JP2586350B2 - Failure detection processing method and apparatus in OSI management - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for detecting a failure of a transmission line or a device in OSI management, and more particularly to a failure detection means and a CP for managing a failure as OSI management.
U, in order to search for a fault occurrence / recovery object and a fault type in a CPU that manages a fault in the OSI management in a fault detection process in the OSI management including a memory for notifying the CPU of a fault detected by the fault detection means. The present invention relates to a method and apparatus for detecting a failure in OSI management, characterized by reducing the size of the table and reducing the time required to search the table.

[0002]

2. Description of the Related Art A description will be given of a failure detection processing method for managing a failure in the conventional OSI management. As shown in FIG. 3, a CPU that manages a fault as OSI management
1, a fault detecting means 2 for detecting various faults of a transmission line or a device, and one of faults detected by the fault detecting means 2 for transmitting the fault detected by the fault detecting means 2 to the CPU 1. And a memory 3 which is written from the failure detecting means 2 and which is read by the CPU 1 for occurrence / recovery of the failure. The memory 3
Has the same number of bits as the number of faults detected by the fault detecting means 2, and the fault detecting means 2 has a write address table 21 having a bit position on the memory 3 for every fault detected. . In addition, the CPU
Reference numeral 1 denotes a plurality of managed objects 14 for managing a fault, a managed object having a fault type and a fault type search table 13 for each bit position of the memory 3, and a readout from the memory 3. Memory read data 15 which is the information obtained.

[0003] A table 131 shown in FIG. 4 is an example of a management object and fault type search table 13. In this example, au3CTPSink, vc3TTPSink, ... tu3C
It is assumed that each of the k types of TPSink has three, ie, k × 3, managed objects, and two, ie, k × 2, fault types per managed object. The number of faults to be detected is represented by the product of the number of managed objects and the number of fault types per managed object, and is k × 3 × 2. In the embodiment of the table 131, the table 131 is allocated to each bit position of k × 3 × 2 bits in the memory 3 by a combination of a management target object and a failure type.

Next, the operation will be described. When the failure detecting means 2 detects the occurrence / recovery of the failure, the bit position on the memory 3 is obtained from the write address table 21 and the obtained bit position is set to "1" when a failure occurs and "0" when the failure is recovered. Write to 3.

The CPU 1 periodically reads out all of the memory 3 and compares it with the previously read memory read data 15 to determine the bit position at which the value has changed. If there is a bit position in which the value has changed, the managed object 14 in which the fault has occurred / recovered and the type of the fault are obtained from the bit position in which the value has changed using the managed object and fault type search table 13. The failure type is notified to the managed object 14 in which the failure has occurred / recovered. Finally, the contents of the memory read data 15 are replaced with the data read from the memory 3 this time.

[0006]

In this conventional failure detection processing method, the OSI-managed object to be notified of the failure after the failure detection means detects the occurrence / recovery and the failure type are converted. The CPU needs to have a table for assigning a managed object and a fault type to each of the faults detected by the fault detecting means.
When the number of faults to be detected is large, there is a problem that the table becomes large and the search time of the table becomes long.

The present invention has been made in order to solve such a problem. By providing a CPU for managing faults in OSI management with two types of tables, one for searching for a managed object and the other for searching for a fault type, the OSI management system has two types of tables. It is an object of the present invention to reduce the size of a table in a CPU that manages a failure in management and to reduce the time required to search the table.

[0008]

In order to solve the above-mentioned problems, a method for detecting a failure in OSI management according to the present invention uses a bit position on a memory obtained by searching a write address table when a failure occurs / recovers. Failure occurs in /
The recovery data is written, the entire memory is periodically read, and the bit position where the value has changed is searched by comparing with the previously read memory read data, and if there is a bit position where the value has changed, the value has changed. A failure information storage block including a bit position is obtained, a managed object in which a failure has occurred / recovered is determined from the obtained failure information storage block by using a management object search table, and failure information storage of a bit position having a changed value is further performed. A position within the block is determined, a failure type of the occurred / recovered failure is determined from the position within the failure information storage block using the failure type search table, and the determined failure type is notified to the managed object. It is characterized in that the contents of the memory read data are replaced with data read from the memory this time.

