JPH07168729A - Restart control system - Google Patents

Abstract

PURPOSE:To automatically start an information processor again when fault is generated and to set an environment for restarting the processor corresponding to the cause of the fault at such a time. CONSTITUTION:When the fault is generated, a fault restart control part 28 collects the required fault information and stores it in a freeze file 7 without erasing fault information in the past. On the other hand, a level for restart is set correspondingly to the cause of a fault at the time. When the restart level is significant, start processing similar to the case of activating a device is performed but when the restart level is non significant, many start processings are omitted and only the required initial processing is executed. At the same time, the restart environment is selected corresponding to the restart level as well. Therefore, according to the kind of the fault, restart processing can be executed as fast as possible.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a restart control system used for control when a restarting process or the like is performed after a failure occurs during operation of an information processing apparatus.

[0002]

2. Description of the Related Art In an information processing apparatus, when software or hardware failure occurs and the operation of the apparatus cannot be continued, the system goes down. In this way, when the operation of the device is interrupted, maintenance personnel collect failure information and restart the operation by pressing down the reset key or the like.

In such a case, specifically, the following processing is performed. FIG. 2 shows a flowchart of conventional processing.
First, when the information processing apparatus is activated for the first time, an activation memory provided in the information processing apparatus, that is, an IPL (Initial Program Loader) ROM controls the initial operation. That is, in step S1, the initial diagnosis by the IPLROM is performed. Furthermore, in step S2,
Initial processing by the IPLROM is executed, and step S3
At, system startup processing is executed.

In such a state, the start-up of the system is completed and the system is in a running state. Here, in step S4, for example, it is assumed that some kind of failure has occurred. In this case, in the conventional system, the maintenance staff waits for the arrival of the maintenance staff, and the maintenance staff performs the operation of collecting the failure information in step S5. That is, the data necessary for searching for the cause of the failure are collected from each unit. Further, in step S6, a restarting operation is performed by the maintenance staff. As a result, the process returns to the state of step S1 again, after which the startup process by the IPLROM is executed.

[0005]

The conventional restart control system as described above has the following problems to be solved. In the conventional system, a maintenance staff gives an instruction for restarting the information processing apparatus after a failure occurs. In this case, maintenance personnel are required to collect fault information,
Failure information may disappear and failure analysis may not be possible due to an operation error in that case. Further, at the time of failure occurrence, the restart cannot be performed unless the maintenance staff is on standby near the information processing apparatus. In addition, when the maintenance staff is at a long distance, the operation of the apparatus must be suspended for a long time until the maintenance staff arrives. In addition, when the occurrence of a failure is not noticed or when the operation for restarting is troublesome, the operation interruption time of the device becomes long.

When the information processing apparatus is highly utilized, it is desirable to shorten the time from system down due to failure occurrence to restart. However, the procedure shown in FIG. 2 requires the same startup time as when starting the information processing apparatus, and it is not easy to shorten the time for restarting. Further, depending on the type of failure factor, it may be preferable to change the operating environment of the apparatus rather than restarting the operation as it is. However, it is not easy to reconstruct such an operating environment in an information processing apparatus in which a program for start-up control is stored in the IPLROM and the procedure is fixed.

The present invention has been made by paying attention to the above points. When a failure occurs, the information processing apparatus is automatically started up, and the environment for restarting according to the failure factor can be easily set. It is an object to provide a restart control system that can be performed.

[0008]

The restart control system of the present invention automatically executes a startup process when a failure occurs,
A failure restart control section that performs control to restart the operation, a failure post-processing control section that collects failure information each time a failure occurs, and that accumulates failure information for multiple generations, and And a system parameter memory storing information including a level for restart set by the fault restart controller, referring to the system parameter memory, from among a plurality of fault restart procedures. It is characterized by selecting an arbitrary procedure.

[0009]

In this system, when a failure occurs, the failure start control unit collects necessary failure information and stores the past failure information in the freeze file without erasing it. On the other hand, the level for restarting is set according to the failure factor at that time. If the restart level is deep, the startup process similar to that at the time of starting the apparatus is performed, and if the restart level is shallow, many startup processes are omitted and only necessary initial processes are executed.
At this time, the startup environment is also selected according to the restart level. As a result, the restart process can be performed as quickly as possible according to the type of failure.

