System, device, and method for improving communication network reliability using trunk splitting
View Patent ↗A multilink trunk from a client device in a communication network is split between two aggregation devices such that each aggregation device is connected to at least one link of the multilink trunk. The aggregation devices work in conjunction to appear to the client device as a single device connected to the client device over the multilink trunk. The aggregation devices transmit identical bridge protocol data units to the client device over their respective links of the multilink trunk. The aggregation devices exchange operational information and data packets over inter-device communication ports. A failure of the split multilink trunk at one aggregation device is restored through the other aggregation device.
1. A method for improving communication network reliability, the method comprising: splitting a multilink trunk from a client device of a network between two aggregation devices such that each aggregation device is coupled to at least one link of the split multilink trunk; and operating the aggregation devices such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk by assigning a unique identifier for the split multilink trunk and using the unique identifier by each aggregation device to identify the split multilink trunk.
2. The method of claim 1 , wherein splitting a multilink trunk from a client device of a network between two aggregation devices such that each aggregation device is coupled to at least one link of the split multilink trunk comprises: coupling at least one link of the multilink trunk to each of the aggregation devices; and configuring the respective links of the multilink trunk for each aggregation device as a split multilink trunk.
3. The method of claim 2 , wherein configuring the respective links of the multilink trunk for each aggregation device as a split multilink trunk comprises: configuring each port associated with the split multilink trunk as a split multilink trunk logical port type.
4. The method of claim 1 , wherein operating the aggregation devices such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk comprises: establishing a communication connection between the aggregation devices; and exchanging information and data packets between the aggregation devices over the communication connection using a communication connection protocol in order to operate the aggregation devices such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk.
5. A method for improving communication network reliability, the method comprising: splitting a multilink trunk from a client device of a network between two aggregation devices such that each aggregation device is coupled to at least one link of the split multilink trunk: and operating the aggregation devices such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk wherein operating the aggregation devices such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk comprises: transmitting substantially identical bridge protocol data unit (BPDU) messages by the aggregation devices to the client device over their respective links of the split multilink trunk.
6. The method of claim 5 , wherein transmitting substantially identical bridge protocol data unit (BPDU) messages by the aggregation devices to the client device over their respective links of the split multilink trunk comprises: determining a value for each of a number of BPDU message parameters; and using the determined value for each of the number of BPDU message parameters by each aggregation device.
7. The method of claim 6 , wherein determining a value for each of a number of BPDU message parameters comprises: negotiating the value for at least one of the number of BPDU message parameters by the aggregation devices.
8. The method of claim 6 , wherein the number of BPDU message parameters comprises a bridge identifier parameter, and wherein determining a value for the bridge identifier parameter comprises: determining a relative priority for each aggregation device; determining the higher priority aggregation device; determining a bridge identifier for the higher priority aggregation device; and determining the value for the bridge identifier parameter to be the bridge identifier for the higher priority aggregation device.
9. The method of claim 6 , wherein the number of BPDU message parameters comprises a root path cost parameter, and wherein determining a value for the root path cost parameter comprises: determining a relative priority for each aggregation device; determining the higher priority aggregation device; determining a root path cost for the higher priority aggregation device; and determining the value for the root path cost parameter to be the root path cost for the higher priority aggregation device.
10. The method of claim 6 , wherein the number of BPDU message parameters comprises a port identifier parameter, and wherein determining a value for the port identifier parameter comprises: determining a relative priority for each aggregation device; determining the higher priority aggregation device; determining a highest priority port identifier from among ports of the higher priority aggregation device associated with the split multilink trunk; and determining the value for the port identifier parameter to be the highest priority port identifier from among ports of the higher priority aggregation device associated with the split multilink trunk.
11. The method of claim 1 , wherein operating the aggregation devices such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk comprises: maintaining an address table by each aggregation device; and forwarding packets by each aggregation device based upon its respective address table.
12. The method of claim 11 , wherein maintaining an address table by each aggregation device comprises: learning an address by one of the aggregation devices; maintaining the address by said one of the aggregation devices in its address table; sending the learned address by said one of the aggregation devices to the other aggregation device; and maintaining the address by the other aggregation device in its address table.
13. The method of claim 12 , wherein learning an address by one of the aggregation devices comprises at least one of: learning an address of a directly connected device; and learning an address of an end device over the split multilink trunk.
14. The method of claim 11 , wherein forwarding packets by each aggregation device based upon its respective address table comprises: receiving a packet by one of the aggregation devices from a device other than the other aggregation device; searching the address table for a packet destination; determining that the packet destination is unknown; and forwarding the packet to at least the other aggregation device.
15. The method of claim 14 , further comprising: forwarding the packet over the split multilink trunk, provided the packet was not received over the split multilink trunk.
