JPH0821945B2 - Failure determination method - Google Patents

Description

TECHNICAL FIELD The present invention relates to a failure determination method for collecting failure information generated from a communication network and identifying a failure location based on the collected failure information.

[Conventional technology]

As a conventional fault determination method of this type, Japanese Patent Application No. 60-2507
45 There is a method described in the specification of “Multi-section fault judgment control method”. The conventional failure determination method will be described below by taking this method as an example.

The communication network is configured by including various communication devices, and what collects and codes alarm information generated from these devices is called an alarm pattern. The alarm pattern is a bit pattern representing individual alarm information of each device as "1" when there is a failure and "0" when there is no failure. That is, if the device arranged at the installation location i is ai and the alarm generated by the device ai is bi, the alarm pattern AP is expressed by the following equation (1).

AP = b _I 2 ^I + b _I-1 2 ^I-1 + ... b _i 2 ⁱ + ... + b ₁ 2 ¹ + b ₀ ... (1) However, the installation location i indicates (I + 1) locations from 0 to I. , Alarm bi is normal bi = 0, failure bi =
Set to 1.

In a communication network that multiplexes and transmits signals according to a prescribed hierarchy, when a failure occurs in a certain device, the effect of the failure propagates to all devices connected to the lower side of this device, and alarm information spreads. . For this reason, the collected alarm information itself does not directly represent the failure location. For example, as shown in FIG. 3, device a1
From until a9 is connected, if the device a3 fails, the alarm b ₃ are generated from the device a3. Device at the same time
a7, a8, since the signal from the a9 device to a3 does not reach the alarm b _7, b _8, b ₉ occurs and spread of alarm information occurs. In this way, an alarm pattern including the spread of alarm information when a failure occurs in each part of the communication network is called a spread pattern.

In this way, the spread pattern when a failure occurs at each location of the communication network is stored in advance in association with the failure location, and the alarm pattern currently occurring is searched from this spread pattern, It is possible to find the location of the failure while the alarm information spreads. That is, the spread pattern XP (i) is assigned to each point i. When the failure occurs, the device ak that satisfies AP = XP (k) (3) can be specified as the failure point in this ripple pattern when the failure occurs. In this way, by comparing the alarm pattern with the pre-stored spillover patterns of all the failure points and finding a matching spillover pattern, the failure point that is the cause of the currently occurring failure can be specified.

For example, in the communication network shown in FIG. 3, the propagation pattern corresponding to each fault location can be expressed as follows.

Further, when the alarm pattern (AP = 111000100) is searched from this ripple pattern, AP = XP (3) ... (5), and it is possible to specify that the device a3 is the failure point.

[Problems to be Solved by the Invention]

The conventional failure determination method described above creates and stores an alarm propagation pattern when it is assumed that each location in the communication network has failed in advance, and successively compares the alarm pattern generated at the time of failure with this propagation pattern. However, if an alarm pattern that does not match any of the ripple patterns created in advance is generated, there is a drawback that the determination cannot be made. Further, as the size of the communication network increases, there is a drawback that a storage area for storing an expected alarm pattern is required multiple times and the time required to successively compare the alarm patterns also greatly increases.

An object of the present invention is to provide a failure determination method that solves these problems.

[Means for solving the problem]

According to the method of the present invention, the failure information generated by each device constituting the communication network is divided into three types of signal input disconnection from the upper side, signal output disconnection to the upper side, and signal input disconnection from the lower side in the hierarchy. A classifying unit for classifying, a first extracting unit for extracting the highest one in the hierarchy from the devices in which a failure has occurred, and an opposite device logically opposite to the extracted device. Second extracting means for extracting; third extracting means for extracting a device physically connected to a higher level than the devices extracted by the first and second extracting means; Through determination means for identifying a failure location according to a predetermined determination rule based on the failure information of the equipment extracted by the third extraction means, and storage means for storing the determination rule in advance ing.

[Action]

According to the method of the present invention, 1. Among the devices in which a failure has occurred, there is a cause of failure in the alarm information generated by the highest-level device on the hierarchy.

2. Communication is done on a one-to-one basis, and alarm information can be obtained from the other devices.

3. The alarm information generated by each device can be classified into three types: signal input disconnection from the upper side, signal output disconnection to the upper side, and signal input disconnection from the lower side.

Paying attention to the above three points, the device that is generating the alarm at the highest level on the hierarchy, the device that logically opposes the device, and the device that is physically connected are extracted, and generated from these devices. The failure determination is performed based on the above-mentioned three types of alarms of signal disconnection.

 The operation of the method of the present invention will be described below.

Each communication device has a counterpart device that is logically connected to itself and communicates on a one-to-one basis. This partner device is called a facing device, and this relationship is called a facing device. The two devices in the facing relationship are located in the same hierarchy and communicate by logical connection. However, it is rare that two devices in an opposing relationship are directly connected to each other, and physically, they are connected to a device one level higher in the hierarchy, and the opposing connection is formed by a multiplexed connection line.

