Method and apparatus for managing the interconnection between network domains

1. A method of managing an interconnect region between a first network domain and a second network domain, the method comprising the steps of:

establishing a control instance by first, second, third, and fourth network elements in the interconnect region, the interconnect region including a plurality of links extending between the first and second network elements on the first network domain and the third and fourth network elements on the second network domain, the control instance being implemented as a spanning tree that does not extend beyond the interconnect region and which allows the first, second, third, and fourth network elements to select at most one of the links extending between the first and second network domains to be selected as an active link and causing remaining links extending between the first and second network domains to be designated by the first, second, third, and fourth network elements as backup links; and

configuring the interconnect region to avoid loop formation under a failure condition;

wherein the interconnect region includes four links interconnecting the first and second network elements with the third and fourth network elements, wherein a fifth link interconnects the first and second network elements, wherein a sixth link interconnects the third and fourth network elements, and wherein the control instance in the interconnect region allows at most one of the four links interconnecting the first and second network elements with the third and fourth network elements to be selected as the active link between the two domains.

2. The method of claim 1 , wherein the control instance is implemented to cause the selection of the active link and designation of backup links to be implemented on a per-port basis.

3. The method of claim 1 , wherein multiple control instances are implemented to cause the selection of the active link and designation of backup links to be implemented on a per-VLAN basis.

4. The method of claim 1 , wherein the control instance is a single instance of a spanning tree.

5. The method of claim 1 , wherein the control instance includes multiple spanning trees to enable the active link to be selected on a per-VLAN basis.

6. The method of claim 1 , wherein the first network domain is an Ethernet network domain and the second network domain is a MultiProtocol Label Switch (MPLS) domain.

7. The method of claim 1 , wherein multiple control instances are implemented, each of the control instances being a spanning tree that does not extend beyond the interconnect region, and wherein the multiple control instances are used to enable the selection of the active link and designation of backup links to occur on a per-VLAN basis.

8. The method of claim 1 , further comprising a step of engineering the interconnect by assigning a lower weight to the fifth and sixth links than is assigned to the first through fourth links to cause the control instance in the interconnect region to automatically select only one of the four links interconnecting the first and second network elements with the third and fourth network elements as the active link.

9. The method of claim 8 , wherein the control instance is a single instance of a spanning tree, the spanning tree including the fifth link, the sixth link, and only one of the four links interconnecting the first and second network elements with the third and fourth network elements as the active links in the interconnect region.

10. The method of claim 1 , wherein the control instance includes multiple instances of a spanning tree, and wherein each of the multiple instances of the spanning tree are enabled to select one of the four links interconnecting the first and second network elements with the third and fourth network elements as the active links in the interconnect region on a per-VLAN basis.

11. A method of managing an interconnect region between a first network domain and a second network domain, the interconnect region including a plurality of links extending between first nodes in the first network domain and second nodes in the second network domain, the method comprising the steps of:

establishing a control instance by the first nodes and the second nodes in the interconnect region, the control instance allowing the first nodes and second nodes to select at most one of the links extending between the first and second network domains to be selected as an active link and causing remaining links extending between the first and second network domains to be designated by the first nodes and second nodes as backup links;

wherein the control instance is a spanning tree that is used to select at most one of the links interconnecting the first and second network domains as the active link;

wherein the interconnect region includes first and second network elements on the first network domain and third and fourth network elements on the second network domain, wherein the interconnect region includes four links interconnecting the first and second network elements with the third and fourth network elements, wherein a fifth link, interconnects the first and second network elements, wherein a sixth link interconnects the third and fourth network elements, and wherein the control instance in the interconnect region allows at most one of the four links interconnecting the first and second network elements with the third and fourth network elements to be selected as the active link between the two domains;

wherein the control instance includes multiple instances of a spanning tree, and wherein each of the multiple instances of the spanning tree are enabled to select one of the our links interconnecting the first and second network elements with the third and fourth network elements as the active links in the interconnect region on a per-VLAN basis; and

wherein the third and fourth network elements on the second network domain have access control lists associated with the sixth link configured to allow control Protocol Data Units (PDUs) to pass over the sixth link but not to allow data PDUs to be transmitted over the sixth link.