[0010]

The present invention will be described in detail below with reference to the embodiments shown in the drawings. FIG. 1 is a block diagram showing a system embodiment of the present invention. This system includes a bus line 1, a user operation panel (UOP) 2 and a processor (CP).
U) 3, IPLROM 4, system parameter memory (SPM) 5, RAM 6, and freeze file 7 are connected. The user operation panel 2 is a panel for a maintenance person to operate for maintenance and operation of the apparatus, and is a conversational operation panel including a keyboard, a display and the like.

The processor 3 is a part for executing control processing of the entire restart control system. IPLROM4
Is a memory that stores a program for initial operation when the apparatus is started up. In this embodiment, an IPLROM loading file 21 and a system initial processing file control unit 22 are stored.

FIG. 3 is a diagram for explaining the structure of the IPLROM loading file. This IPLROM loading file is composed of a program which is first executed when the power of the apparatus is turned on. This is, for example, as shown in the figure, common loader 11, boot loader 12, SG
It is composed of a file 13, a driver file 14, and a system file 15. These files are IPLR
In accordance with the OM program, for example, a hard disk (not shown) or the like is sequentially loaded into the RAM 6 shown in FIG. 1 and executed.

In the present invention, when the power is turned on, the process basically executed is the deepest process at the restart level, and these files have file names as shown in the figure. And an identifier indicating that the restart level is "A".

FIG. 4 shows the structure of the system initial processing file. These files are program files that are subsequently executed after the start-up process is executed by the IPLROM loading file. These are also stored in a hard disk device or the like (not shown). And I
It is loaded by the system initial processing file controller 22 of the PLROM. This includes the initial process definition and the startup environment definition. The initial processing definition includes the definitions of hardware diagnostic information and check information. The hardware diagnosis information specifies whether or not to carry out an initial diagnosis of the hardware to be mounted, the range of the diagnosis when it is carried out, and the like.

Further, the check information includes designation of whether a dose check of the mounted memory, a check of the memory, a check of the external cache, a control such as incorporation and disconnection, a check of the BBM or the like is necessary. The startup environment definition includes the environment definition and the program definition. The environment definition is composed of information for defining the operating environment of the program. Further, the program definition defines the application program to be loaded.

In order to perform setting of these information and definitions and constant processing when the system is started up, as shown in the figure, each of these files has an identifier whose restart level is "B" together with a file name. use. That is, depending on the cause of the failure, it may be possible to restart from the processing such as the setting of the system initial processing file without starting again from the re-execution of the IPLROM loading file. In order to execute the processing according to such a difference in restart level, the identifiers of the respective file names are made different in advance. Note that the system initial processing file can be rewritten as necessary, as described later, in order to prevent the occurrence of a failure again.

Returning to FIG. 1 again, in the system parameter memory 5, the fault type 23 and the restart counter 24 are stored.
The restart code 25, the restart level 26, and the restart detailed information 27 are stored. In Figure 6,
The system parameter memory storage content explanatory drawing is shown. As shown in the figure, for example, for the start / restart type, when a failure occurs, the same start-up as when the power is turned on is performed, or according to the restart level newly set in the present invention. Stored is information indicating whether or not restart restart is performed. For example, in the case of start-up, this information is "0", and in the case of restart-startup, this information is "1". In addition,
Start-up is a conventional start-up method that is executed by pressing the reset key by turning on the power of the system or intervention of the maintenance person. In addition, restart startup is startup according to the present invention depending on the cause of failure.

The restart counter is a counter which is initialized at the time of start-up, and "1" is added at the time of restart-startup. After the restart start-up, a failure occurs again, and when the restart start-up is executed thereafter, "1" is added again. That is, the number of times when restart startup is continuously repeated is displayed on the restart counter. In the case of this embodiment, when the value of the restart counter exceeds a certain number of times, the system goes down.

The restart code is information set according to the cause of restart, that is, the cause of failure. The restart level is set based on this information. The restart levels A and B in this embodiment are set according to the combination of the value of the restart counter and the content of the restart code. If this level is deep, startup processing close to the startup startup is executed, and if this level is shallow, startup is completed and restarted by only executing part of the initial settings. The detailed restart information is for maintenance personnel, and even if the restart processing is automatically performed after a failure occurs, the maintenance staff will read the content and recognize the details of the failure and the coping procedure, etc. can do. In addition,
The information after the restart counter is valid only when restarting.

