method as recited in claim 1 , wherein the request comprises a TCP anycast request.

3. A method as recited in claim 1 , wherein the request is received at the first node from an Internet Service Provider node.

4. A method as recited in claim 1 , further comprising monitoring at the first node sessions currently being serviced by the first node.

5. A method as recited in claim 1 , further comprising monitoring at the first node sessions being serviced at other nodes of the content delivery network including the second node.

6. A method as recited in claim 1 , further comprising determining that the first node is operating at capacity.

7. A method as recited in claim 1 , further comprising determining that the first node is servicing a prescribed number of sessions.

8. A method as recited in claim 1 , further comprising determining that the first node is in a maintenance or backup mode.

9. A method as recited in claim 1 , further comprising determining that the request is a part of a session that originated at the second node.

10. A method as recited in claim 1 , further comprising determining that the second node has capacity to service the request.

11. A method as recited in claim 1 , wherein the second node is part of a set of available nodes of the content delivery network that have capacity to service the request and further comprising determining that in the set of available nodes the second node provides a most efficient route to an end user who issued the request.

12. A method as recited in claim 1 , wherein the request is redirected to the second node via a virtual tunnel.

13. A method as recited in claim 1 , wherein the request is redirected to the second node via an HTTP 302 redirect.

14. A method as recited in claim 1 , wherein the request is redirected to the second node using a unicast IP address associated with the second node.

15. A method as recited in claim 1 , further comprising servicing the request at the second node.

16. A system for managing traffic, comprising:

a processor at a first node of a content delivery network configured to:

receive a request;

determine that the request should be serviced by a second node of the content delivery network; and

redirect the request to the second node; and

a communication interface coupled to the processor and configured to receive the request;

wherein the first node and the second node of the content delivery network comprise endpoints for serving content and wherein the content delivery network is at least in part an anycast network anycast network, and

wherein the request is serviceable by any of a plurality of nodes including the first node and the second node that share a same anycast IP address to which the request is directed.

17. A system as recited in claim 16 , wherein the request is received at the first node from an Internet Service Provider node.

18. A computer program product for managing traffic, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving a request at a first node of a content delivery network;

determining that the request should be serviced by a second node of the content delivery network; and

redirecting the request to the second node;

wherein the first node and the second node of the content delivery network comprise endpoints for serving content and wherein the content delivery network is at least in part an anycast network anycast network, and

wherein the request is serviceable by any of a plurality of nodes including the first node and the second node that share a same anycast IP address to which the request is directed.