Further, the apparatus according to the present invention comprises a fault detecting means for detecting a fault, a memory in which data is written by the fault detecting means, and a memory for reading data from the memory.
A failure management means for managing failures; the failure detection means has a write address table having bit positions on the memory to be written for all failures to be detected, and the memory is managed by the CPU; The fault information storage block has the same number of fault information storage blocks as the object to be managed, and the fault information storage block has the same number of bits as the fault type of the fault detected by the fault detection means.
U is a managed object search table for searching the managed object for each of the fault information storage blocks, and a fault type search table for searching a fault type from all bits in the fault information storage block. It is characterized by having a table, a plurality of managed objects for managing faults by OSI, and memory read data read from the memory.

[0010]

The fault detecting means for various faults in the transmission path and the device, when a fault occurs / recovers, sets a bit position on a memory obtained by searching a write address table in the fault detecting means when a fault occurs. Write "1" and "0" at the time of failure recovery. The CPU periodically reads out all of the memory, searches the memory for the bit position where the value has changed by comparing with the previously read memory read data, and if there is a bit position where the value has changed, if there is a bit position where the value has changed, the bit position where the value has changed A failure information storage block in the memory including the following is obtained, and from the failure information storage block, a management target object in which a failure has occurred / recovered is determined using a management target object search table in the CPU. The location in the failure information storage block is determined, the failure type of the fault that has occurred / recovered is determined from the location in the failure information storage block using the failure type search table in the CPU, and the failure type is determined by the above method. Notify the managed object obtained in step 2. Then, the contents of the memory read data are replaced with the data read from the memory this time.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing one embodiment of the present invention. Referring to FIG. 1, an embodiment of the present invention includes a CPU 1 that manages a fault as OSI management, and a transmission path such as disconnection of an input signal that occurs in an OSI management target such as a transmission device or a communication device such as an exchange. Fault detecting means 2 for detecting a fault of the apparatus such as a fault of the device or an abnormal reading of data from the memory, and a fault detected by the fault detecting means 2 for transmitting the fault detected by the fault detecting means 2 to the CPU 1. Memory 3 in which one bit is assigned to one
Consists of The memory 3 comprises the same number of fault information storage blocks 31 as the number of managed objects 14 managed by the CPU 1. The fault information storage block 31 has the same number of bits as the fault type of the fault detected by the fault detecting means 2. The fault detecting means 2 has a write address table 21 having bit positions on the memory 3 for all faults to be detected. The CPU 1
Is a management object search table 11 having a management object 14 for each failure information storage block 31 of the memory 3 and a failure management table for all the bits of the failure information storage block 31. 1 for the failure type search table 12 having the type
First, it has a plurality of management target objects 14 for managing faults and memory read data 15 which is information read from the memory 3.

Table 11 shown in FIGS. 2A and 2B
Reference numerals 1 and 121 denote an embodiment of a management object search table 11 and a failure type search table 12, respectively. In this embodiment, au3CTPSink, vc3TTPSink,.
It is assumed that each of the three types of 3CTPSink has three, i.e., k * 3 managed objects, and two types of managed objects, i.e., k * 2 fault types, per managed object. Therefore, the number of faults detected is represented by the product of the number of managed objects and the number of fault types per one type of managed object, and is k × 3 × 2.

In the embodiment using the tables 111 and 121,
The memory 3 has k × 3 failure information storage blocks 31 of the same number as the number of objects to be managed, and the table 111 has management objects assigned to the k × 3 failure information storage blocks. Each failure information storage block 31 has the same number of k × 2 bits as the number of types of failure types, and the table 121 is obtained by assigning failure types to the k × 2 bits.