16. The method of claim 11 , wherein forwarding packets by each aggregation device based upon its respective address table comprises: receiving a packet by one of the aggregation devices from the other aggregation device;
searching the address table for a packet destination;
determining that the packet destination is known;
determining that the packet destination is associated with an active dual-homed split multilink trunk link; and
dropping the packet without forwarding the packet over any port.
17. The method of claim 11 , wherein forwarding packets by each aggregation device based upon its respective address table comprises :-receiving a packet by one of the aggregation devices from the other aggregation device;
searching the address table for a packet destination;
determining that the packet destination is known;
determining that the packet destination is not associated with an active dual-homed split multilink trunk link; and
forwarding the packet to all ports that are not associated with an active dual-homed SMLT port or with the other aggregation device.
18. The method of claim 11 , wherein forwarding packets by each aggregation device based upon its respective address table comprises: receiving a packet by one of the aggregation devices from the other aggregation device;
searching the address table for a packet destination; determining that the packet destination is unknown; and
forwarding the packet to all ports that are not associated with an active dual-homed SMLT port or with the other aggregation device.
19. The method of claim 12 , wherein maintaining an address table by each aggregation device comprises: determining that an address has been aged by at least one of the aggregation devices; and removing the aged address by each aggregation device from its respective address table.
20. The method of claim 1 , wherein operating the aggregation devices such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk comprises: load sharing by the aggregation devices over their respective links of the split multilink trunk.
21. The method of claim 1 , wherein operating the aggregation devices such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk comprises: determining by one of the aggregation devices that all links of its split multilink trunk have failed; and restoring communication between said one of the two aggregation devices and the client device through the other aggregation device.
22. The method of claim 21 , wherein restoring communication between said one of the aggregation devices and the client device through the other aggregation device comprises: sending a message by said one of the aggregation devices to the other aggregation device, the message informing the other aggregation device that all links of its split multilink trunk have failed; and redirecting all affected flows from the failed split multilink trunk to the other aggregation device.
23. The method of claim 22 , wherein redirecting all affected flows from the failed split multilink trunk to the other aggregation device comprises: changing the status of all ports associated with the failed split multilink trunk.
24. The method of claim 22 , wherein restoring communication between said one of the aggregation devices and the client device through the other aggregation device further comprises: receiving said message by said other aggregation device; and reconfiguring the links of the split multilink trunk for said other aggregation device from a split multilink trunk to a multilink trunk.
25. The method of claim 24 , wherein reconfiguring the links of the split multilink trunk for said other aggregation device from a split multilink trunk to a multilink trunk comprises: configuring each port associated with the split multilink trunk from a split multilink trunk logical port type to a multilink trunk logical port type.
26. The method of claim 4 , wherein the communication connection protocol comprises at least one of: a hello message that is used to verify connectivity between the aggregation devices; a notification of learned or migrated MAC addresses message that is used to exchange learned or migrated MAC addresses between the aggregation devices; and a notification of aged address message that is used to exchange address aging information for addresses associated with the split multilink trunk between the aggregation devices.
27. An apparatus for operating as an aggregation device in a communication network having a multilink trunk that is split between two aggregation devices, the apparatus comprising: at least one inter-device communication port for communicating with the other aggregation device using a communication connection protocol; at least one split multilink trunk communication port for communicating with a client device over at least one link of a multilink trunk from the client device that is split between the aggregation devices; and aggregation logic operably coupled to operate in conjunction with other aggregation device so that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk wherein the aggregation logic is operably coupled to maintain a unique identifier for the split multilink trunk and to use the unique identifier to identify the split multilink trunk.
28. The apparatus of claim 27 , wherein the aggregation logic is operably coupled to configure each port associated with its split multilink trunk as a split multilink trunk logical port type.
29. The apparatus of claim 27 , wherein the aggregation logic is operably coupled to exchange information and data packets with the other aggregation device over the inter-device communication port using the communication connection protocol.
30. An apparatus for operating as an aggregation device in a communication network having a multilink trunk that is split between two aggregation devices, the apparatus comprising: at least one inter-device communication port for communicating with the other aggregation device using a communication connection protocol at least one split multilink trunk communication port for communicating with a client device over at least one link of a multilink trunk from the client device that is split between the aggregation devices; and aggregation logic operably coupled to operate in conjunction with other aggregation device so that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunk wherein the aggregation logic is operably coupled to negotiate with the other aggregation device to determine values for each of a number of bridge protocol data unit (BPDU) parameters and to transmit substantially identical bridge protocol data unit (BPDU) messages as the other aggregation device to the client device over its links of the split multilink trunk.
