Method and apparatus for preventing loops in a network by controlling broadcasts

• US 8,615,599 B1
• Filed: 03/31/2006
• Issued: 12/24/2013
• Est. Priority Date: 03/31/2006
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Takefman, Michael, Ervin, Jimmy P., Alexander, James S., Bhagwatula, Parveen, Faulkner, Steven J.
• Primary Examiner(s): Pollack, Melvin H
• Application Number:
  US11/278,314
• Time in Patent Office: 2,825 Days
• Field of Search: 709/236
• US Class Current: 709/236
• CPC Class Codes: H04L 12/437 Ring fault isolation or reconfiguration for SDH/SONET ring networks H04J3/085
• CPC Class Codes: H04L 12/4637 Interconnected ring systems
• CPC Class Codes: H04L 41/0654 using network fault recovery ring fault isolation or reconfiguration in loop networks without recovery actions by a network management system H04L12/437
• CPC Class Codes: H04L 41/0816 the condition being an adaptation, e.g. in response to network events
• CPC Class Codes: H04L 45/18 Loop-free operations
• CPC Class Codes: H04L 45/28 using route fault recovery

Abstract

Methods and apparatus for eliminating loops associated with interconnects between clusters are disclosed. According to one aspect of the present invention, a method for operating a cluster of an interconnect group that includes at least one node and is in communication with a plurality of interconnects includes broadcasting a first frame on the interconnects. The method also includes receiving a second frame, the second frame being received on only a first interconnect, and blocking the second frame from being broadcast on the first interconnect. Finally, the method includes broadcasting the received second frame throughout the cluster such that no node in the cluster receives more than one copy of the second frame.

Claims

1. A method comprising:

at a node that is part of a first cluster of nodes arranged in a ring configuration in a network and configured to communicate with every other one of the nodes of the first cluster, broadcasting a first frame to a second cluster of nodes in the network on a plurality of point-to-point interconnects that are configured to carry broadcast and unicast communications and that are connected between the first cluster and the second cluster;

at the node that is part of the first cluster, receiving from a node in the second cluster a second frame on a first interconnect of the plurality of interconnects;

broadcasting the second frame to nodes within the first cluster;

at nodes in the first cluster, recognizing the first interconnect of the plurality of interconnects and that the second frame is a received frame; and

at nodes in the first cluster that control a corresponding one of the plurality of interconnects and which receive the second frame, blocking the second frame from being returned to the second cluster from the first cluster on any of the plurality of interconnects after broadcasting the second frame to nodes within the first cluster and recognizing the first interconnect of the plurality of interconnects.

2. The method of claim 1, and further comprising at a node in the second cluster that receives the first frame, broadcasting the first frame directly to nodes in a subcluster of nodes within the second cluster over one of the interconnects.

3. The method of claim 1, and further comprising broadcasting the second frame received from a node in a particular subcluster of nodes of a plurality of subclusters of nodes within the first cluster to remaining plurality of subclusters within the first cluster.

4. The method of claim 1, further comprising:

determining if one of the plurality of interconnects has failed; and

if it is determined that one of the plurality of interconnects has failed, removing the failed interconnect as an available interconnect for communications between the first cluster and the second cluster and reallocating the remaining interconnects.

8. The method of claim 1, and further comprising:

sending a first unicast message on one of the plurality of interconnects from the node that is part of the first cluster to a filter device in the second cluster;

receiving at a filter device in the first cluster a second unicast message from a node in the second cluster that is intended to be delivered to a node in a particular subcluster of nodes within the first cluster;

identifying the intended node of the particular subcluster of nodes; and

directing the unicast message from the filter device in the first cluster to the identified intended node of the particular subcluster and not to any other nodes of subclusters within the first cluster.

9. The method of claim 1, and further comprising:

receiving at a filter device in the first cluster a message from a particular subcluster of a plurality of subclusters in the first cluster;

determining at the filter device which of the plurality of interconnects the message was sent over to reach the filter device in the first cluster;

identifying that the message was sourced from the particular subcluster based on the determining; and

blocking the message at the filter device in the first cluster from being sent back to the particular subcluster.

10. The method of claim 1, and further comprising sending messages from nodes in corresponding ones of a plurality of subclusters of nodes in the first cluster via an associated one of the plurality of interconnects and receiving messages from nodes in the first cluster intended for nodes in corresponding subclusters in the first cluster via an associated one of the plurality of interconnects.

