JPH0321935B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a data change method in a duplex system, and more particularly to a data change method in a duplex system that simultaneously executes the same data change on an active storage device and a standby storage device.

Generally, electronic exchanges store various information (subscriber data) regarding accommodated subscribers in a main storage device, and refer to the information as needed by a processing unit to proceed with the exchange process. Such subscriber data needs to be changed immediately when the class of accommodated subscribers is changed or when the number of subscribers is increased. On the other hand, in electronic exchanges that require high reliability, the central processing system consisting of the processing device, main storage device, etc. is often duplicated. In this type of duplex system, the subscriber data is stored in each main storage device, and both must be changed whenever necessary.

FIG. 1 is a diagram showing an example of a conventional data change method in this type of duplex system. In Figure 1, the processing device CPU0 and the main storage device
MM0 constitutes a 0-system system SYS0, and the processing device CPU1 and main storage device MM1 constitute a 1-system system SYS1. Now 0 system system SYS
System 0 executes the exchange process as the active system, and system 1 SYS1 stands by as the standby system. In such a case, the gates of the system components in the processing unit CPU0 are made conductive, and the data accumulated in the buffer register in the processing unit CPU0 is transferred to both systems.
Main memory MM0 and SYS0 and SYS1
Stored simultaneously in MM1. (See Figure 1)
Therefore, the various data accumulated in the buffer register during the process by the exchange processing section in the processing unit CPU0 that controls the exchange process are passed through the gate in the conductive state,
It is stored simultaneously in main storage devices MM0 and MM1. As a result, even if the 0 system SYS0 is unable to continue the exchange process due to a failure of the processing unit CPU0, the data stored in the main memory MM1 from the processing unit CPU0 can be used to
The processing device CPU1 of the system system SYS1 can continue to execute the exchange process, and service stoppage of the electronic exchange can be prevented. On the other hand, if it is necessary to change subscriber data, the maintenance personnel use a typewriter device.
Data change commands and change data are input from TYP to the active processing unit CPU0. Since the changed data is accumulated in the buffer register 3 by the data changing section 2, it is simultaneously stored in the main memories MM0 and MM1 via the gates in the conductive state.

As is clear from the above explanation, in the data change method in the conventional redundant system, 0
The subscriber data stored in the main storage device MM0 of the system SYS0 and the subscriber data stored in the main storage device MM1 of the system SYS1 are as follows.
It will be changed at the same time by the change data input from the typewriter device TYP. Therefore, if the active system 0 system SYS0 is unable to continue the exchange process because change data containing an error has been input, the 1 system system SYS1, which received the same subscriber data change at the same time, is also an obstacle,
The exchange process will stop.

That is, in the conventional control procedure, (i) data in both the active system and the standby system are changed simultaneously from the normal operating state (FIG. 1-) by a data change command; (Transition 11 to Figure 1-) (ii) In the state where the data has been changed (Figure 1-),
When a fault occurs in the active system due to an error in the changed data, the processor is switched. (Transition to Figure 1-
12) (iii) Processor switched state (Figure 1-)
In this case, since the data of the new active system remains incorrect, normal continued operation is impossible, and in order to resume operation, it is necessary to load the system file back into memory, and during this time online processing is interrupted. Resulting in.

In this manner, in the conventional system, data in both systems is changed simultaneously, and therefore, if the processor is switched due to a failure in the active system due to the data change, there is a drawback that operation cannot be continued in the new active system.

The purpose of the present invention is to eliminate the drawbacks of the data change method in the conventional duplex system as described above,
The purpose of the present invention is to provide a means for preventing the system from stopping its functions even when erroneous data changes are made.

The purpose of this is to store the same data in the main memory of the active system and the main memory of the standby system, and to simultaneously change the data stored in the main memory of the active system in the standby system. In a redundant data processing system that changes data stored in the main storage device, when changing the data stored in the main storage device based on a specific command, only the data in the active main storage device is changed. After confirming that the active system operates normally with the changed data, the data in the main storage device of the corresponding backup system is changed, and it is confirmed that the active system operates abnormally due to the changed data. In some cases, this is achieved by switching between the active system and the standby system.

An embodiment of the present invention will be described below with reference to FIGS. 2 and 3.

First, FIG. 2 shows the procedure of a data change method in a duplex system according to an embodiment of the present invention.
In Fig. 2, in the case of the present invention, (i) Only the active system data is changed from the normal operating state (Fig. 2-) by a data change command (transition to Fig. 2-13) (ii) Active system data The normality of the changed data is confirmed in the changed state (Fig. 2-).

