JPH04302240A - Fault recovery system for loop type communication system - Google Patents

Abstract

PURPOSE:To recover a fault without using any artificial means when a station disconnected by the generation of the fault is normal by confirming whether there is the information of the fault generation a[ the station informed for disconnection of a port or not and informing the port separation again when there is the information. CONSTITUTION:When the fault generation is detected, a station 2 informs the fault generation of a control station 1 and stops monitor for prescribed time. When the fault generation is informed from the station 2, the control station 1 informs the port disconnection of the station 2 having the possibility of the fault generation until the information of the fault generation is eliminated. Afterwards, the port recovery is successively informed of the station 2 informed the port disconnection after the information of the fault generation is eliminated, and it is confirmed whether there is the information of the fault generation or not. When there is the information of the fault generation, the port disconnection is informed of the station again. Therefore, when the station disconnected by the fault generation is normal, the fault can be recovered without requiring any artificial means.

Description

[Detailed description of the invention]

[Object of the invention]

0002

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a failure recovery method in a loop communication system.

0003

2. Description of the Related Art A loop communication system, which is a type of so-called LAN, is constructed by connecting a control station and stations in a loop via a transmission path. A terminal is connected to each station directly or via a PBX or bridge (BRG). When a station connects a terminal via a BRG, it uses a packet-switched media access controller (PXMA
C) accommodates.

[0004] Each station communicates by circulating frames on a transmission path. Furthermore, terminals connected via the BRG communicate using the packet exchange area of this frame.

By the way, such a loop communication system has a fault recovery processing routine in case an abnormality occurs in the station.

[0006] First, each station monitors whether or not a fault detection frame is flowing from a station logically one upstream on the loop (hereinafter simply referred to as the station one upstream).

[0007] When a station detects a fault detection frame, it notifies the station one upstream to disconnect the PX MAC port and below of that station.

[0008] Furthermore, even after this notification is made, if the station that has detected the failure detection frame does not recover from the failure, it performs processing to disconnect the ports below the PX MAC connected to the own station.

Furthermore, when multiple failures occur,
Such processing is continued until the failure is recovered.

However, in processing such a plurality of failures, a normal station may exist among the disconnected stations. After this point, even if the station is normal, it will not recover unless artificial measures are taken.

[0011]

[Problem to be Solved by the Invention] In the above-described fault recovery processing in the loop communication system, when a fault occurs in multiple stations, a normal station may be included in the separated stations. Therefore, even if the station is normal, there is a problem that it will not recover unless artificial measures are taken.

[0012] The present invention has been made based on the above circumstances, and provides loop-type communication that recovers without taking any artificial means if the station that has been disconnected due to the occurrence of a failure is a normal station. The purpose is to provide a failure recovery method for systems. [Structure of the invention]

[0013]

[Means for Solving the Problems] In the present invention, a plurality of stations including a control station are connected in a loop via a transmission line, a terminal is connected to each of the stations in a bus via a port, and the transmission line is connected to a terminal. In a loop type communication system, communication is performed between the stations by circulating a frame, and terminals connected to the station in a bus form communicate using a packet exchange area of the frame. When detecting the occurrence of a fault, the control station notifies the control station of the fault occurrence and suspends monitoring for a predetermined period of time, and when the control station receives a notification of the fault occurrence from the station, the control station The station is notified to disconnect the port until there is no more notification of the failure occurrence, and after the notification of the failure occurrence is gone, the stations that have been notified to disconnect the port are sequentially notified to recover the port. The above-mentioned problem is solved by checking whether there is a notification of the occurrence of a failure, and when there is a notification of the occurrence of a failure, notifying the station again to disconnect the port.

[0014]

[Operation] That is, in the present invention, when the station detects the occurrence of a fault, it notifies the control station of the occurrence of the fault and stops monitoring for a predetermined period of time. Inform the station where the problem may have occurred to disconnect the port until there is no more notification of the problem.
After there are no more failure notifications, the station that notified the port to be disconnected is sequentially notified to recover the port, and then checks to see if there is a failure notification. Since the station is notified to disconnect the port, if the station disconnected due to the occurrence of a failure is a normal station, recovery can be made without taking any artificial measures.