15. The method of claim 1, further comprising receiving from a plurality of nodes in the second cluster a plurality of frames on more than one of the plurality of interconnects.

16. The method of claim 4, wherein reallocating comprises reallocating the plurality of interconnects that have not failed to the nodes of the first cluster and second cluster that correspond to the one of the plurality of interconnects that has failed.

17. The method of claim 16, further comprising load balancing the plurality of frames across the plurality of interconnects such that the plurality of frames are received only at nodes of interest in the first cluster.

18. The method of claim 16, further comprising identifying subclusters of nodes of the first cluster based on logical communication attributes and forwarding the frames by two or more nodes to the identified subclusters of nodes of the first cluster.

19. The method of claim 17, further comprising forwarding the plurality of frames by two or more nodes in the first cluster to remaining nodes in the first cluster based on network topology.

20. The method of claim 19, wherein forwarding comprises identifying network topology based on specific virtual local area networks (VLANs).

21. The method of claim 2, further comprising broadcasting the first frame to multiple subclusters of nodes within the first cluster on the interconnects such that each one of the interconnects is associated with a particular one of the multiple subcluster of nodes within the first cluster.

22. The method of claim 1, wherein broadcasting the first frame comprises broadcasting the first frame to the second cluster on all of the plurality of interconnects between the first cluster and the second cluster of nodes in the network such that the first frame is transmitted to multiple subclusters of nodes within the second cluster, wherein receiving the second frame at the first cluster comprises receiving the second frame that is broadcast on all of the plurality of interconnects to the first cluster, and wherein blocking comprises blocking the second frame from being returned to any of the subclusters of nodes in the second cluster from the first cluster on any of the plurality of interconnects.

5. A system comprising:

a plurality of nodes arranged in a ring configuration in a cluster within a network; and

a plurality of interconnects configured to connect the cluster to a corresponding one of a plurality of subclusters of nodes in an other cluster of nodes in the network, wherein the plurality of interconnects are configured to carry broadcast and unicast communications between nodes in the cluster and nodes in the other cluster to receive messages at nodes in the cluster from nodes in the other cluster and to send messages from the nodes in the cluster to nodes in the other cluster, wherein the plurality of interconnects are configured to receive a frame broadcast to nodes within the cluster from the other cluster; and

wherein nodes of the cluster that are configured to control corresponding interconnects are configured to determine and recognize one of the plurality of interconnects on which the frame was transmitted from the other cluster to the cluster and to block the frame from being returned back to the other cluster on any one of the plurality of interconnects once the nodes of the cluster determine that the frame originated on one of the plurality of interconnects.

6. The system of claim 5, wherein the plurality of nodes in the cluster are configured as a resilient packet ring.

7. The system of claim 5, wherein the plurality of interconnects includes an Ethernet channel.

11. A system comprising:

a first cluster of nodes in a network comprising a plurality of subclusters each of which comprises one or more nodes;

a second cluster of nodes in the network in a ring configuration; and

a plurality of interconnects between the first cluster and the second cluster, wherein the interconnects are configured to carry broadcast and unicast communications;

wherein each of the interconnects is designated to send messages to and receive messages from one or more nodes in a corresponding subcluster;

wherein the plurality of interconnects are configured to receive a first frame broadcast to the plurality of interconnects from one or more nodes in the first cluster and to send the frame to one or more nodes in the second cluster, and to receive a second frame broadcast from one or more nodes in the second cluster to a particular subcluster of the first cluster on a particular interconnect that is dedicated to the particular subcluster; and

wherein nodes in the first cluster which are configured to control a corresponding one of the plurality of interconnects are configured to recognize particular interconnects that receive the first frame and the second frame and to block the second frame from being returned to the second cluster from the first cluster on any of the plurality of interconnects.

12. The system of claim 11, wherein nodes in the second cluster that receive the first frame are configured to broadcast the first frame directly to nodes in their subcluster over one of the interconnects.

13. The system of claim 11, wherein nodes in the first cluster that receive the second frame are configured to broadcast the second frame to nodes within their corresponding subclusters.

14. The system of claim 11, and further comprising a filter device in the first cluster, wherein the filter device is configured to:

receive a message from the second cluster;

determine which of multiple interconnects of the plurality of interconnects the message was sent over to reach the filter device in the first cluster; and

block the message at the filter device in the first cluster from being sent back to the second cluster.