Protected switching ring

• US 20040223503A1
• Filed: 05/06/2004
• Published: 11/11/2004
• Est. Priority Date: 05/06/2003
• Status: Active Grant

• Current Assignee: Adtran Networks SE (ADTRAN Holdings, Inc.)
• Original Assignee: Overture Networks, Inc. (ADTRAN Holdings, Inc.)
• Inventors: Lynch, Robert Leroy, Griswold, David Stephen, Swope, Glenn Russell, Pate, Prayson Will
• Granted Patent: US 6,928,050 B2
• Time in Patent Office: Days
• US Class Current: 370/404
• CPC Class Codes: H04L 12/42 Loop networks
• CPC Class Codes: H04L 12/437 Ring fault isolation or reconfiguration for SDH/SONET ring networks H04J3/085

Abstract

Normal 802.3 Ethernet requires a tree topology. If a ring or a loop exists, then packets will be forwarded around the ring indefinitely. If the ring is broken, then there is no possibility of packets being propagated forever. This invention shows how to quickly impose a virtual break in the ring such that all nodes can communicate with each other, and how to remove the virtual break when a real failure occurs. This is accomplished by placing intelligent nodes on the ring that work together to virtually break and restore the ring. An embodiment is disclosed that handles a unidirectional break in a communication link. This abstract is provided as an aid to those performing prior art searches and not a limitation on the scope of the claims.

Claims

1. A network ring that is manipulated to provide a virtual break in order to operate within a network requiring a tree topology, the network carrying data packets, the network ring comprising:

A) a master arbiter comprising;

a MA East port;

a MA West port; and

B) a Slave Arbiter comprising;

a SA East Port normally connected to the MA East Port by a full duplex communication connection;

a SA West port normally connected to the MA West Port by a full duplex communication connection;

a SA control system;

the SA control system comprising controls over the introduction and removal of a virtual break to prevent the formation of a communication ring for data packets within the network requiring a tree topology through the use of at least one non-forwarding state wherein control traffic is forwarded but data packets are not forwarded;

the SA control system comprising;

at least one rule causing the operation of a SA Port in a slave-forwarding state wherein the port forwards data packets and control traffic received at the port;

at least one rule causing the operation of a SA Port in one of at least one non-forwarding states wherein control traffic is forwarded but data packets are not forwarded;

at least one rule causing the Slave Arbiter to operate when the ring is fully operational such that one SA Port is in a slave-forwarding state and the other SA Port is in a non-forwarding state wherein control traffic is forwarded but data packets are not forwarded;

at least one rule for changing the state of the Slave Arbiter to have both the SA East Port and the SA West port acting to forward data packets after a functional break in the network ring is discerned; and

at least one rule for changing the state of one of the two SA Ports to a non-forwarding state wherein control traffic is forwarded but data packets are not forwarded after detection that the functional break has been removed.

2. The network ring of claim 1 further comprising:

C) at least one East Side Ring Relay on the ring segment between the MA East Port and the SA East Port;

 

the at least one East Side Ring Relay comprising two network ring ports and at least one user port in data communication with the network ring through the East Side Ring Relay;

D) a First Ring Relay Port on an East Side Ring Relay directly linked to the MA East Port, the First Ring Relay Port having a partner port on the Slave Arbiter Side of the East Side Ring Relay with the First Ring Relay Port; and

E) a Second Ring Relay port on an East Side Ring Relay directly linked to the SA East Port;

the Second Ring Relay Port having a partner port on the Master Arbiter Side of the East Side Ring Relay with the Second Ring Relay Port;

wherein the First Ring Relay Port is the partner port to the Second Ring Relay Port when there is only one East Side Ring Relay;

