System for establishing new path in ring network and dividing ring shaped path into plurality of ring shaped paths by being rid of faulty path

• US 5,712,968 A
• Filed: 07/08/1992
• Issued: 01/27/1998
• Est. Priority Date: 07/08/1991
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Mori, Kenjiro, Nakayama, Yoshiyuki, Nakamura, Fumio, Yamamitsu, Tadashi, Ishizaki, Takeshi
• Primary Examiner(s): Meky, Moustafa M.
• Application Number:
  US07/910,454
• Time in Patent Office: 2,029 Days
• Field of Search: 395/200, 395/575, 395/200.11, 395/200.21, 395/182.02, 395/200.04, 370/14, 370/16, 370/16.1, 370/85.5, 370/216, 370/221-228, 371/8.2, 371/11.1, 371/11.2, 371/20.6
• US Class Current: 714/4.4
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: H04L 12/1813 for computer conferences, e.g. chat rooms instant messaging H04L51/04; protocols for multimedia communication H04L65/1101; arrangements for multi-party communication H04L65/403; telephonic conference arrangements H04M3/56; television conference systems H04N7/15
• CPC Class Codes: H04L 12/437 Ring fault isolation or reconfiguration for SDH/SONET ring networks H04J3/085

Abstract

A conference system is formed by multiple workstations which have individual conference control programs and are connected to a network, with each workstation having a destination workstation determined in advance so that a ring-shaped logical communication path is formed on the network. A conference control program which detects a fault on its receiving path during a conference sends a fault recovery request message including an identifier of the conference control program itself and the address of the workstation to the neighboring workstation, and the conference control program of the neighboring workstation relays the received fault recovery request message to the next workstation on the ring-shaped logical communication path so that the fault recovery request message is propagated on the ring-shaped logical communication path. The conference control program of a workstation which is located immediately upstream of the fault and is therefore unable to send the received fault recovery request message to the neighboring workstation carries out a procedure of establishing a new communication path to the conference control program of the workstation which issued the fault recovery request message.

Claims

1. A method of controlling a ring-shaped logical communication path formed in a network which connects a plurality of node means through respective inter-node communication paths, each of said node means operating to relay a message to a predetermined adjacent node means on a respective inter-node communication path so that said ring-shaped logical communication path is formed, said method comprising the steps of:

generating, by one of said plurality of node means that has detected a fault on its message receiving communication path, a control message indicative of a fault recovery request including a node identifier assigned to the node means that has detected the fault and sending the control message to an adjacent node means on a respective inter-node communication path of said ring-shaped logical communication path;

transferring, by each node means that has received the control message, the received control message to the adjacent node means on the respective inter-node communication path of said ring-shaped logical communication path; and

establishing, by a node means that has failed to transfer the received control message to the adjacent node means on the respective inter-node communication path, a new inter-node communication path for message transmission directly to the node means on the respective inter-node communication path of the downstream side thereof indicated by the node identifier included in the received control message, the new inter-node communication path not including any node means which has transferred the received control message to the adjacent node means.

2. A control method for a ring-shaped logical communication path according to claim 1 including:

releasing, by said node means that has failed to transfer the received control message to the adjacent node means, the inter-node communication path on the message reception side thereof so that said node means that has failed to transfer the received control message to the adjacent node means retreats from said ring-shaped logical communication path if said node means that has failed to transfer the received control message to the adjacent node means has judged that a new inter-node communication path to said downstream node means indicated by the node identifier included in the control message cannot be established;

generating, by said node means that has generated the fault recovery request message, a second fault recovery request message including a node identifier assigned to the node means that has detected the fault and sending said second message to the adjacent node means on the respective inter-node communication path of said ring-shaped logical communication path if a new inter-node communication path is not established on the message reception side thereof within a prescribed time limit following the generation of the first control message or following the transmission of the message to the adjacent node means;

transferring, by each node means that has received the second control message, the received second control message to the adjacent node means on the respective inter-node communication path of said ring-shaped logical communication path; and

establishing, by a node means that has failed to transfer the received second control message to the adjacent node means on the respective inter-node communication path, a new inter-node communication path for message transmission to the node means on the respective inter-node communication path of the downstream side thereof indicated by the node identifier included in the received second control message.

3. A control method for a ring-shaped logical communication path according to claim 1 including:

establishing, by said node means that has failed to transfer the received control message to the adjacent node means on the respective inter-node communication path, a new communication path for message transmission to the downstream node means on the respective inter-node communication path indicated by the node identifier included in the control message by using a communication network different from the network in trouble.