31. The apparatus of claim 30 , wherein the number of BPDU message parameters comprises a bridge identifier parameter, and wherein the aggregation logic is operably coupled to determine a relative priority for each aggregation device, determine the higher priority aggregation device, determine a bridge identifier for the higher priority aggregation device, and determine the value for the bridge identifier parameter to be the bridge identifier for the higher priority aggregation device.
32. The apparatus of claim 30 , wherein the number of BPDU message parameters comprises a root path cost parameter, and wherein the aggregation logic is operably coupled to determine a relative priority for each aggregation device, determine the higher priority aggregation device, determine a root path cost for the higher priority aggregation device, and determine the value for the root path cost parameter to be the root path cost for the higher priority aggregation device.
33. The apparatus of claim 30 , wherein the number of BPDU message parameters comprises a port identifier parameter, and wherein the aggregation logic is operably coupled to determine a relative priority for each aggregation device, determine the higher priority aggregation device, determine a highest priority port identifier from among ports of the higher priority aggregation device associated with the split multilink trunk, and determine the value for the port identifier parameter to be the highest priority port identifier from among ports of the higher priority aggregation device associated with the split multilink trunk.
34. The apparatus of claim 27 , wherein the aggregation logic is operably coupled to maintain an address table and forward packets based upon the address table.
35. The apparatus of claim 34 , wherein the aggregation logic is operably coupled to learn an address, maintain the address in its address table, and send the learned address to the other aggregation device over the inter-device communication port.
36. The apparatus of claim 34 , wherein the aggregation logic is operably coupled to receive a packet from a device other than the other aggregation device, search the address table for a packet destination, determine that the packet destination is unknown, and forward the packet to at least the other aggregation device over the inter-device communication port.
37. The apparatus of claim 36 , wherein the aggregation logic is operably coupled to forward the packet to the client device over the split multilink trunk, provided the packet was not received over the split multilink trunk.
38. The apparatus of claim 34 , wherein the aggregation logic is operably coupled to receive a packet from the other aggregation device over the inter-device communication port, search the address table for a packet destination, determine that the packet destination is known, determine that the packet destination is associated with an active dual-homed split multilink trunk, and drop the packet without forwarding the packet over any port.
39. The apparatus of claim 34 , wherein the aggregation logic is operably coupled to receive a packet from the other aggregation device over the inter-device communication port, search the address table for a packet destination, determine that the packet destination is unknown, and forward the packet to all ports that are not associated with an active dual-homed SMLT port or with the other aggregation device.
40. The apparatus of claim 34 , wherein the aggregation logic is operably coupled to determine that an address has been aged by at least one aggregation device and to remove the address from its address table.
41. The apparatus of claim 40 , wherein the aggregation logic is operably coupled to determine that the address has been aged locally and to send a message to the other aggregation device over the inter-device communication port indicating that the address has been aged locally.
42. The apparatus of claim 40 , wherein the aggregation logic is operably coupled to receive a message from the other aggregation device over the inter-device communication port indicating that the address has been aged by the other aggregation device.
43. The apparatus of claim 27 , wherein the aggregation logic is operably coupled to determine that all links of its split multilink trunk have failed, send a message to the other aggregation device over the inter-device communication port informing the other aggregation device that all links of its split multilink trunk have failed, and redirecting all affected flows from the failed split multilink trunk to the inter-device communication port so that all packets previously destined for its split multilink trunk are forwarded to the other aggregation device over the inter-device communication port.
44. The apparatus of claim 27 , wherein the aggregation logic is operably coupled to receive a message from the other aggregation device indicating a split multilink trunk failure and to reconfigures each port associated with the split multilink trunk from a split multilink trunk logical port type to a multilink trunk logical port type.
45. The apparatus of claim 27 , wherein the communication connection protocol comprises at least one of: a hello message that is used to verify connectivity between the aggregation devices; a notification of learned or migrated MAC addresses message that is used to exchange learned or migrated MAC addresses between the aggregation devices; and a notification of split multilink trunk address age expired message that is used to exchange address aging information for addresses associated with the split multilink trunk between the aggregation devices.
46. The apparatus of claim 34 , wherein the aggregation logic is operably coupled to receive a packet from the other aggregation device over the inter-device communication port, search the address table for a packet destination, determine that the packet destination is known, determine that the packet destination is not associated with an active dual-homed split multilink trunk link, and forward the packet to all ports that are not associated with an active dual-homed SMLT port or with the other aggregation device.
47. A system comprising a client device coupled to two aggregation devices over a multilink trunk that is split between the aggregation devices, wherein the aggregation devices are operably coupled to work in conjunction such that the aggregation devices appear to the client device as a single device coupled to the client device through the multilink trunks by associating a unique identifier for the split multilink trunk and to using the unique identifier to identify the split multilink trunk.