[0015]

Embodiments Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a diagram showing the configuration of a loop communication system according to an embodiment of the present invention, in which control station 1, station 2...
are connected in a loop via a transmission line 3. The terminal 4 is connected to the station 2 directly, the terminal 4 is connected via a PBX 5, or the terminal 4 is connected via a bridge (BRG) 6 in the form of a bus. When terminal 4 is connected via BRG 6, a packet switching media access controller (
PX MAC) 7 is inserted.

FIG. 2 is a diagram showing the configuration of the PX MAC 7, which includes a station side interface circuit 8, a terminal side interface circuit 9, a control section 10, and a MAC controller (MA
CC) 11, memory 12, and buses 13 and 14.

FIG. 3 shows the format of a frame circulating on the transmission line 3, which consists of a frame synchronization area, a control area, and a data communication area. The frame synchronization area indicates the beginning of the frame and is composed of a unique pattern that does not appear in other areas. The control area is an area for performing system control such as loop control, circuit switching connections, and time slot management. The data communication area consists of a circuit switching area consisting of a plurality of time slots and a packet switching area for performing packet communication.

In normal times, each station 2 communicates by circulating frames on the transmission path 3. Also, BRG
Terminal 4 connected via 6 performs communication using the packet exchange area of this frame.

Next, a description will be given of the failure recovery processing performed when an abnormality occurs in the station in the system of this embodiment.

Here, the control station 1, as shown in FIG.
It has a status management table that manages the status of each station 2. The status management table shown in the figure has an entry (area) corresponding to each station 2, and each entry includes at least the serial number of the station, the recovery required FLAG, and the disconnection processing F.
Information such as LAG is stored.

The failure recovery processing operation in this system will be explained below based on FIGS. 5 to 7. Note that FIG. 5 shows the processing flow of control station 1, FIG. 6 shows the processing flow of station 2, and FIG. 7 shows the processing flow of the entire system when an abnormality occurs in one station. .

A message is received from upstream (step 601), and the result of analyzing this message (step 60
2), the station where the MACC abnormal station (Figure 7 (1a)) was detected (
In FIG. 7(1b)), it is determined whether loop control is in progress (step 603).

[0024] During loop control, for example, the system is performing processing such as loopback.

Therefore, when loop control is in progress, no further processing is performed and the normal reception state is returned.

On the other hand, when loop control is not in progress, PX
A command to stop monitoring is output to the MAC (step 60).
4) Sends a message indicating that there is an abnormal station to the control station (step 605, FIG. 7 (2)), and sets a monitoring stop command timer (step 606, FIG. 7 (3))
.

Message received (step 501)
If the processing for the abnormal station is not continuing (step 502), the control station (A-SVS) analyzes the message (step 503), and then determines whether loop control is in progress (step 504).

Here, if loop control is in progress, no further processing is performed and the normal reception state is returned.

On the other hand, if loop control is not in progress, refer to the status management table, specifically refer to the serial number, and select the station one upstream from the station that sent the MACC abnormal station message. Confirm (step 505).

[0030] Subsequently, the detachment processing FLAG of the state management table corresponding to the station one upstream is set (step 506).

[0031]Furthermore, a port disconnection request is notified to the station one upstream (step 507, FIG. 7(
4)).

[0032] The station that received the port disconnection request is PXM.
Processing is performed to disconnect the own port from the AC (step 607).

[0033] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609).
Figure 7 (5)).

The control station that receives the port disconnection response sets a monitoring timer (step 510).

On the other hand, the station whose monitoring stop command timer has timed out requests the PX MAC to start monitoring (step 610, FIG. 7(6)).

[0036] In this case, an abnormality has occurred in one station and the port of this station has been disconnected.
A message indicating that there is a CC abnormal station will not be sent.

[0037] Therefore, after the control station times out,
A search is made to see if the recovery necessary FLAG is set in the state management table, and in this case, since it is not set, the process is terminated.

Next, FIGS. 5, 6, and 8 show the failure recovery processing operation when an abnormality occurs in two stations, and in particular, when the station that has notified that there is an abnormal station is also malfunctioning. 8
The explanation will be based on.

Note that FIG. 8 shows the processing flow of the entire system in this case.

A message is received from upstream (step 601), and the result of analyzing this message (step 60
2), the station that detected the MACC abnormal station (Figure 8 (1a)) (
In FIG. 8(1b)), it is determined whether loop control is in progress (step 603).

[0041] If loop control is in progress, no further processing is performed and the normal reception state is returned.