4. A control method for a ring-shaped logical communication path according to claim 1 including:

establishing, by said node means that has failed to transfer the received control message to the adjacent node means on the respective inter-node communication path, a new communication path to the downstream node means on the respective inter-node communication path indicated by the node identifier included in the control message, the new communication path to the downstream node means passing through a relay device having a relay function for relaying messages from the node means that has failed to transfer the received control message to the adjacent node means on the respective inter-node communication path to the downstream node means.

5. A method of controlling a ring-shaped logical communication path formed in a network which connects a plurality of node means, each of said node means operating to relay a message to a predetermined adjacent node thereof so that said ring-shaped logical communication path is formed, said method comprising:

generating, by one of node means that has detected a fault on its message receiving communication path, a control message indicative of a fault recovery request including a node identifier assigned to the node means that has detected the fault and sending the control message to the adjacent node on said ring-shaped logical communication path;

transferring, by each node means that has received the control message, the received control message to the adjacent node means on said ring-shaped logical communication path; and

establishing by a node means that has failed to transfer the control message to the adjacent node, a new communication path for message transmission to the node means indicated by the node identifier included in the received control message, so that said ring-shaped logical communication path is divided into a plurality of ring-shaped logical communication paths by being rid of a faulty path.

6. A conference system including a plurality of workstations connected in a network, each of said workstations comprising:

display means having a multi-window display function;

input means for entering data and/or a command to an arbitrary window created on the display screen of said display means;

data processing means for processing the data and/or command entered through said input means in accordance with a program which corresponds to the window where the data and/or command has been entered, and displaying the result of process in said window of said display screen; and

interface means for transacting data with said network, said data processing means including conference control means which correlates one of said workstations to an input logical communication path and another workstation to an output logical communication path, sends a message which includes data and/or command entered from said input means and said interface means to the workstation corresponding to the output logical communication path through said interface means, and operates on a program to process said entered data and/or command,

said conference control means including

first means for generating, on detecting abnormality of data reception of said interface means from said network, a control message including a control command for fault recovery which includes an identifier assigned to said conference control means, and sending the message to the workstation corresponding to the output logical communication path, and

second means for establishing, if said first means fails to send a fault recovery control message generated by any workstation and received by said interface means to the workstation corresponding to the output logical communication path, a new communication path to the workstation having the conference control means indicated by the identifier included in the control message, and correlating, to the output logical communication path, the workstation having the conference control means indicated by the identifier included in the control message.

7. A conference system according to claim 6, wherein said second means includes means for dissolving the input logical communication path if said second means fails to establish a new communication path to the workstation having the conference control means indicated by the identifier included in the control message so that the workstation having the second means which fails to establish a new communication path to the workstation having the conference control means indicated by the identifier included in the control message retreats from the conference system.

8. A conference system according to claim 6, wherein said first means includes means for sending recurrently the fault recovery control message if said first means fails to receive a signal for establishing a new communication path from said interface means within a prescribed time limit following the generation or transmission of said control message.

9. A conference system according to claim 6, wherein said conference control means includes means for displaying on said display screen, on detecting an abnormality due to communication failure, a message for inquiring of a user of the conference control means whether or not automatic recovery of a communication path is to be tried, said first means generating and sending the control message in response to entry by the user of a command indicative of automatic recovery from said input means.

10. A method of controlling a ring-shaped logical communication path formed in a network which connects a plurality of node means through respective inter-node communication paths, each of said node means operating to relay a message to a predetermined adjacent node means on a respective inter-node communication path so that said ring-shaped logical communication path is formed, said method comprising:

generating, by each of at least two of said plurality of node means that have detected respective faults on their respective message receiving communication paths, a control message indicative of a fault recovery request including a node identifier assigned to the node means that has detected the fault and sending the control message to an adjacent node means on a respective inter-node communication path of said ring-shaped logical communication path;

transferring, by each node means that has received the control message, the received control message to the adjacent node means on the respective inter-node communication path of said ring-shaped logical communication path; and

establishing, by each of at least two node means that have failed to transfer the control message to the adjacent node means on the respective inter-node communication path, a new inter-node communication path for message transmission directly to the node means on the respective inter-node communication path of the downstream side thereof indicated by the node identifier included in the received control message, each new inter-node communication path not including any node means which has transferred the received control message to the adjacent node means, so that said ring-shaped logical communication path is divided into a plurality of ring-shaped logical communication paths by being rid of at least two faults.