 

the operation of the network ring further characterized by;

the First Ring Relay Port relaying to the Master Arbiter'"'"'s MA East Port data packets received from its partner port that are not directed to exit over one of the at least one user ports are sent out the first port;

the Second Ring Relay Port relaying to the Slave Arbiter'"'"'s SA East Port data packets received from its partner port that are not directed to exit over one of the at least one user ports are sent out the first port;

each of the at least one East Side Ring Relay relaying data packets received from the at least one Subscriber Port connected to that East Side Ring Relay out both ring ports;

the First Ring Relay Port ceasing forwarding data packets upon detection of a functional break between the First Ring Relay Port and the Master Arbiter'"'"'s MA East Port;

the First Ring Relay Port resuming forwarding data packets after the detection of the removal of the functional break between the First Ring Relay Port and the Master Arbiter'"'"'s MA East Port;

the resumption delayed by a delay period sufficient to allow the Slave Arbiter to create a virtual break to prevent formation of a communication ring;

the Second Ring Relay Port ceasing forwarding data packets upon detection of a functional break between the Second Ring Relay Port and the Slave Arbiter'"'"'s SA East Port; and

the Second Ring Relay Port resuming forwarding data packets after the detection of the removal of the functional break between the Second Ring Relay Port and the Slave Arbiter'"'"'s SA East Port, the resumption delayed by a delay period sufficient to allow the Slave Arbiter to create a virtual break to prevent the formation of a communication ring.

3. The network ring of claim 1 further comprising:

C) at least one West Side Ring Relay on the ring segment between the MA West Port and the SA West Port;

 

the at least one West Side Ring Relay comprising two network ring ports and at least one user port in data communication with the network ring through the West Side Ring Relay;

D) a First Ring Relay Port on an West Side Ring Relay directly linked to the MA West Port, the First Ring Relay Port having a partner port on the Slave Arbiter Side of the West Side Ring Relay with the First Ring Relay Port; and

E) a Second Ring Relay port on an West Side Ring Relay directly linked to the SA West Port;

the Second Ring Relay Port having a partner port on the Master Arbiter Side of the West Side Ring Relay with the Second Ring Relay Port;

wherein the First Ring Relay Port is the partner port to the Second Ring Relay Port when there is only one West Side Ring Relay;

 

the operation of the network ring further characterized by;

the First Ring Relay Port relaying to the Master Arbiter'"'"'s MA West Port data packets received from its partner port that are not directed to exit over one of the at least one user ports are sent out the first port;

the Second Ring Relay Port relaying to the Slave Arbiter'"'"'s SA West Port data packets received from its partner port that are not directed to exit over one of the at least one user ports are sent out the first port;

each of the at least one West Side Ring Relay relaying data packets received from the at least one Subscriber Port connected to that West Side Ring Relay out both ring ports;

the First Ring Relay Port ceasing forwarding data packets upon detection of a functional break between the First Ring Relay Port and the Master Arbiter'"'"'s MA West Port;

the First Ring Relay Port resuming forwarding data packets after the detection of the removal of the functional break between the First Ring Relay Port and the Master Arbiter'"'"'s MA West Port;

the resumption delayed by a delay period sufficient to allow the Slave Arbiter to create a virtual break to prevent formation of a communication ring;

the Second Ring Relay Port ceasing forwarding data packets upon detection of a functional break between the Second Ring Relay Port and the Slave Arbiter'"'"'s SA West Port; and

the Second Ring Relay Port resuming forwarding data packets after the detection of the removal of the functional break between the Second Ring Relay Port and the Slave Arbiter'"'"'s SA West Port, the resumption delayed by a delay period sufficient to allow the Slave Arbiter to create a virtual break to prevent the formation of a communication ring.

