JPH07319502A - Method and system for restart processing for duplex system - Google Patents

Abstract

PURPOSE:To provide the method and system for shortening rising time in the case of any emergency operation at a system provided with duplexed processors and generation managed backup files. CONSTITUTION:This system is provided with an emergency operating device 20 equipped with a means for storing the order of access to backup files 30-3n, means for receiving the restart processing result after initial program load, order changing means for changing the rank of access to any relevant backup file into the first rank at the time of the next emergency operation generation when restart processing is normally finished, means for switching the system configuration by starting the emergency operation again when the restart processing is abnormally finished, means for receiving the results of hardware diagnosis and latest file writing to the backup file of abnormal end in the restart processing during an on-line operation after the restart processing is normally finished, and means for recovering the order of access to backup files 30-3n into the initial state only when the result shows normal end.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dual processor system, a generation-controlled backup file restart processing method, and a method thereof.

[0002]

2. Description of the Related Art Generally, as a method for improving system availability, a method of duplicating a processor is most widely adopted. Particularly in a system requiring high reliability such as an electronic exchange system, an emergency operation device is installed to automatically start up from any system when an emergency failure occurs. When the emergency operation device detects a failure, the system is initialized, the system configuration is performed in a preset order, the initial program load (hereinafter referred to as IPL) is executed, and the restart process is executed. Also, from the viewpoint of recent remote maintenance, the system backup file is generation-managed and stored in a plurality of devices, enabling automatic system startup.

As a method of restarting these generation-managed files, there is a method of automatically restarting from the file one generation before, as in JP-A-63-18443.
Also, a system for setting and displaying the system configuration order as in Japanese Patent Laid-Open No. 52550/1990, and Japanese Patent Laid-Open No. 1072/1990.
A method for controlling access to the IPL and the multi-CH file, such as No. 54, has been published.

[0004]

In the above-described conventional emergency operation apparatus and restart processing method, since the system configuration order is constant, the backup file of the newest generation is temporarily started up, and the system is restarted. It had the drawback of a long downtime.

An object of the present invention is to provide a restart processing method and system for a duplex system in which the system down time is shortened as much as possible.

[0006]

A method for resuming a duplex system according to the present invention is a method for resuming a duplex system having a duplicate processor system and a generation-managed backup file. If the restart process ends normally after loading the initial program that accompanies the restart process, the first priority is to access the backup file for which the restart process ended normally at the next emergency operation. If the step of changing and the restart process are abnormally terminated, the emergency operation is performed again, and the system configuration is switched. The first priority is to access the backup file that ended normally. And a step of the order change to so that.

Further, the restart processing method of the duplex system of the present invention is the restart processing method of the duplex system having a redundant processor system, an emergency operation device for switching the system configuration, and a generation-managed backup file. The operating device stores the access sequence to the backup file, the means to receive the restart processing result after loading the initial program from the processor, and if the restart processing is completed normally, the backup will be performed at the next emergency operation. The means for changing the order of accessing the file to the first rank, the means for switching to the system configuration again when the restart processing is abnormally terminated, and the restart processing following the abnormal termination are performed. After the normal end, the restart process ends normally when the next emergency operation occurs. Means for changing the order so that accessing the backup file becomes the first order, and hardware diagnosis and backup of the backup file that is abnormally terminated by the restart processing during online operation after the restart processing is normally completed. It has means for receiving the writing result of the latest file from the processor and means for restoring the access order to the backup file to the initial state only when the result is received.

[0008]

[Operation] During an emergency operation, a backup file is selected and restart processing is performed according to a predetermined order.
If it ends abnormally, the backup files of the previous generation or earlier are accessed and restart processing is performed, and once normal restart processing is performed, the next emergency operation,
Since the backup process is executed with the backup file as the first order, the duplication with the shortest rise time is possible.

[0009]

Embodiments of the present invention will now be described with reference to the drawings.

FIG. 1 is a block diagram of an embodiment of a restart processing system of a duplex system according to the present invention. This system includes processors 10 and 11 for dual operation and an emergency operation device 20.
And backup files 30, 31, to 3n stored in order of freshness.

The processors 10 and 11 are always operated in a duplicated state, and when an important failure occurs, the emergency operation device 2
0 detects a failure factor and enters emergency operation (EMA),
Reconfigure and restart the system. Processor 1
When the restart by 0 or 11 itself is impossible, the emergency operation device 20 performs the IPL from the backup file 30,
Thereafter, the restart processing program loads the file, and the restart processing including the system initialization is executed.