On the other hand, when loop control is not in progress, PX
A command to stop monitoring is output to the MAC (step 60).
4) Send a message indicating that there is an abnormal station to the control station. (Step 605, FIG. 8(2)) and sets a monitoring stop command timer (Step 606, FIG. 8(3))
).

Message received (step 501)
Control station 1 continues processing for the abnormal station (step 5).
02), the message is analyzed (step 503), and then it is determined whether loop control is in progress (step 504).

[0044] Here, if loop control is in progress, no further processing is performed and the normal reception state is returned.

On the other hand, if loop control is not in progress, refer to the status management table, specifically refer to the serial number, and check the station one upstream from the station that sent the MACC abnormal station message ( Step 505).

Subsequently, the disconnection processing FLAG of the state management table corresponding to the station one upstream is set (step 506).

Continuing further, a port disconnection request is notified to the station one upstream (step 507, FIG. 7(
4)).

[0048] The station that received the port disconnection request is PXM.
Processing is performed to disconnect the own port from the AC (step 607).

[0049] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609).
Figure 8 (5)).

The control station that receives the port disconnection response sets a monitoring timer (step 510).

On the other hand, the station whose monitoring stop command timer has timed out requests the PX MAC to start monitoring (step 610).

[0052] In this case, since an abnormality has also occurred in the station that started monitoring, the station that started monitoring while the monitoring timer times out sends a message indicating that there is an abnormal station to the control station again. (Step 605, Figure 8
(7)) and set the monitoring stop command timer (
Step 606, FIG. 8(8)).

The control station, which has received the message indicating that there is an abnormal station again, refers to the status management table, specifically refers to the serial number, and confirms the station that transmitted the message that there is an abnormal MACC station (step 512).

[0054] Subsequently, the disconnection processing FLAG of the state management table corresponding to this station is set (step 51).
3).

[0055] Further, a port disconnection request is notified to this station (step 514, FIG. 8 (9)), and the recovery necessary FLAG of the state management table corresponding to the station one upstream of this station is set. (step 515)
.

[0056] The station that received the port disconnection request is PXM.
Processing is performed to disconnect the own port from the AC (step 607).

[0057] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609).
Figure 8 (10)).

The control station that receives the port disconnection response sets a monitoring timer (step 510).

[0059] Since the ports of the two stations where the abnormality has occurred have been disconnected, the station that started monitoring will not send a message indicating that there is an abnormal MACC station until the monitoring timer times out.

[0060] Therefore, after the control station times out,
Search whether the recovery required FLAG is set in the status management table. In this case, since the recovery required FLAG corresponding to the upstream MACC abnormal station is set, port recovery is performed on the upstream station where the abnormality occurred. A request message is notified (step 516, FIG. 8 (13)), and the recovery necessary FLAG of the state management table corresponding to the downstream station among the stations where the abnormality has occurred is set.

[0061] The station that received the port recovery request message:
Processing is performed for the PX MAC to recover its own port (step 611).

[0062] Then, it waits for a response from the PX MAC (step 612), and when a response is received, it notifies the control station of the port disconnection response (step 613).
Figure 8 (14).

[0063] Here, since an abnormality has occurred in the recovered station, a normal station downstream of the two stations where the abnormality has occurred notifies the control station of a MACC abnormal station presence message (step 605, Figure 8 (15)). In addition, the downstream normal station outputs a monitoring stop command to the PX MAC (
At the same time as step 604), a monitoring stop command timer is set (step 606, FIG. 8(16)).

The control station then refers to the status management table and confirms the upstream MACC abnormal station (step 50).
5).

Subsequently, the disconnection processing FLAG of the state management table corresponding to the upstream MACC abnormal station is set (step 506).

[0066] Further, a port disconnection request is notified to this upstream MACC abnormal station (step 507).
, Figure 8(17)).

[0067] The station that received the port disconnection request is PXM.
Processing is performed to disconnect the own port from the AC (step 607).

[0068] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609,
Figure 8 (18)).

On the other hand, the station whose monitoring stop command timer has timed out requests the PX MAC to start monitoring (step 610, FIG. 19).

In this case, since the ports of the two MACC abnormal stations are separated, the station that started monitoring will not send a message indicating that there is an abnormal MACC station until the monitoring timer times out.

[0071] Therefore, after the control station times out, the control station searches the state management table to see if the recovery necessary FLAG is set, and in this case, since the recovery necessary FLAG corresponding to the downstream MACC abnormal station is set, an abnormal A port recovery request message is sent to downstream stations among the stations that have been accessed (step 516, (21) in FIG. 8).