4. The network ring of claim 2 further comprising:

F) at least one West Side Ring Relay on the ring segment between the MA West Port and the SA West Port;

 

the at least one West Side Ring Relay comprising two network ring ports and at least one user port in data communication with the network ring through the West Side Ring Relay;

G) a First Ring Relay Port on an West Side Ring Relay directly linked to the MA West Port, the First Ring Relay Port having a partner port on the Slave Arbiter Side of the West Side Ring Relay with the First Ring Relay Port; and

H) a Second Ring Relay port on an West Side Ring Relay directly linked to the SA West Port;

the Second Ring Relay Port having a partner port on the Master Arbiter Side of the West Side Ring Relay with the Second Ring Relay Port;

wherein the First Ring Relay Port is the partner port to the Second Ring Relay Port when there is only one West Side Ring Relay;

 

the operation of the network ring further characterized by;

the First Ring Relay Port relaying to the Master Arbiter'"'"'s MA West Port data packets received from its partner port that are not directed to exit over one of the at least one user ports are sent out the first port;

the Second Ring Relay Port relaying to the Slave Arbiter'"'"'s SA West Port data packets received from its partner port that are not directed to exit over one of the at least one user ports are sent out the first port;

each of the at least one West Side Ring Relay relaying data packets received from the at least one Subscriber Port connected to that West Side Ring Relay out both ring ports;

the First Ring Relay Port ceasing forwarding data packets upon detection of a functional break between the First Ring Relay Port and the Master Arbiter'"'"'s MA West Port;

the First Ring Relay Port resuming forwarding data packets after the detection of the removal of the functional break between the First Ring Relay Port and the Master Arbiter'"'"'s MA West Port;

the resumption delayed by a delay period sufficient to allow the Slave Arbiter to create a virtual break to prevent formation of a communication ring;

the Second Ring Relay Port ceasing forwarding data packets upon detection of a functional break between the Second Ring Relay Port and the Slave Arbiter'"'"'s SA West Port; and

the Second Ring Relay Port resuming forwarding data packets after the detection of the removal of the functional break between the Second Ring Relay Port and the Slave Arbiter'"'"'s SA West Port, the resumption delayed by a delay period sufficient to allow the Slave Arbiter to create a virtual break to prevent the formation of a communication ring.

5. The network ring of claim 1 wherein the Master Arbiter can operate as a slave arbiter and the Slave Arbiter can operate as a master arbiter and a set at least one control system rules operate to limit the network ring to a single master arbiter at any one time.

6. The network ring of claim 1 wherein the Master Arbiter further comprises at least one Subscriber Port and the data packets received from the at least one Subscriber Port are transmitted out both the MA East Port and the MA West Port.

7. The network ring of claim 1 wherein:

the MA East Port sends a control traffic message of a first type that is conveyed to the SA East Port;

and the SA control system is further characterized by comprising;

at least one rule for changing the state of the Slave Arbiter to have both the SA East Port and the SA West port acting to forward data packets after a functional break in the network ring is discerned; and

a process of discernment comprising monitoring the receipt of the control traffic messages of the first type received at the SA East Port and detecting a gap in the receipt of an incoming control traffic message of the first type at the SA East Port for a period in excess of a first time interval.

8. The network ring of claim 1 wherein:

the data packets within the network ring are forwarded using a switching technique; and

at least one type of control traffic is forwarded using a bridging technique.

9. The network ring of claim 1 wherein:

the data packets within the network ring are forwarded using a switching technique; and

at least one type of control traffic is forwarded using a switching technique.

10. The network ring of claim 1 wherein the control system further comprises an extended delay so that a port recovering from a port down state does not forward data packets until after the execution of the at least one rule for changing the state of one of the two SA Ports to a non-forwarding state wherein control traffic is forwarded but data packets are not forwarded whereby the removal of a functional break does not create a temporary loop for data packets and thus provides a safeguard against the duplication of data packets.