Further, as shown in FIG. 4, the alarms generated from the respective devices constituting these communication networks are alarms Iu (i) indicating that the signal from the upper side is not input, and that the signal to the upper side is not output. Alarm Ou
(I) and alarm Id (i) indicating that a signal from the lower side is not input, these alarms can be classified into three types.
The input / output relationship of a signal can be traced by Iu (i), Ou (i), Id (i).

For example, in FIG. 5, the devices a10 and a20 have a facing relationship, and similarly, the devices a1x and a2x have a facing relationship. The output signal of the device a13 is multiplexed with the output signal of the device a14 by the device a11, further multiplexed with the output signal of the device a12 by the device a10, and sent to the device a20. Device a20 is the opposite
a21 and device a22 are branched to respective output signals, and device a21 is further branched to device a23 and device a24, respectively, and eventually the output signal of device a13 reaches device a23, which is in the opposite relationship with this. To do. In this way, device a13 and device a2
3 is logically connected and communicates. Further, in the communication network of FIG. 5, when a failure occurs between the device a12 and the device a10, an alarm I is issued to the device a10.
d (10) occurs and alarms Iu (2
2), Iu (25), Iu (26) are generated, but no alarm is generated in other devices.

The spread of alarm information described in the prior art is caused by the multiplexing of signals, and the influence of the spread always appears on the lower side of the hierarchy and does not affect the upper side. Therefore, among the devices that generate the alarm, there is a cause of the ripple to the highest device on the hierarchy. Also, in the device that is generating the alarm, the input or output of the signal is lost, and look for where the signal is lost between this device and the device connected in a physical or facing relationship. The fault location can be determined by.

Actually, the failure location is determined by the procedure described below.

(1) Among the devices generating the alarm Iu (i), the highest device B on the hierarchy is obtained. (Here, the alarms Ou (i) and Id (i) are excluded.) (2) The device A having a facing relationship with the device B is obtained.

(3) Obtain the device C physically connected to the upper side of the device A.

(4) The alarms generated from the device A, the device B, and the device C are Ou = Ou (A), Iu = Iu (B), Id = Id (C), and the determination rule shown in Table 1 (described later) is used. Pinpoint the fault location.

(5) Among the devices generating the alarm Ou (i), the device A which is the highest in the hierarchy is obtained. (Here, the alarms Iu (i) and Id (i) are excluded.) (6) The device B having a facing relationship with the device A is obtained.

(7) Obtain the device C physically connected to the upper side of the device A.

(8) The alarms generated from the device A, the device B, and the device C are set to Ou = Ou (A), Iu = Iu (B), Id = Id (C), and the failure point is determined using the determination rule shown in Table 1. Specify.

The determination rule shown in Table 1 describes a rule for determining a fault location from the signal flow in the three basic devices A, B, C shown in FIG. 6, and performs the fault determination. In the procedure, these three basic devices are extracted, and then this determination rule is applied to determine the failure location.

In FIG. 5, an example of determination when a failure occurs between the device a10 and the device a12 will be described below. In this case, the device
Alarm Id (10) from a10, alarm Iu (2 from device a22
2) The alarm Iu (25) is generated from the device a25 and the alarm Iu (26) is generated from the device a26.

(1) The device a22 generating the alarm Iu (22) is set as the device B, which is the highest hierarchy in the hierarchy.

(2) The device a12 that is in a facing relationship with the device a22 is referred to as device A.

(3) The device a10 physically connected to the upper side of the device a12 is referred to as a device C.

(4) Alarms generated from the devices a12, a22, and a10 are Iu = Iu (22) = 1, Ou = Ou (12) = 0, Id = Id (1
0) = 1 and using the determination rule shown in Table 1, A
→ C failure, that is, the failure point can be specified to be between the device a10 and the device a12.

(5) Since there is no device generating the alarm Ou (i), the determination is ended.

The principle of the present invention has been described above. In the above description, the signal transmission direction is one-way as an example in FIG. 5, but even when the signal transmission direction is two-way, this method can be used in exactly the same manner without changing the alarm classification and determination procedure and determination rule. Further, the multiplexing hierarchy in the present description can be used not only for the currently realized hierarchy but also for all the materials for which predetermined hierarchization is performed.

〔Example〕

 Hereinafter, the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram of an embodiment of the present invention.
In the figure, the alarms generated in the communication network 100 are collected by the alarm collecting unit 200 and sent to the alarm classifying means 300, and the signal input interruption alarm Iu from the upper side and the signal output interruption to the upper side in consideration of the hierarchy. The alarm Ou and the signal input interruption alarm Id from the lower side are classified into three types and stored in the alarm table 400.