Further, in this embodiment, the contents of the system parameter memory 5 can be rewritten using software control or the user operation panel 2. Also, IP
It may be rewritten by the operation of the LROM 4. The RAM 6 is provided with a fault restart control unit 28 for controlling the restart startup operation of the system of the present invention. The fault restart control unit 28 reads the above-described various information stored in the system parameter memory 5, and selects a restart process with an appropriate depth according to the setting content of the restart level 26. This is the part to be executed. Note that this is specifically composed of a program that executes an operation for restarting, and the actual operation is executed by the processor 3.

The post-fault processing control section 29 is a section for collecting fault information from each section of the information processing apparatus and storing it in the freeze file 7 when a fault occurs. This is also composed of a program for information storage processing. It should be noted that in the present invention, the freeze file 7 has a structure in which the failure information collected at the time of failure that has been repeated in the past is sequentially stored without being erased over a plurality of generations.
This is for the purpose of deciding the procedure for diagnosing and restoring the device based on the past failure information and the like, and when the data amount becomes enormous, the data compression process or the like is executed. The upper limit of the storage amount is selected according to the allowable amount of hardware.

The operation of the system of the present invention will be described below with reference to a specific flow chart. FIG. 6 is a flowchart of the startup operation of the information processing apparatus according to the system of the present invention. First, steps S1 and S2 show a start-up operation when the power of the apparatus is turned on. In step S1, it is determined whether the system will be installed. When it is installed, necessary operation start information and the like are set in step S2. IPL after installation is unnecessary or start information is set
The start-up operation by the ROM is executed. It should be noted that steps S1 and S2 are executed by turning on the power or the like by the intervention of the maintenance personnel.

The processes after step S3 are executed at the restart start-up. First, in step S3,
Initial diagnosis is performed by the IPLROM. Next, in step S4, initial processing by the IPLROM is executed. After that, in step S5, the restart information is initialized. The restart information indicates various kinds of information stored in the system parameter memory 5. Here, for example, the type 23 is set to restart startup, and the restart counter 24 is initialized. For other information,
Do not initialize because it does not affect the operation of the system.

Next, in step S6, the system is started up, and the system startup is completed. Then, in step S7, the system is in operation and the fault monitoring is started. As mentioned above, type 2
The reason why 3 is set as the restart start-up and the restart counter 24 is reset is that the restart start-up is executed instead of the start-up if an unexpected failure occurs thereafter. . Further, the reason why the restart detailed information 27, the restart level 26, and the restart detailed information 27 are left as they are is that they are always left as a failure history so that the failure content that occurred immediately before can be understood.

If the operating state is normal in step S7, the restart counter is initialized. This is because the restart counter counts the number of retries when the occurrence of a failure is repeated immediately after the system is restarted, and confirms the seriousness of the failure.

FIG. 7 shows a fault monitoring operation flowchart following the process of FIG. If a failure occurs during system operation, first in step S8, it is determined whether or not automatic restart startup is to be executed. If it is not the automatic restart start-up, the restart instruction is executed by the maintenance staff as usual. In this case, the maintenance person manually sets the restart code in step S9, and the system is reset in step S10. After that, the routine goes to Step S3 shown in FIG.

On the other hand, in the case of automatic startup, step S
At 11, restart information is set. That is,
The restart counter 24 of the system parameter memory 5 is updated and, for example, the value that has been 1 up to now is incremented by 2 such as 2. Further, the restart code 25 is set, and the content of the restart detailed information 27 is set. Next, step S
At 12, it is determined if the restart counter has exceeded. If the restart counter is over, as described above, it is judged that normal startup is impossible even if retry is restarted for a certain amount or more, and the restart level, which is the deepest level, is set again. (Step S13). In this way, the process returns to the restart start-up process shown in FIG. 6, that is, step S3 in FIG.

On the other hand, if it is determined in step S12 that the restart counter has not exceeded, the process proceeds to step S14, and branch processing is performed according to the content of the restart code. For example, when the restart code indicates the content of hardware failure, the process proceeds to step S15, and it is determined whether or not the software interrupt is possible. If interrupting the software is not possible, the process proceeds to step S16, the restart level is set to the same level as in step S13, and the process is restarted.

On the other hand, when it is judged that this is a soft fault when judging the branch by the restart code, and when it is judged in step S15 that the software can be interrupted, the process proceeds to step S17. Set the restart level according to the content of the restart code. Furthermore, in step S18, the restart counter level and the restart level are compared and checked. Then, for example, when the restart counter indicates a large numerical value, the level is set to a level deeper than the restart level set by the restart code, if necessary. As a result, the restart start-up process is performed.