11. A network ring that is manipulated to provide a virtual break in order to operate within a network requiring a tree topology, the network carrying data packets, the network ring comprising:

A) a Slave Arbiter comprising;

a SA East Port normally connected to a first Ethernet device by a full duplex communication connection;

a SA West port normally connected to a second Ethernet device by a full duplex communication connection;

a SA control system;

the SA control system comprising controls over the introduction and removal of a virtual break to prevent the formation of a communication ring for data packets within the network requiring a tree topology through the use of at least one non-forwarding state wherein control traffic is forwarded but data packets are not forwarded;

the SA control system comprising;

at least one rule causing the operation of a SA Port in a slave-forwarding state wherein the port forwards data packets and control traffic received at the port;

at least one rule causing the operation of a SA Port in one of at least one non-forwarding states wherein control traffic is forwarded but data packets are not forwarded;

at least one rule causing the Slave Arbiter to operate when the network ring is fully operational such that one SA Port is in a slave-forwarding state and the other SA Port is in a non-forwarding state wherein control traffic is forwarded but data packets are not forwarded;

at least one rule for changing the state of the Slave Arbiter to have both the SA East Port and the SA West port acting to forward data packets after a functional break in the network ring is discerned; and

at least one rule for changing the state one of the two SA Ports to a non-forwarding state wherein control traffic is forwarded but data packets are not forwarded after detection that the functional break has been removed.

12. The network ring of claim 11 wherein the first Ethernet device is also the second Ethernet device.

13. A network ring that is manipulated to have at least one virtual break for data packets;

the network ring comprising;

a first arbiter acting as Master Arbiter, the Master Arbiter having at least a MA West Port, and a MA East Port;

a second arbiter acting as Slave Arbiter, the Slave Arbiter having at least a SA West Port, and a SA East Port;

a First full duplex communication path between the Master Arbiter and the Slave Arbiter;

a Second full duplex communication path between the Master Arbiter and the Slave Arbiter;

the Master Arbiter and the Slave Arbiter serving as nodes on the network ring;

the Master Arbiter sending out a HELLO protocol packet frequently onto the First full duplex path and on the Second full duplex path;

the Slave Arbiter noting the arrival of the HELLO protocol packets and noting a ring timeout when the interval between arrivals of two sequential HELLO protocol packets on a particular port exceeds a first threshold value;

the Slave Arbiter reacting to a ring timeout by removing at least one virtual break within the Slave Arbiter and communicating data packets received from the First Communication Path to the Second Communication Path and from the Second Communication Path to the First Communication Path; and

the Slave Arbiter reacts to receipt of a HELLO protocol packet from the communication path deemed to have a ring timeout by introducing at least one virtual break so that data packets received from the First Communication Path are not communicated to the Second Communication Path and the data packets received from the Second Communication Path are not communicated to the First Communication Path so that the network ring is not operated to provide a data communication loop.

14. A network ring that is manipulated to have at least one virtual break for data packets;

the network ring comprising;

a first arbiter acting as Master Arbiter, the Master Arbiter having at least a MA West Port, and a MA East Port;

a second arbiter acting as Slave Arbiter, the Slave Arbiter having at least a SA West Port, and a SA East Port;

a First full duplex communication path between the Master Arbiter and the Slave Arbiter comprising at least an “

A”

Ring Relay and a “

B”

Ring Relay;

a Second full duplex communication path between the Master Arbiter and the Slave Arbiter comprising at least an Initial Ring Relay;

the A Ring Relay, the B Ring Relay and the Initial Ring Relay all comprising a Master Arbiter Side Port, a Slave Arbiter Side Port, and a Subscriber Port;

the Master Arbiter;

the Slave Arbiter; and

the ring relays serving as nodes on the network ring;

the Master Arbiter sending out a HELLO protocol packet frequently onto the First full duplex path and on the Second full duplex path;

the ring relays passing the HELLO protocol packet from the Master Arbiter Side Port to the Slave Arbiter Side Port;

the Slave Arbiter noting the arrival of the HELLO protocol packets and noting a ring timeout when the interval between arrivals of two sequential HELLO protocol packets on a particular port exceeds a first threshold value;

the Slave Arbiter reacting to a ring timeout by removing at least one virtual break within the Slave Arbiter and communicating data packets received from the First Communication Path to the Second Communication Path and from the Second Communication Path to the First Communication Path; and

the Slave Arbiter reacts to receipt of a HELLO protocol packet from the communication path deemed to have a ring timeout by introducing at least one virtual break so that data packets received from the First Communication Path are not communicated to the Second Communication Path and the data packets received from the Second Communication Path are not communicated to the First Communication Path so that the network ring is not operated to provide a data communication loop.