[0072] The station that received the port recovery request message:
Processing is performed for the PX MAC to recover its own port (step 611).

[0073] Then, it waits for a response from the PX MAC (step 612), and when a response is received, it notifies the control station of the port disconnection response (step 613).
Figure 8 (22)).

[0074] Here, since an error has occurred in the recovered station, the station where the error has occurred notifies the control station of a MACC error station presence message (step 604,
Figure 8 (23)). Additionally, this station outputs a monitoring stop command to the PX MAC (step 604) and sets a monitoring stop command timer (step 606).
Figure 8 (24)).

[0075] Upon receiving the message indicating that there is an abnormal MACC station, the control station refers to the status management table and checks the downstream MACC
The abnormal station is confirmed (step 505).

Subsequently, the disconnection processing FLAG of the state management table corresponding to the downstream MACC abnormal station is set (step 506).

[0077] Further, a port disconnection request is notified to this downstream MACC abnormal station (step 507).
, Figure 8(25)).

[0078] The station that received the port disconnection request is PXM.
Processing is performed to disconnect the own port from the AC (step 607).

[0079] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609,
Figure 8 (26)).

The control station receiving the port disconnection response sets a monitoring timer (step 510).

[0081] The station that received the command to start monitoring is the PX MA
A monitoring start request command is issued to C (step 609).

On the other hand, the station whose monitoring stop command timer has timed out requests the PX MAC to start monitoring (step 610, FIG. 7 (27)).

In this case, since the ports of the two MACC abnormal stations are separated, the station that started monitoring will not send a message indicating that there is an abnormal MACC station until the monitoring timer times out.

[0084] Therefore, after timeout, the control station
A search is made to see if the recovery necessary FLAG is set in the state management table, and in this case, since it is not set, the process is terminated.

Next, FIGS. 5 and 6 show the failure recovery processing operation when an abnormality occurs in two stations, especially when the two stations upstream of the station that has notified that there is an abnormal station are abnormal. This will be explained based on FIG.

Note that FIG. 9 shows the processing flow of the entire system in this case.

A message is received from upstream (step 601), and the result of analyzing this message (step 60
2) The station (FIG. 9(1c)) that has detected the MACC abnormal station (FIGS. 9(1a), (1b)) determines whether it is under loop control (step 603).

[0088] If loop control is in progress, no further processing is performed and the normal reception state is returned.

On the other hand, when loop control is not in progress, PX
A command to stop monitoring is output to the MAC (step 60).
4) Sends a message indicating that there is an abnormal station to the control station (step 605, FIG. 9 (2)), and sets a monitoring stop command timer (step 606, FIG. 9 (3))
.

Message received (step 501)
If the processing for the abnormal station is not continuing (step 502), control station 1 analyzes the message (step 502).
After step 503), it is determined whether loop control is in progress (step 504).

[0091] Here, if loop control is in progress, no further processing is performed and the normal reception state is returned.

On the other hand, if loop control is not in progress, refer to the status management table, specifically refer to the serial number, and check the station one upstream from the station that sent the MACC abnormal station message ( Step 505).

[0093] Subsequently, the disconnection processing FLAG of the state management table corresponding to the station one upstream is set (step 506).

[0094] Further, a port disconnection request is notified to this station one upstream (step 507, FIG. 9(
4)).

[0095] The station that received the port disconnection request is PXM
Processing is performed to disconnect the own port from the AC (step 607).

[0096] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609,
Figure 9 (5)).

The control station that receives the port disconnection response sets a monitoring timer (step 510).

On the other hand, the station whose monitoring stop command timer has timed out requests the PX MAC to start monitoring (step 610, FIG. 9(6)).

[0099] In this case, since an abnormality has also occurred in two stations upstream from the station where monitoring has started, the MACC abnormal station is detected from the normal station downstream of the two stations where the abnormality has occurred. message is notified to the control station (step 60
5, Figure 9 (7)). Also, this downstream normal station is PX
Outputs a monitoring stop command to the MAC (step 60)
4), and also sets a monitoring stop command timer (step 606, FIG. 9(8)).

[0100] While the monitoring timer times out, the control station that receives the message indicating that there is an abnormal station again refers to the status management table, specifically refers to the serial number, and
The station that sent the MACC abnormal station presence message is confirmed (step 512).

