JPH06338896A - Fault processing method for loop transmission system - Google Patents

Abstract

PURPOSE:To attain a stuck fault operation from which a faulty node is excluded by temporarily separating a self node from a transmission line, allowing its own node to perform a self-diagnostic test when an abnormality is detected by a fault processing, resubscribing it to the transmission line when the result is normal, and separating it from the transmission line as it is when the result is abnormal. CONSTITUTION:A node control part 22 being the host processor of a control processor 17 is provided with the transmitting and receiving function of the data frame of its own node destination and a compare check state corresponding to it, so that a processing can be attained even when the fault of a loop transmission control part 10 occurs in addition to the conventional fault processing. The control part 22 transmits the data frame of its own node destination to the processor 17 at a prescribed interval, so that the reception monitor and compare check of the data frame circulating through a transmission line 1 can be performed and when the frame is not matched, the self- diagnostic test can be executed. Thus, it is possible to confirm not only the normality of the transmission line 1 but also the normality of the processor 17 and to facilitate a countermeasure at the time of the occurrence of the fault of the control part 10 and the processor 17.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a workstation,
The present invention relates to a failure processing method for a local area network (local area network) system that is configured by connecting computer devices such as personal computers in a loop, and it is particularly required to minimize the influence on the entire loop transmission system when a failure occurs. The present invention relates to a fault processing method for a loop transmission system.

[0002]

2. Description of the Related Art Conventionally, in a loop type transmission system in which computer devices such as workstations and personal computers are connected in a loop, various measures have been taken to prevent a failure of one node from bringing down the entire loop. It was

In particular, one of the loop transmission systems, 1
FDDI (Fiber Distributed Data Int) of 00 Mbps
erface) LAN (Local Area Netwark) ANSI
In accordance with the FDDI standard (ANSI X3T9 / SMT Rev6.2) stipulated by the (American National Standards Institute), the following failure processing methods are used to reduce failures.

FIG. 3 is a block diagram showing an example of a conventional loop transmission system technique. Three nodes 2 are provided in a loop transmission line 1 composed of a primary transmission line 1a and a secondary transmission line 1b. , 3, 4 are connected.

Here, the downstream node for node 4 is node 2, and the upstream node is node 3.

The configuration of each node will be described on behalf of the node 4.

The node 4 includes primary and secondary system reception circuits 6 and 9 and transmission circuits 7 and 8, and a transmission line interface section 5 for controlling a transmission line interface, and transmission control and FDDI in its own node. It is composed of a loop transmission control unit 10 for controlling the failure processing defined by the standard.

The loop transmission control unit 10 includes a changeover switch 1 for switching between transmission and reception of the transmission line interface unit 5.
1, a reception control unit 12, a transmission control unit 13, a beacon reception monitoring timer 14, a token reception monitoring timer 15, an internal bus 16, and a control processor 17 for controlling the loop transmission control unit 10.

When the ring-shaped loop transmission line 1 is formed, any node issues a control frame called a token. This token orbits the loop transmission line 1 at regular intervals, but when it becomes necessary for each node to transmit some data through the loop transmission line 1, the token is acquired to acquire the transmission right,
A data frame is added to the token and sent to the loop transmission line 1. The receiving node fetches only the data frame and sends the token to the loop transmission line 1.

Since this token circulates around the loop transmission line 1 at regular intervals, it is used to detect a network abnormality.

In the above configuration, each node 2, 3, 4
The token monitor timer 15 monitors whether or not the token circulates on the loop transmission line 1 at regular intervals, but a failure occurs in any node in the network and the token does not arrive within the regular time. In the case, the time-out occurs, and a time-out interrupt is issued to the control processor 17.

The control processor 17, which has received the interrupt, transmits a control frame called a beacon 18 to the transmission controller 13 in order to reconfirm the network failure. The beacon 18 is transmitted to the loop transmission line 1 through the transmission control unit 13, the changeover switch 11 and the transmission circuit 7.

At this time, the control processor 17 activates the beacon reception monitoring timer 14 to monitor whether the beacon 18 can be correctly received via the loop transmission line 1.
At the same time, the control processor 17 sends a control frame called an MLS (master line state) 19 to the node immediately above, for example, the node 3 immediately above in the node 4 to request self-diagnosis to the upstream node. 13, the changeover switch 11, the transmission circuit 8, and the loop transmission line 1.