15. The network ring of claim 14 wherein the Slave Arbiter notes a ring failure when the interval between arrivals of two sequential HELLO protocol packets exceeds a second threshold value, larger than the first threshold value.

16. The network ring of claim 14 wherein the first threshold value is selected to allow the declaration of a ring timeout and removal of at the least one virtual break within the Slave Arbiter in less than 50 milliseconds.

17. The network ring of claim 14 wherein the Slave Arbiter has at least one Subscriber Port and the Slave Arbiter continues to transmit data packets received from at least one Subscriber Port onto the First Communication Path and the Second Communication Path after noting the ring timeout, but after receipt of a HELLO protocol packet from the communication path deemed to have a ring timeout, the Slave Arbiter acts to prevents subsequent retransmission of these data packets by inserting at least one virtual break within the Slave Arbiter.

18. The network ring of claim 17 wherein the nodes closest to a failure in the network ring stop transmitting received data packets towards the failure in the network ring and do not resume transmission of data packets towards the failure in the network ring after correction of the failure until after the Slave Arbiter has an opportunity to insert at least one virtual break.

19. The network ring of claim 14 wherein the HELLO protocol packets are transported within the network ring through use of bridging procedures to direct the HELLO protocol packets to particular nodes on the network ring, but data packets received from user ports are transported around the network ring through the use of a switching technique.

20. The network ring of claim 14 wherein the HELLO protocol packets and data packets are transported within the network ring through use of a switching technique to direct the packets to particular nodes on the network ring.

21. A network node device for selectively creating and removing at least one virtual break in a network ring;

the node device comprising;

a first ring connection and a second ring connection for use in connecting the node device to two ends of a network ring;

the node device operation within a network ring characterized by;

initializing so that the node device contains a virtual break to prevent data packets received on the first ring connection from continuing on the second ring connection and data packets on the second data connection from continuing on the first ring connection;

creating and transmitting a series of HELLO protocol packets out the first ring connection and the second ring connection;

receiving HELLO protocol packets from a device other than itself on the first ring connection and monitoring the interval since the last HELLO protocol packet was received on the first ring connection;

receiving HELLO protocol packets from a device other than itself on the second ring connection and monitoring the interval since the last HELLO protocol packet was received on the second ring connection;

discerning the existence of a break in the network ring by detecting an interval in excess of a first value since the receipt of the last HELLO packet on the first ring connection;

discerning the existence of a break in the network ring by detecting an interval in excess of a second value since the receipt of the last HELLO packet on the second ring connection where the second value can equal the first value;

responding to the discernment of the existence of a break in the network ring by removing the at least one virtual break within the node device;

detecting the resumed receipt of a HELLO protocol packet on a ring connection used to discern a break in the network ring, and concluding the discerned break is gone; and

introducing at least one virtual break within the node device after concluding that the discerned break is gone.

22. The node device of claim 21 wherein the node device inspects incoming HELLO protocol packets to discern whether the node device has the duty of selectively imposing at least one virtual break or whether that duty is given to a device other than itself.

23. The node device of claim 21 wherein the node device does use dynamic layer 2 learning tables for transmitting data packets.

24. The node device of claim 21 further comprising at least one user port wherein:

during a period when the node device has imposed a virtual break, there is only one path for data packets to travel from the network ring to egress the network ring to the at least one user port;

during the period when the node device has imposed a virtual break, there is only one path for data packets to travel from the at least one user port to the network ring; and

during a period when the node device has removed the virtual break, data packets arriving at the node device from one of the at least one user port are sent out both the first ring connection and the second ring connection.