[0101] Subsequently, the disconnection processing FLAG of the state management table corresponding to this station is set (step 51).
3).

[0102] Further, a port disconnection request is notified to this station (step 514, FIG. 9 (9)), and the recovery necessary FLAG of the state management table corresponding to the station one upstream of this station is set. (step 515)
.

[0103] The station that received the port disconnection request is PXM.
Processing is performed to disconnect the own port from the AC (step 607).

[0104] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609).
Figure 9 (10)).

[0105] Upon receiving the port disconnection response, the control station sets a monitoring timer (step 510). here,
In this case, since the abnormality is still occurring at three stations upstream from the station that started monitoring, the station that started monitoring will send a message indicating that there is an abnormal MACC station again while the monitoring timer times out. (Step 605, FIG. 9(12)). In addition, this downstream normal station is PX M
Outputs a monitoring stop command to the AC (step 604)
At the same time, a monitoring stop command timer is set (step 606, (13) in FIG. 9).

[0106] The control station that receives the message that there is an abnormal station again refers to the status management table, specifically refers to the serial number, and checks the stations three upstream from the station that sent the message that there is an abnormal MACC station. (step 512).

[0107] Subsequently, the disconnection processing FLAG of the state management table corresponding to this station is set (step 51).
3).

[0108] Further, a port disconnection request is notified to this station (step 514, FIG. 9 (14)), and the recovery necessary FLAG of the state management table corresponding to this station is set (step 515).

[0109] The station that received the port disconnection request is PXM
Processing is performed to disconnect the own port from the AC (step 607).

[0110] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609,
Figure 9 (15)).

[0111] Upon receiving the port disconnection response, the control station sets a monitoring timer (step 511).

[0112] Since each port of the two stations where the abnormality has occurred has been disconnected, the station that started monitoring will not send a message indicating that there is an abnormal MACC station until the monitoring timer times out.

[0113] Therefore, after the control station times out,
Search whether the recovery required FLAG is set in the status management table. In this case, since the recovery required FLAG corresponding to the upstream MACC abnormal station is set, port recovery is performed on the upstream station where the abnormality occurred. A request message is notified (step 516, FIG. 9 (18)), and the recovery necessary FLAG of the state management table corresponding to the downstream station among the stations where the abnormality has occurred is set.

[0114] The station that received the port recovery request message:
Processing is performed for the PX MAC to recover its own port (step 611).

[0115] Then, it waits for a response from the PX MAC (step 612), and when a response is received, it notifies the control station of a port recovery response (step 613, FIG. 9 (19)).

The control station that receives the port recovery response sets a monitoring timer.

[0117] In this case, since an error has occurred in the station that recovered its own port, the downstream station sends a message indicating that there is an abnormal MACC station while the monitoring timer times out. (Step 605, FIG. 9(2)
0)). In addition, this downstream normal station outputs a monitoring stop command to the PX MAC (step 604) and sets a monitoring stop command timer (step 606).
, Figure 9(21)).

[0118] The control station that receives the message that there is an abnormal station again refers to the status management table, specifically refers to the serial number, and checks the stations three upstream from the station that sent the message that there is an abnormal MACC station. (step 512).

[0119] Subsequently, the disconnection processing FLAG of the state management table corresponding to this station is set (step 51).
3).

[0120] Further, a port disconnection request is notified to this station (step 514, FIG. 9 (22)).

[0121] The station that received the port disconnection request is PXM.
Processing is performed to disconnect the own port from the AC (step 607).

[0122] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609).
Figure 9 (23)).

[0123] Upon receiving the port disconnection response, the control station sets a monitoring timer (step 510).

[0124] In this case, since the ports of the two MACC abnormal stations are separated, the station that started monitoring will not send a message indicating that there is an abnormal MACC station until the monitoring timer times out.

[0125] Therefore, after the control station times out,
Search whether the recovery required FLAG is set in the status management table. In this case, since the recovery required FLAG corresponding to the downstream MACC abnormal station is set, port recovery is performed on the downstream station where the abnormality occurred. A request message is notified (step 516, (25) in FIG. 9).

[0126] The station that received the port recovery request message:
Processing is performed for the PX MAC to recover its own port (step 611).

[0127] Then, it waits for a response from the PX MAC (step 612), and when a response is received, it notifies the control station of a port recovery response (step 613, FIG. 9 (26)).