Therefore, when the monitoring timer 14 detects a time-out in the monitoring by the beacon reception monitoring timer 14, it issues a time-out interrupt to the control processor 17. Control processor 17 that received this interrupt
Issues a self-diagnosis test command to the transmission interface unit 5 and the loop transmission control unit 10 in the own node 4.

As a result, the node 4 leaves the loop transmission line 1, and a self-diagnosis test is performed to confirm the normality of each circuit section in the node 4.

On the other hand, the upstream node 3b that receives the MLS 19 also leaves the loop transmission line 1 and performs a self-diagnosis test to confirm the normality of each circuit section of the own node 3.

A node for which the result of such self-diagnosis test is normal rejoins the loop transmission line 1, but a node for which an abnormality is detected does not rejoin the loop transmission line 1 and remains disconnected. become.

[0018]

However, in the following failure modes, the nodes 2 and 34 cannot detect the failure, and the failed node cannot be separated from the transmission line 1, and the entire system becomes unusable. There is a problem that it becomes inoperable.

That is, when a failure occurs in the failure points 20 and 21 shown by "x" in FIG. 3, the reception control unit 12 cannot receive the token in the failure point 20 first.
Therefore, the token monitoring timer 15 times out and issues a timeout interrupt to the control processor 17. Control processor 17 that received this interrupt
Transmits a beacon 18 to the transmission control unit 13, and the transmission circuit 7
The beacon 18 is transmitted to the loop transmission line 1 through.

On the other hand, the beacon reception monitoring timer 14
Monitors whether the beacon 18 transmitted to the loop transmission line 1 can be correctly received via the loop transmission line 1, but the reception control unit 12 receives the beacon 18 because there is a failure at the failure point 20. Can not. Therefore, the beacon reception monitoring timer 14 issues a timeout interrupt to the control processor 17 at the same time as the timeout.

In response to this interrupt, the control processor 17 normally issues a self-diagnosis test command to the transmission interface section 5 and the loop transmission control section 10, but a failure occurs at the failure point 21 on the interrupt signal line. Therefore, the control processor 17 cannot receive the timeout interrupt. Therefore, the self-diagnosis test command cannot be issued to the transmission interface unit 5 and the loop transmission control unit 10.
Similarly, the MLS 19 cannot be transmitted.

From the above, the node 4 in which the failure occurs in both the failure points 20 and 21 and the node 3 on the upstream side cannot leave the loop transmission line 1 and cannot perform the self-diagnosis test of the own node. , The faulty node cannot be separated from the transmission line 1. As a result, the entire system becomes inoperable.

An object of the present invention is to perform failure processing even if a failure occurs in the token and beacon reception paths, the interrupt path to the control processor, etc., and to reliably disconnect the failed node from the loop transmission path. It is to provide a failure processing method for a loop transmission system.

[0024]

In order to achieve the above object, the present invention provides a data transmission from each node to its own node in a loop transmission system in which a plurality of communication nodes are connected in a loop by a transmission line. Is sent every predetermined time, and it is determined whether or not the data frame that has circulated on the transmission path is normally received by the own node, and the predetermined failure processing is executed in the own node according to the result of the judgment. is there.

[0025]

According to the above means, normally, the failure processing of the transmission line is performed by the failure processing method defined by the standard of the loop transmission system, but the loop transmission controlling the failure processing method defined by the standard. Data frames addressed to the own node are sent out every predetermined time so that failure processing can be performed even when the control unit or the like fails, and it is determined whether or not the data frames that have circulated on the transmission path are normally received by the own node. The predetermined failure processing in the own node is executed according to the determination result.

As a result, not only the normality of the transmission path but also the normality of the control processor can be confirmed, and fault processing can be performed even when the loop transmission control unit and the control processor have a failure.

When an abnormality is detected by the failure processing, the self node is once disconnected from the transmission line and the self diagnosis test of the self node is performed. As a result, if the test result is normal, the transmission line is rejoined, and if an abnormality is detected, it remains disconnected from the transmission line, and the system performs degenerate operation excluding the faulty node. As a result, the damage can be reduced.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A fault processing method in a loop transmission system according to the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of a loop transmission system to which the present invention is applied, and the same portions as those in FIG. 3 are represented by the same symbols. Further, FIG. 2 is a flow chart for explaining the operation.

In FIG. 1, a point different from the conventional one is that a node controller 22 which is a higher-level processor of the control processor 17 is used.
Is added.

The node control unit 22 comprises a supervisory frame transmission / reception control unit 23, a supervisory frame reception timer 24, a supervisory frame check unit 25, and a reception frame reception timer 26.