25. The node device of claim 24 wherein:

the data packets received at the node device from the at least one user port are modified to add a ring tag specifying a connection between two ring nodes; and

the data packets received at the node device from the network ring where the data packets are destined for one of the at least one user ports on the node device are stripped of the ring tag added to the data packet at the node where the data packet entered the network ring.

26. The node device of claim 21 wherein the at least one virtual break is always placed on the first ring connection so that:

data packets destined for the at least one user port on the node device reach the user port through the second ring connection when the at least one virtual break is imposed;

data packets from the at least one user port are transmitted exclusively on the second ring connection when the at least one virtual break is imposed; and

at least one type of network control packet is sent out of the first ring connection even when the at least one virtual break is imposed in the node device.

27. A ring relay for use in a network ring that is manipulated to provide a virtual break in order to operate within a network requiring a tree topology, the network ring carrying data packets;

the ring relay comprising the following attributes;

having a first ring port and a second ring port connected to the network ring;

detection of a type of network ring failure and moving the relevant port to a port down state; and

delaying resumption of operation after detecting the removal of the ring failure so that the transition from the port down state to operating is delayed sufficiently to allow another node in the network ring to impose virtual breaks to prevent the formation of a communication loop for data packets.

28. The ring relay of claim 27 further comprising at least one user port connected to the network ring through the ring relay;

the characteristics of the ring relay further comprising;

adding a network ring VLAN tag, unique on the network ring, to any data packet entering the network ring through the ring relay; and

removing a network ring VLAN tag, previously added by a network ring node, from any data packet leaving the network ring through the ring relay.

29. The ring relay of claim 28 where data packets received on the ring first port are simply transferred to the second ring port, without modification, if a switch table, based on the network ring VLAN tag, indicates that the ring relay is a transit node for that data packet, rather than an endpoint for that data packet.

30. The ring relay of claim 27 where data packets received on the first ring port are simply transferred to the second ring port, without modification, if a switch table indicates that the ring relay is a transit node for that data packet rather than an endpoint for that data packet.

31. The ring relay of claim 27 wherein the ring relay periodically receives discovery messages on the first ring port and adds information about the ring relay to the discovery message before sending it out the second ring port.

32. The ring relay of claim 31 wherein the ring relay will create a discovery message indicating that the ring relay is the originator of the discovery message and send that discovery message out the second ring port if a discovery message is not received on the first ring port within a specified time interval.

33. The ring relay of claim 31 wherein the ring relay periodically receives hello messages on the first ring port and sends the hello message out the second ring port without processing so that hello messages travel around a portion of the network ring faster than discovery messages travel around that same portion.

34. A ring relay for use in a network ring that is manipulated to provide a virtual break in order to operate within a network requiring a tree topology, the network ring carrying data packets;

the ring relay comprising the following attributes;

having a first ring port and a second ring port connected to the network ring;

detecting a type of network ring failure and moving the relevant port to a port down state; and

delaying resumption of operation of the port in the port down state until after receiving a control signal message from another node indicating the insertion of at least one virtual break to prevent the formation of a communication loop for data packets.

35. A network having a redundant path for increasing the availability of a segment of a network ring, the network comprising:

an existing network ring;

a Master Arbiter at a first access point on the existing network ring;

a Slave Arbiter at a second access point, different from the first access point, on the existing network ring;

a second network ring formed with a ring side and an extension side;

the ring side comprising the Master Arbiter, the Ring Side port on the Master Arbiter, the Ring-Side Port on the Slave Arbiter, and the Slave Arbiter;

the extension side comprising the Master Arbiter, the Extension Side Port on the Master Arbiter;

the first access point, the portion of the existing network ring between the first access point and the second access point, the second access point, the Extension Side Port on the Slave Arbiter, and the Slave Arbiter;

the Master Arbiter passing data packets received on the Extension Side Port to the Ring Side Port and the data packets received on the Ring Side Port to the Extension Side Port;