12. The method of claim 11 , wherein the third and fourth network elements terminate control PDUs so that the control PDUs are not forwarded across the second network domain.

13. A method of managing an interconnect region between a first network domain and a second network domain, the interconnect region including a plurality of links extending between first nodes in the first network domain and second nodes in the second network domain, the method comprising the steps of:

establishing a control instance by the first nodes and the second nodes in the interconnect region, the control instance allowing the first nodes and second nodes to select at most one of the links extending between the first and second network domains to be selected as an active link and causing remaining links extending between the first and second network domains to be designated by the first nodes and second nodes as backup links;

establishing an OAM management entity between the first and second network elements;

transmitting OAM frames by the first network element on the management entity over the active link between the first and second network domains and addressed to the second network element; and

interpreting receipt of the OAM frames at the second network element as an indication of loop formation;

wherein the control instance is a spanning tree that is used to select at most one of the links interconnecting the first and second network domains as the active link; and

wherein the interconnect region includes first and second network elements on the first network domain and third and fourth network elements on the second network domain, wherein the interconnect region includes four links interconnecting the first and second network elements with the third and fourth network elements, wherein a fifth link interconnects the first and second network elements, wherein a sixth link interconnects the third and fourth network elements, and wherein the control instance in the interconnect region allows at most one of the four links interconnecting the first and second network elements with the third and fourth network elements to be selected as the active link between the two domains.

14. The method of claim 13 , further comprising the step of administratively disabling, by the second network element, a link over which the OAM frame was received by the second network element.

15. A network, comprising:

a MultiProtocol Label Switch (MPLS) domain;

a first Ethernet domain connected to the MPLS domain at a first interconnect region, the first interconnect region including a first Ethernet node in the first Ethernet domain that is connected to a first MPLS node in the MPLS domain and a second Ethernet node in the first Ethernet domain that is connected to a second MPLS node in the MPLS domain; and

a second Ethernet domain connected to the MPLS domain at a second interconnect region, the second interconnect region including a third Ethernet node in the second Ethernet domain that is connected to a third MPLS node in the MPLS domain and a fourth Ethernet node in the second Ethernet domain that is connected to a fourth MPLS node in the MPLS domain;

wherein a separate common control protocol instance is established to include the first interconnect region and the second interconnect region, the common control protocol instance including the first Ethernet node, the second Ethernet node, the third Ethernet node, and the fourth Ethernet node, but not any of the MPLS nodes, and enabling the Ethernet nodes in the first and second interconnect regions to operate as a single link state protocol controlled Ethernet network.

16. The network of claim 15 , and wherein routes between the first through fourth Ethernet nodes are installed to follow shortest paths between them via pseudowires in the MPLS domain.

17. A method of managing an interconnect region between an Ethernet network domain and a MultiProtocol Label Switch (MPLS) network domain, the interconnect region comprising a first service provider edge (SPE) node and a second SPE node operative to interconnect the Ethernet network domain to the interconnect region, and a first multi-service edge (MSE) node and a second MSE node operative to interconnect the MPLS network domain to the interconnect region, the method comprising:

periodically transmitting, by the first SPE node, OAM packets to the first MSE node, the OAM packets being addressed to the second SPE node;

receiving, by the second SPE node, one of the OAM packets from the second MSE node; and

based on receiving the one of the OAM packets, determining that a loop exists between the Ethernet network domain and the MPLS network domain.

18. The method of claim 17 , wherein:

the first MSE node is coupled to the second MSE node via a first link;

the first MSE node is coupled to the MPLS network domain via a second link; and

wherein the first MSE node communicates the OAM packets to the MPLS network domain via the second link and does not communicate the OAM packets to the second MSE node via the first link.

19. The method of claim 18 , further comprising:

accessing, by the first MSE node, an access control list that indicates data packets may not be communicated to the second MSE node;

interpreting the OAM packets to be data packets; and

wherein, based on the access control list, the first MSE node communicates the OAM packets to the MPLS network domain via the second link and does not communicate the OAM packets to the second MSE node via the first link.