The operation of the highest-level alarm extraction means 500 is roughly divided into two steps. First, the highest alarm extracting means 500 refers to the alarm table 400, searches for the highest device B among the devices which generate the signal input interruption alarm Iu from the upper side, and obtains the alarm information thereof. Iu buffer
1020 and the alarm information of the device A facing the device B is stored in the Ou buffer 1010. Similarly, the connected device alarm extracting means 600 refers to the alarm table 400 and stores the alarm information of the host device C to which the device A is connected in the Id buffer 1030. Ou buffer 1010, Iu buffer 1020,
When the alarm information is stored in the three buffers of the Id buffer 1030 and the Id buffer 1030, the failure point determination unit 700 refers to the determination rule storage unit 800 to determine the failure point indicated by the three alarms Ou, Iu, and Id, and The judgment result R1 is displayed in the result display section 900
Output to.

Next, the highest alarm extraction means 500 is
By referring to the alarm table 400, a device A, which is the highest among the devices generating the signal output interruption alarm Ou from the lower side, is searched for, the alarm information is stored in the Ou buffer 1010, and the device A is opposed to the device A. Iu alarm information of device B
Store in buffer 1020. Similarly, the connected device alarm extracting means 600 refers to the alarm table 400 and stores the alarm information of the host device C to which the device A is connected in the Id buffer 1030. Same as last time, Ou buffer 1010, Iu buffer 102
When the alarm information is stored in three buffers of 0 and the Id buffer 1030, the failure point determination unit 700 refers to the determination rule storage unit 800 to determine the failure point indicated by the three alarms Ou, Iu, and Id. , The judgment result R2 is displayed on the result display section.
Output to 900. In the result display section 900, the judgment results R1 and R2
The failure points of are displayed together.

Top level alarm extraction means 500, connected device alarm extraction means 600, fault location determination means 700, determination rule storage means 80
The function of the failure determination unit 1100 including 0 is usually realized by a computer program. FIG. 2 shows a flowchart of a program example for realizing this function, and Table 2 shows the format of the alarm table.

Each device constituting the communication network 100 has a device number m peculiar to the device. Corresponding to the device number, the device number T of the device facing the device and the device number J of the connected upper device belong The hierarchy L on the hierarchy and the presence / absence of alarms Ou, Iu, Id are stored in the alarm table (AT) 400 as shown in Table 2. The alarm table 400 actually constitutes a two-dimensional array, and when information is to be taken out from this, for example, when it is desired to know the layer L of the device m, it can be read out as 1 = AT (m, L). By the alarm collection unit 200 and the alarm classification means 300,
After the data is stored in this alarm table, the determination is executed according to the procedure of the flowchart shown in FIG.

In the above description, the highest alarm extraction means
When the signal from 500 is not input from upper side (that is, Iu
The device that is the highest level in the hierarchy is extracted only from the devices that have a failure (when the error occurs) and the signal to the upper side is not output (that is, when Ou occurs). Then, when the signal from the lower side is not input (that is, when Id is generated), Id occurs because the highest device in the hierarchy is not extracted from the faulty devices. This is because in the case, Iu always occurs in the device on the opposite side, and the fault in which Id occurs is included in the fault in which Iu occurs.

〔The invention's effect〕

According to the present invention, unlike the conventional method, an accurate judgment can be performed without depending on a failure pattern assumed in advance, and a failure judgment method that can be executed with a small amount of calculation and a small storage area can be obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG.
FIG. 3 is a flow chart of an embodiment of the present invention, and FIGS. 3 to 6 are connection diagrams for explaining the embodiment of the present invention. 100 …… Communication network, 200 …… Alarm collector, 30
0 …… Alarm classification means, 400 …… Alarm table,
500 ... Top level alarm extraction means, 600 ... Connected device alarm extraction means, 700 ... Fault location determination means, 800 ... Judgment rule storage means, 900 ... Result display section, 1010 ... Ou buffer, 1020 ... Iu buffer, 1030 …… Id buffer, 1100
...... Disability determination section.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical indication H04L 12/26 29/14 9466-5K H04L 11/08

Claims (1)

[Claims] 1. Fault information generated by each of the devices constituting a network for communication between opposing devices is signaled from the upper side in the hierarchy, the signal is disconnected from the upper side, and the signal from the lower side is disconnected. The classification means for classifying into three types of input disconnection, and the highest rank in the hierarchy among the devices in which a signal from the upper side is not input or a signal to the upper side is not output First extraction means for extracting a device, second extraction means for extracting an opposite device that logically opposes the extracted device, and devices extracted by the first and second extraction means A third determination means for extracting a device physically connected to the upper level, and a predetermined determination based on the failure information of the device extracted by the first to third extraction means. According to the rules A failure determination method, comprising: a determination unit that identifies a failure location; and a storage unit that stores the determination rule in advance.