Here, the timer time of the supervisory frame reception timer 24 is set sufficiently longer than the timer time of the beacon reception supervisory timer 14. The reason is that if the self-diagnosis test by the node control unit 22 occurs frequently, the self-diagnosis test using the beacon 18 will be adversely affected.

Reference numeral 27 is a supervisory frame transmission / reception control unit 2
The monitoring frames a and 28 addressed to the own node transmitted from the transmission control unit 3 to the transmission control unit 13 at predetermined time intervals are the monitoring frames a.
Is the supervisory frame a ′ received by the node 4 after going around the loop transmission line 1.

Although there are nodes other than 2, 3, and 4, they are omitted in FIG.

Next, the failure processing according to the above configuration will be described with reference to the flowchart of FIG.

First, in the normal state, each node 4 performs the failure processing of the transmission line 1 by the failure processing method defined by the standard of the loop transmission system, as described in FIG. On the other hand, the supervisory frame transmission / reception control unit 23 of the node 4 transfers the supervisory frame a27 addressed to the self node to the transmission control unit 13 via the internal bus 16 at every predetermined time so as not to adversely affect the existing failure processing. At the same time
The supervisory frame reception timer 24 is started.

The transmission controller 13 transfers the supervisory frame a27 to the transmission circuit 7 via the changeover switch 11. The transmission circuit 7 transmits the monitoring frame a27 to the loop transmission line 1 (step S10).

The receiving circuit 6 of the node 4 is connected to the loop transmission line 1
The supervisory frame a27 that has passed through one round is received and transferred to the reception control unit 12 via the changeover switch 11.

The reception control unit 12 transfers the received supervisory frame a27 to the supervisory frame transmission / reception control unit 23 as the supervisory frame a'28 via the internal bus 16.

The supervisory frame transmission / reception control unit 23 uses the supervisory frame reception timer 24 to monitor whether or not a time-out occurs (step S11). However, if the supervisory frame reception timer 24 detects a time-out, other than the supervisory frame The reception frame reception timer 26 is activated to monitor whether or not the data frame has been received, and monitors whether or not there is a reception frame. That is, the reason why the supervisory frame does not come back may be that the system is overloaded and the supervisory frame does not come back within the timer time of the supervisory frame reception timer 24. Therefore, some data frame is received. If so, the reception frame reception timer 2 is used to judge that the transmission path 1 is normal.
Start 6. The timer time of the reception frame reception timer 26 is set shorter than the timer time of the supervisory frame reception timer 24.

Therefore, if the reception frame reception timer 26 times out, a self-diagnosis test command is issued to the transmission line interface unit 5 and the loop transmission control unit 10 (steps S12 and S13).

That is, the node control unit 22 issues a self-diagnosis test command to the transmission line interface unit 5 and the loop transmission control unit 10 under the condition that both the supervisory frame reception timer 24 and the reception frame reception timer 26 have timed out. .

Therefore, the node 4 composed of the transmission line interface unit 5 and the loop transmission control unit 10, which has received the self-diagnosis test command, temporarily leaves the loop transmission line 1 to perform the self-diagnosis test, and the test result is normal. If so, it rejoins the loop transmission line 1 and returns to the normal state (steps S14, S15, S16).

However, when the supervisory frame reception timer 24 does not detect a time-out, the supervisory frame transmission / reception control unit 23 monitors whether or not the supervisory frame a'28 has been received, and if the supervisory frame a'28 has not been received. , And returns to step S10 (step S17).

On the contrary, if the supervisory frame a'28 has been received, the supervisory frame check unit 25 determines that the supervisory frame a'28 has been received.
27 and the check frame a'28 are compared (step S18).

In this compare check, if the monitor frame a27 and the monitor frame a'28 match, the process returns to step S10 (step S19).

However, if they do not match, and step S
If the self-diagnosis test of 14 is abnormal, the node control unit 22 issues a general reset to the loop transmission control unit 10 and the transmission line interface 5, and also performs a general reset (step S20).

By this general reset, the node 4 once leaves the loop transmission line 1, initializes the contents of the memory and the register value, and then performs a self-diagnostic test. If the result is normal, the loop transmission line 1 Rejoin and return to normal state.

However, if an abnormality is detected as a result of the self-diagnosis test, the loop transmission line 1 is not rejoined. As a result, the node 4 remains disconnected from the loop transmission line 1, and a transmission line that loops back is formed by the nodes on both sides of the node 4, that is, the nodes 2 and 3, and degenerate operation is performed excluding the node 4 in the fault state. (Steps S21, S22, S23).