[0128] Here, since an abnormality has occurred in the recovered station, a downstream station notifies the control station of a message stating that there is an abnormal MACC station (step 605, step 28 in Fig. 9).
)). In addition, this downstream normal station outputs a monitoring stop command to the PX MAC (step 604) and sets a monitoring stop command timer (step 606).
Figure 9 (29)).

[0129] The control station that receives the message indicating that there is an abnormal MACC station refers to the status management table and checks the downstream MACC
The abnormal station is confirmed (step 505).

[0130] Subsequently, the disconnection processing FLAG of the state management table corresponding to the downstream MACC abnormal station is set (step 506).

[0131] Further, a port disconnection request is notified to this downstream MACC abnormal station (step 507).
, Figure 9(30)).

[0132] The station that received the port disconnection request is PXM
Processing is performed to disconnect the own port from the AC (step 607).

[0133] Then, it waits for a response from the PX MAC (step 608), and when a response is received, it notifies the control station of the port disconnection response (step 609,
Figure 9 (31)).

[0134] Upon receiving the port disconnection response, the control station sets a monitoring timer (step 510).

[0135] In this case, since the ports of the two MACC abnormal stations are separated, the message indicating that there is an abnormal MACC station will not be sent until the monitoring timer times out.

[0136] Therefore, after the control station times out,
Search whether the recovery required FLAG is set in the status management table. In this case, since the recovery required FLAG of the station one downstream from the two MACC abnormal stations is set, a port recovery request message is sent to this station. Notify (step 516, FIG. 9 (33)).

[0137] The station that received the port recovery request message
Processing is performed for the X MAC to recover its own port (step 611).

[0138] Then, it waits for a response from the PX MAC (step 612), and when a response is received, it notifies the control station of a port recovery response (step 613, FIG. 9 (34)).

[0139] The control station that receives the port disconnection response sets a monitoring timer (step 510). In this case, since the ports of the two MACC abnormal stations have been disconnected, the control station sets the monitoring timer until the monitoring timer times out. , a message indicating that there is an abnormal MACC station will not be sent. Therefore, the control station ends the process.

[0140]

As explained above, according to the present invention, when a station detects the occurrence of a fault, it notifies the control station of the fault occurrence, and also suspends monitoring for a predetermined period of time. When the station receives the notification, it notifies the station where the failure may have occurred to disconnect the port until there is no more notification of the failure.
The station that notified the station to disconnect the port is sequentially notified to recover the port, and checks to see if there is a notification of a failure, and when there is a notification of failure, the station disconnects the port again. Therefore, if the station that was disconnected due to a failure is a normal station, it can be recovered without taking any artificial measures.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the configuration of a loop communication system according to an embodiment of the present invention.

FIG. 2 is a diagram showing the configuration of PX MAC.

FIG. 3 is a diagram showing a format of a frame circulating around a transmission path.

FIG. 4 is a diagram showing a state management table.

FIG. 5 is a diagram showing a processing flow of a control station when a failure occurs.

FIG. 6 is a diagram showing a processing flow of a station when a failure occurs.

FIG. 7 is a diagram showing the processing flow of the entire system when a failure occurs.

FIG. 8 is a diagram showing the processing flow of the entire system when a failure occurs.

FIG. 9 is a diagram showing the processing flow of the entire system when a failure occurs.

[Explanation of symbols]

1...Control station 2...Station 3...Transmission line 4...Terminal 5...PBX 6...RGG 7...PX MAC 8...Station side interface circuit 9... Terminal side interface circuit 10...Control unit 11...MAC controller (MACC) 12...
Memory 13, 14...bus

Claims (1)

[Claims] Claim 1: A plurality of stations including a control station are connected in a loop via a transmission line, a terminal is connected to each station via a port in the form of a bus, and each station is connected by circulating a frame on the transmission line. In a loop communication system, in which terminals connected to the station in a bus form communicate using a packet exchange area of the frame, when the station detects the occurrence of a failure, the station The control station is notified of the occurrence of a fault, and monitoring is stopped for a predetermined period of time. When the control station receives a notification of the fault from the station, it notifies the station where the fault may have occurred. The station notifies the port to be disconnected until the failure is no longer reported, and then notifies the station that notified the port to be disconnected to recover the port one by one, and then receives a notification of the failure. 1. A failure recovery method in a loop communication system, characterized in that when the station is notified of the occurrence of a failure, the station is notified again to disconnect the port.