At this time, if a failure occurs in the active system due to an error in the changed data, the processor is switched (Figure 2-
Transition to 15) Furthermore, if the normality of the changed data is confirmed, the data contents of the active system are transferred to the backup system by the backup system data change command, and the data contents of both systems are made to match. (Transition to Figure 2-14) (iii) Processor switched state (Figure 2-)
Since the new active system data is the data before the change, normal continued operation is possible.

Although the new backup system at this time is incorrect data,
It is possible to transfer and match the new active system data while continuing online processing on the new active system.

In this way, this method has the advantage that even if the processor is switched due to a fault in the active system due to incorrect data modification, operation can be continued in the new active system.

Next, a more detailed explanation will be given with reference to FIG. FIG. 3 is a diagram showing a data change method in a duplex system according to an embodiment of the present invention. In FIG. 3, the same symbols as in FIGS. 1 and 2 indicate the same objects. The 0 system SYS0 is the active system, the 1 system SYS1 is the standby system, and the system component 4 in the processing device CPU0 has gates 5 and 6 in a conductive state. Therefore, the various data that the exchange processing unit 1 accumulates in the buffer register 3 during the processing process are as follows:
It is stored in the main storage devices MM0 and MM1 at the same time, and it is possible to prevent service stoppage of the electronic exchange even when the active system shifts from the 0 system SYS0 to the 1 system SYS1. On the other hand, when a maintenance person inputs a data change command and change data from the typewriter device TYP in order to change subscriber data, the data change unit 20 that operates based on the data change command transfers the changed data to the buffer register 3. and gives an instruction to the system component 4 to make only the gate 5 conductive.
The gate 6 is placed in a blocking state. As a result, the modified data accumulated in buffer register 3 is transferred to gate 5.
The subscriber data is stored only in the main memory device MM0 via the subscriber data. The maintainer executes the exchange process using the 0 system SYS0 that stores the changed subscriber data, and if it is confirmed that it is operating normally, the main memory device MM0 is replaced using a known method.
The same subscriber changes made to main memory MM
1 and both systems SYS0 and SYS
Match the subscriber data of 1. If the 0 system SYS0, which stores changed subscriber data, cannot operate normally, the active system is switched to the 1 system SYS1, and the 1 system SYS1, which stores unchanged subscriber data, Continue the exchange process normally. The main memory device MM0, which stores the subscriber data that has been changed and includes errors, can receive the unchanged subscriber data from the main memory device MM1 using a known method.

Note that the detailed configuration of the 1-system system SYS1 in FIG. 3 is the same as the 0-system system SYS0, so some parts are omitted.

As is clear from the above explanation, according to this embodiment, even when changed data containing an error is input,
Only the 0 system SYS0 is changed, and the 1
The system SYS1 can continue the exchange process normally with the unchanged subscriber data.

Note that FIG. 3 is only one embodiment of the present invention, and for example, subscriber data change processing is performed by a processing device.
The present invention is not limited to being carried out by the CPU0, and may be carried out by the processing device CPU1 which is currently in operation, but the effects of the present invention do not change in such a case. Also processing unit CPU
It is also considered that the exchange processing unit 1, data modification unit 2, and system configuration unit 4 in 0 can be realized by a program, but the effects of the present invention will not change in such a case. Furthermore, the object of the present invention is not limited to subscriber data of electronic exchanges, and its effects will remain the same even if applied to other duplex systems or other data.

As described above, according to the present invention, failures caused by data changes in a redundant system are limited to the active system, and the service can be continued to be transferred to the standby system that is not changed, thereby preventing the redundant system from stopping service. be done.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an example of a data change method procedure in a conventional duplex system, FIG. 2 is a diagram showing a data change method procedure in a duplex system according to an embodiment of the present invention, and FIG. FIG. 2 is a diagram of a data change method explaining a detailed configuration. In the figure, SY0 is the 0-system system, SYS1 is the 1-system system, CPU0 and CPU1 are processing units,
MM0 and MM1 are main storage devices, TYP is a typewriter device, 1 is an exchange processing section, 20 is a data change section, 3 is a buffer register, 4 is a system component section, and 5 and 6 are gates.

Claims (1)

[Scope of Claims] 1. Storing the same data in the main storage device of the active system and the main storage device of the backup system, and responding simultaneously to changes in the data stored in the main storage device of the active system. In a redundant data processing system in which data stored in the backup main storage device is changed, when data stored in the main storage device is changed based on a specific command, the active main storage device After confirming that the active system operates normally with the changed data, the corresponding data in the main storage of the backup system is changed, and the changed data causes the active system to operate normally. A data change method in a duplex system characterized by switching between a working system and a standby system when abnormal operation is confirmed.