FIG. 2 is an example of an instruction sequence for instructing the system configuration order in the emergency operation device 20. The system configuration control counter 21 includes processors 10, 11 and backup files 30, 31, ..., 3n. , 2n at the time of the initial emergency operation.
File loading and restart processing are executed in the order of -1, 2n. That is, the new generation backup file is accessed first, and if the backup file has a hardware error or the file contents are incorrect, the restart process fails, the emergency operation is restarted, and the next backup file is restarted. Try processing.

The emergency operation device 20 receives from the processor 10 or 11 whether the result of the restart processing is normal or abnormal, and if normal, sets the pointer of the system configuration control counter 21 to the normally completed backup file to perform online operation. to go into. If abnormal, the emergency operation is executed again as described above. When the restart process ends normally and some trouble factor occurs in the online operation, the emergency operation device 20
Executes the loading from the backup file set in the previous restart processing and the restart processing as the first order. However, when the processor 10 or 11 notifies the emergency operation device 20 of the result of the hardware diagnosis of the backup file that failed in the previous restart processing during online operation and the result of the file recovery operation during online operation, and the result is normal. In addition, the pointer of the system configuration counter is returned to the initial state, that is, the backup file 30, and the restart processing in the file is executed at the next emergency operation.

[0014]

As described above, the present invention has the effect of shortening the system down time by setting the start-up from the backup file whose contents are guaranteed by the previous restart processing as the first order.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of a restart processing method of a duplex system according to the present invention.

FIG. 2 is a diagram showing an instruction example of a system configuration control counter of the emergency operation device 20 shown in FIG.

[Explanation of symbols]

 10 processor 11 processor 20 emergency operation device 21 system configuration control counter of emergency operation device 30 backup file # 1 31 backup file # 2. ． ． 3n Backup file #n

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] June 1, 1994

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be amended] Claims

[Correction method] Change

[Correction content]

[Claims]

[Procedure Amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0006

[Correction method] Change

[Correction content]

[0006]

A method for resuming a duplex system according to the present invention is a method for resuming a duplex system having a duplicate processor system and a generation-managed backup file. step and, after the initials program load due to the restart process, if the restart process has been completed successfully, Luz step to the first order to access to the backup file you restart process was successful at the next emergency operations occur to If the restart process ends abnormally, the emergency operation is performed again and the system configuration is switched, and after the restart process accompanying the abnormal end ends normally, the restart process ends normally when the next emergency operation occurs. Change the rank so that accessing the file is the first rank And a step that.

[Procedure 3]

[Document name to be amended] Statement

[Correction target item name] 0007

[Correction method] Change

[Correction content]

Further, the restart processing method of the duplex system according to the present invention is the restart processing method of the duplex system having a redundant processor system, an emergency operation device for switching the system configuration, and a generation-managed backup file. The operating device stores the access sequence to the backup file, the means to receive the restart processing result after loading the initial program from the processor, and if the restart processing is completed normally, the backup will be performed at the next emergency operation. hand stage you to access the file in the first rank, if the resuming process is abnormally terminated, it enters the emergency operation again, and means for switching the system configuration, after the restart processing due to the abnormal termination is successful , Backup file for which restart processing has completed normally at the next emergency operation A means to change the order so that access is the first order, and a hardware diagnosis and writing of the latest file for the backup file that is abnormally terminated by the restart process during online operation after the restart process is completed normally. And a means for receiving an embedded result from the processor and a means for restoring the access order to the backup file to the initial state only when the result is received.

Claims (2)

[Claims] 1. A restart processing method for a duplex processor system and a duplex system having a generation-managed backup file, the steps of assigning an access order to the backup files, and restarting after loading an initial program accompanying the restart processing. If the process ends normally, the step of changing the order so that the backup file is accessed first when the next emergency operation occurs, and if the restart process ends abnormally, perform the emergency operation again. Then, the first step is to switch the system configuration and to access the backup file for which the restart process has completed normally when the next emergency operation occurs after the restart process after the system configuration has been switched normally due to the abnormal termination. And changing the ranking so that Resuming method of duplicated system. 2. In a restart processing method of a redundant system having a redundant processor system, an emergency operation device for switching a system configuration, and a generation-managed backup file, the emergency operation device sets the access order to the backup file. A means for storing, a means for receiving the restart processing result after loading the initial program from the processor, and if the restart processing is completed normally, the first priority is to access the backup file at the next emergency operation. As described above, the means for changing the order and the means for changing the system configuration when the restart processing is abnormally terminated again, and the means for switching the system configuration, and the restart processing after the system configuration switching due to the abnormal termination are normally completed, In case of an emergency operation of the To change the order so that the access is the first order, and to perform a hardware diagnosis and write the latest file to the backup file that ended abnormally during the restart process during online operation after the restart process ended normally. A method of restarting a duplex system, comprising: a means for receiving an embedded result from a processor; and a means for restoring the backup file access order to the initial state only when the result is received.