the Slave Arbiter when acting in a blocking mode, blocking the passage of data packets from the Extension Side Port to the Ring Side Port and from the Ring Side Port to the Extension Side Port but forwarding at least one type of control packet from the Extension Side Port to the Ring Side Port and from the Ring Side Port to the Extension Side Port;

the Slave Arbiter when acting in a forwarding mode, forwarding data packets from the Extension Side Port to the Ring Side Port and from the Ring Side Port to the Extension Side Port;

the Master Arbiter and the Slave Arbiter monitoring the performance of the second network ring to detect a failure in the second network ring within a first specified period and switching the Slave Arbiter from acting in a blocking mode to acting in a forwarding mode within a second specified period; and

the Slave Arbiter reacting to an indication that the failure in the second network ring has been remedied by switching from acting in forwarding mode to acting in blocking mode; and

at least one port on the second network adjacent to the failure in the second network ring noting the failure in the second network ring and switching to a port down mode, the at least one port delaying recovery from a port down condition long enough for the Slave Arbiter to switch from acting in a forwarding mode to acting in a blocking mode to prevent the formation of a communication loop for data packets.

36. The network of claim 35 wherein the total time for:

the actions of the Master Arbiter and the Slave Arbiter monitoring the performance of the second network ring to detect a failure in the network ring within a first specified period and

the time required for switching the Slave Arbiter from acting in a blocking mode to acting in a forwarding mode is less than 50 milliseconds.

37. A method of providing a redundant connection for at least one user port for increasing the availability access between the at least one user port and an existing network;

the method comprising;

connecting the at least one user port to a slave arbiter node;

connecting a first ring port on the slave arbiter node in bidirectional data communication with a first network access device connected to the existing network;

connecting a second ring port on the slave arbiter node in bidirectional data communication with a second network access device connected to the existing network;

sending HELLO messages out the first ring port to the second ring port via the first network access device and the second network access device;

operating in a blocking mode when the ring comprising the slave arbiter node, the first access device, and the second access device is functioning, wherein blocking mode comprises;

blocking one ring port on the slave arbiter node, by blocking the transmission of data packets from the at least one user port from leaving the blocked ring port to travel to the corresponding access device; and

blocking data packets received on the blocked ring port from reaching the at least one user port;

operating in a forwarding mode when the slave arbiter detects a problem with a communication path between the unblocked ring port and the existing network, wherein forwarding mode comprises;

sending data packets received from the at least one user port out the previously blocked ring port; and

sending data packets received from the previously blocked ring port to the at least one user port.

38. The method of claim 37 wherein the frequency of HELLO messages is sufficient to allow both the detection of a problem with the communication path between the unblocked port and the existing network and switching from blocking mode to forwarding mode in less than 50 milliseconds.

39. The method of claim 37 wherein HELLO messages are sent out the second ring port to the first ring port via the second network access device and the first network access device.

40. A ring relay for use in a bidirectional network ring that is manipulated to provide a virtual break in order to operate within a network requiring a tree topology, the network ring carrying data packets;

the ring relay comprising the following attributes;

having a first ring port in data communication with a first incoming link and a first outgoing link and a second ring port in data communication with a second incoming link and a second outgoing link connected to the network ring; and

after detecting a type of network ring failure on the second incoming link, reacting to the detection of the network failure on the second incoming link by blocking the second outgoing link such that a virtual break is imposed on the second outgoing link after detection of a type of network failure on the second incoming link.

41. The ring relay of claim 40 wherein the ring relay is further characterized by:

receiving HELLO messages on the first incoming link and passing the HELLO message out the second outgoing link;

receiving data packets on the first incoming link destined for the second outgoing link and passing these data packets to the second outgoing link; and

after detecting a type of network ring failure on the second incoming link, reacting to the detection of the network failure on the second incoming link by blocking the HELLO messages received on the first incoming link from passing onto the second outgoing link and blocking data packets received on the first incoming link destined for the second outgoing link from passing to the second outgoing link such that a virtual break is imposed on the second outgoing link after detection of a type of network failure on the second incoming link.