Next, the operation of the embodiment will be described. When the failure detecting means 2 detects the occurrence / recovery of the failure, the bit position on the memory 3 is obtained from the write address table 21 and the obtained bit position is set to "1" when a failure occurs and "0" when the failure is recovered. Write to 3.

The CPU 1 periodically reads out all of the memory 3 and compares it with the previously read memory read data 15 to determine the bit position at which the value has changed. If there is a bit position in which the value has changed, a failure information storage block 31 including the bit position in which the value has changed is determined, and a failure occurrence / recovery is performed from the determined failure information storage block using the management object search table 11. Obtained managed object 14 is obtained. Further, the position of the bit position whose value has changed in the failure information storage block 31 is obtained, and the failure type of the generated / recovered failure is determined from the position in the failure information storage block 31 using the failure type search table 12. Then, the fault type is notified to the managed object 14 in which the fault found / recovered by the above-described method has occurred / recovered. Finally, the contents of the memory read data 15 are replaced with the data read from the memory 3 this time.

Here, the reduction amount of the elements of the table according to the method of the present invention will be calculated. There are k types of managed objects per type m (m ≧ 2), and each type of managed object detects n types of {n ≧ 2} faults, and the fault type is k × 2 in total. Assuming that there are a number, the management object search table 11 has k × m elements and the failure type search table 12 has k × n elements, for a total of k (m + n) elements. On the other hand, in the conventional method, the management object and the failure type search table 13 have k × m × n elements. Therefore, in the system of the present invention, k × m × nk (m + n) ≧ 0 elements can be reduced from the table as compared with the conventional system.

[0017]

As described above, the OS according to the present invention is
In the fault detection processing method in the I management, the CPU that manages the fault in the OSI management has two types of tables, one for searching for a managed object and the other for searching for a fault type. The total amount of tables for searching types can be reduced,
This has the effect of reducing the table search time.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing a failure detection processing method in OSI management according to the present invention.

FIG. 2A shows an example of a management object search table; and FIG. 2B shows an example of a failure type search table.

FIG. 3 is an explanatory diagram showing a failure detection processing method in the conventional OSI management.

FIG. 4 is a conventional example of a management target object and a failure type search table.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 CPU 2 Failure detection means 3 Memory 11 Managed object search table 12 Failure type search table 13 Managed object and failure type search table 14 Managed object 15 Memory read data 21 Write address table 31 Failure information storage block 111 Management One Example of Target Object Search Table 121 One Example of Fault Type Search Table 131 Conventional Example of Managed Object and Fault Type Search Table

Claims (2)

(57) [Claims] 1. A failure occurrence / recovery data is written in a bit position on a memory obtained by searching a write address table at the time of occurrence / recovery of a failure, and all the memories are periodically read out, and the memory read last time is read out. The bit position where the value has changed is searched by comparing with the data, and if there is a bit position where the value has changed, a fault information storage block including the bit position where the value has changed is obtained, and the management target is determined from the obtained fault information storage block. Using the object search table, a managed object in which a failure has occurred / recovered is determined, and furthermore, the position of the bit position whose value has changed in the failure information storage block is determined, and a failure type search is performed from the position in the failure information storage block. The fault type of the fault that has occurred / recovered is determined using the A failure detection method in the OSI management, wherein the failure is notified to the target and the content of the memory read data is replaced with the data read from the memory this time. 2. A fault detecting means for detecting a fault, a memory to which data is written by the fault detecting means, and a CPU which reads data from the memory and manages faults by OSI management, wherein the fault detecting means comprises: A write address table having bit positions on the memory to be written for all detected faults, wherein the memory has the same number of fault information storage blocks as the number of managed objects managed by the CPU; The storage block has the same number of bits as the fault type of the fault detected by the fault detection means, and the CPU is used to search for the management target object for each of the fault information storage blocks. From the table and all bits in the fault information storage block, the fault type A fault type search table for OSI management, comprising: a fault type search table for searching for an error; a plurality of the management target objects managing faults by OSI; and memory read data read from the memory.