As described above, in the present embodiment, in addition to the conventional fault handling, the node control unit 22 which is the host processor of the control processor 17 is provided so that fault handling can be performed even when the loop transmission control unit 10 fails. A data frame transmission / reception function for a node and a compare check function for the data frame are provided, and the node control unit 22 sends a data frame addressed to the own node to the control processor 17 at a predetermined time interval and circulates around the transmission path 1. Also, reception monitoring and compare check of the data frame are performed, and when the data frames do not match, a self-diagnosis test is executed, so that not only the normality of the transmission path 1 but also the normality of the control processor 17 is confirmed. Yes, loop transmission control unit 1
It is possible to perform fault handling even when 0 and the control processor 17 fail.

When a data frame addressed to itself cannot be received, a reception timer 26 for monitoring whether or not a data frame other than the data frame can be received is provided, so that it is unnecessary for a node in a communication overload state. It is possible to avoid a normal check.

When an abnormality is detected by the timeout of the reception timer 26 and the compare check of the data frame, the self node is once disconnected from the transmission line 1 and the self diagnosis test of the self node is performed. As a result, if the test result is normal, the transmission line is rejoined, and if an abnormality is detected, it remains disconnected from the transmission line, and the system is in a degenerate operation excluding the faulty node. As a result, the damage can be reduced.

[0053]

As described above, according to the present invention, in addition to the failure processing function defined by the standard of the loop transmission system, each node sends out a data frame addressed to its own node every predetermined time, It is determined whether or not the data frame that has circulated through the node is normally received by the own node, and the predetermined failure processing in the own node is executed according to the determination result, so that not only the normality of the transmission path but also each Fault processing can be performed even when the loop transmission control unit and the control processor constituting the node have failed, so that the failure can be reduced.

Further, when a data frame addressed to itself cannot be received, a reception timer for monitoring whether a data frame other than the data frame can be received is provided, so that it is unnecessary in a node communication overload state. You can avoid checking for normality.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a flowchart illustrating the operation of the embodiment of the present invention.

FIG. 3 is a block diagram showing a conventional loop transmission system.

[Explanation of symbols]

1 ... Loop transmission line, 2, 3, 4 ... Node, 5 ... Transmission line interface unit, 6, 9 ... Reception circuit, 7, 8 ... Transmission circuit, 10 ... Loop transmission control unit, 11 ... Changeover switch, 1
2 ... Reception control unit, 13 ... Transmission control unit, 14 ... Beacon reception monitoring timer, 15 ... Token reception monitoring timer, 16 ...
Internal bus, 17 ... Control processor, 18 ... Beacon, 1
9 ... MLS, 20, 21 ... Fault location, 23 ... Supervisory frame transmission / reception control unit, 24 ... Supervisory frame reception timer, 25
... supervisory frame check unit, 26 ... reception frame reception timer, 27 ... supervisory frame a.

Front page continuation (72) Inventor Toru Miyagawa 2-6-2 Otemachi, Chiyoda-ku, Tokyo Hitachi Information Network Co., Ltd. (72) Inventor Toyokazu Hashimoto 810 Shimoimaizumi, Ebina, Kanagawa Stock Company Hitachi, Ltd. Office Systems Division (72) Inventor Takahiro Maeba 1 Horiyamashita, Hadano-shi, Kanagawa Hitachi Computer Engineering Co., Ltd. (72) Masahiro Inoue 810 Shimoimaizumi, Ebina-shi, Kanagawa Hitachi Ltd. Office Systems Within the business unit

Claims (4)

[Claims] 1. A loop transmission system in which a plurality of communication nodes are connected in a loop by a transmission line,
A data frame addressed to the own node is sent from each node every predetermined time, and it is judged whether or not the data frame circling the transmission path is normally received by the own node. Based on the judgment result, the predetermined failure in the own node A method of processing a fault in a loop transmission system, characterized by executing processing. 2. The fault processing method for a loop transmission system, wherein the predetermined fault processing is not executed when a data frame other than a data frame addressed to the own node is normally received. 3. When the abnormality of the self-node is detected by the judgment, the self-node diagnostic test for disconnecting the self-node from the transmission line and confirming the normality of the circuit in the self-node is performed as a predetermined failure processing. A loop transmission system fault handling method. 4. When the result of the self-diagnosis function is normal, the self-node is reconnected to the transmission line, and when it is abnormal, the self-node is left disconnected from the transmission line. Loop transmission system fault handling method.