FIG. 8 shows a restart start-up operation flowchart after a failure according to the present invention. When the process shown in FIG. 7 is completed, the restart start-up process as shown in this figure is executed. First, in step S19, failure information when a failure occurs is collected by the failure post-processing control unit 29 shown in FIG. 1 and stored in the freeze file 7. On the other hand, when the processing cannot be continued, such a freeze acquisition processing may be executed after the system reset. After the system reset in step S21,
Freeze can be acquired if the memory is not initialized. This freeze acquisition is generation-managed and stored in the freeze file 7.

Next, in step S20, it is determined whether or not the restart level has dropped to the minimum value. That is, when the restart level is lowered to the minimum value due to a serious failure, the system is not restarted here, and the operation of the information processing apparatus is completely stopped. On the other hand, when the restart level is not the lowest value in step S20, the process proceeds to step S21, and the system reset is automatically executed.

In step S22, initialization diagnosis by IPLROM is performed, and in step S23,
Initial processing by the IPLROM is executed. The process in this case is selectively executed according to the restart level. The same applies to the system startup processing in step S24.

That is, when the restart level is deep, the IPL
All the ROM loading files are executed, and if shallow, only the necessary files among these are selectively executed. Also, there may be cases where this loading file is not executed at all. Further, when the system is started up in step S24, the initial processing file as shown in FIG. 4 is referred to, and only the necessary level of processing is selected.
Therefore, for example, in the case of a fault of the loaded application program itself, only the loading of the program is performed during the startup processing, but in the case of a memory fault of a part of the main storage device, the faulty part is excluded. In addition to such processing, deep startup processing using a loading file, an initial processing file, etc. is executed.

When the system startup in step S24 is completed, the system operation is restarted. Then, in step S25, the failure information is displayed to the maintenance personnel,
If necessary, a buzzer etc. will be sounded to notify maintenance personnel that the restart has been performed automatically.

Since the setting of the restart level at the time of restarting can be performed from the user operation panel as well, when the environment of the information processing apparatus is selected, switched, expanded, transferred, or failed in the transfer. It is also possible to easily carry out the returning work and the like. Further, it is preferable that the maintenance personnel can output the failure information stored in the freeze file 7 whenever necessary while the information processing apparatus is operating.

The present invention is not limited to the above embodiments. Although the system parameter memory 5 has a configuration different from that of the RAM 6, the contents of the system parameter memory 5 may be appropriately moved into the RAM 6. Further, the operation procedure such as the restart processing may be appropriately changed as necessary.

[0037]

The restart control system of the present invention described above collects failure information and a failure restart control section that automatically executes a startup process to restart operation when a failure occurs. The failure post-processing control unit that controls the accumulation of failure information for multiple generations, and the level for restarting according to the cause of the failure. The failure restart control unit refers to this and restarts multiple failure restarts. Since an arbitrary procedure is selected from among the procedures for, it is possible to automatically and optimally start up the apparatus automatically without the intervention of maintenance personnel.

In addition, failure information of a plurality of generations is accumulated, so that failure analysis and the like can be accurately performed, and thereby the startup environment can be freely and appropriately changed depending on the cause of the failure. In addition, if the contents of the system parameter memory that stores the information including the level for restart can be freely changed, not only when a failure occurs, but also when the system installation conditions are changed, switched, selected, transferred, etc. Can be easily performed.

[Brief description of drawings]

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

FIG. 2 is a flowchart of conventional processing.

FIG. 3 is an explanatory diagram of a configuration of an IPLROM loading file.

FIG. 4 is a diagram illustrating the configuration of a system initial processing file.

FIG. 5 is an explanatory diagram of contents stored in a system parameter memory.

FIG. 6 is a flowchart of a restart operation according to the present invention.

FIG. 7 is a flowchart of a fault monitoring operation according to the present invention.

FIG. 8 is a flowchart of a restarting operation after a failure according to the present invention.

[Explanation of symbols]

 1 Bus line 2 User operation panel 3 Processor 4 IPLROM 5 System parameter memory 6 RAM 7 Freeze file

Claims (1)

[Claims] 1. A failure restart control unit that automatically executes a startup process when a failure occurs to control the operation to restart, and collects failure information each time a failure occurs and collects a plurality of pieces of failure information. A failure post-processing control unit that controls accumulation of failure information over generations, and a system parameter memory that stores information including a restart level set according to a failure factor are provided. A restart control system, wherein an arbitrary procedure is selected from a plurality of failure restart procedures with reference to the system parameter memory.