JP2872113B2 - Micro diagnostic system for information processing equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a micro-diagnosis system for an information processing apparatus, and more particularly, to a diagnosis test performed by a standby information processing apparatus in a standby system in a redundant system including an active information processing apparatus and a standby information processing apparatus. The present invention relates to a micro diagnostic system of the information processing apparatus at the time.

[0002]

2. Description of the Related Art In a conventional micro-diagnosis method of an information processing apparatus of this type, when a working apparatus that has been operating so far in a duplex system becomes faulty, a standby system apparatus that is on standby from the working apparatus in which the fault has occurred. The operation is switched to the standby mode, so that the standby system device operates in place of the active system device in which a failure has occurred, thereby preventing the operation of the system from being stopped. In such a system, it is necessary to regularly test whether or not the standby device normally operates, in case a failure occurs in the active device. If it is found that a fault has occurred in the standby system, the fault that has occurred in the standby system must be repaired immediately. For this reason, the Central Processing Unit (Central Processing Unit,
A board circuit on which a memory for storing a micro diagnostic program is stored, that is, a CPU mounting board is provided in a standby device, and a micro diagnostic program start instruction is issued from a higher-level control device that manages the above-mentioned redundant system. Based on the above, the micro diagnostic program periodically tests whether each board in the standby system operates normally. If the test shows that a fault has occurred, the faulty board is replaced with a normal board to identify the faulty board and repair the fault in the standby system. Will be In the case of such a duplex system, the boards mounted on the active system unit and the standby system unit can be broadly classified into the above-described CPU mounted board and the CPU non-mounted board which is a board on which the CPU is not mounted. When replacing these boards, it is necessary to store, in the random access memory of the CPU-mounted board, history information of replacement of all failed boards and a normal spare board for maintenance that replaces those boards. Have been done. That is, information indicating which faulty board of the standby system has been replaced with a normal board is stored in the random access memory. The micro-diagnosis program is started when performing a diagnosis based on a micro-diagnosis program start instruction from a higher-level control device that is periodically performed, and is also replaced when a board is replaced after confirmation of the occurrence of a failure. Therefore, when testing whether or not each of the newly mounted boards operates normally, the board is started in response to a micro diagnostic program start instruction from the host controller. The test by the micro diagnostic program is performed based on the history information of replacement of the failed board and the spare board for maintenance stored in the random access memory.

[0003]

In the above-described conventional micro-diagnosis method for an information processing apparatus, it is found that a failure has occurred in a standby system, and after a faulty board is replaced, a maintenance spare newly mounted for replacement is replaced. When performing a diagnosis based on a micro-diagnosis program start instruction from a higher-level control device that manages a redundant system to perform a board diagnostic test, a board to be subjected to a diagnostic test of a replacement board by the micro-diagnosis program must be selected. The information referred to at this time is the board exchange history information stored in the random access memory managed by the CPU.
However, when a plurality of boards including the CPU mounted board become obstacles and the failed boards are replaced, if the non-CPU mounted boards are replaced before the CPU mounted boards, the replacement history of the previously replaced non-CPU mounted boards is changed. Since the information is not stored in the random access memory of the board on which the CPU is mounted, the board on which the CPU is not mounted is dropped from the target list of the diagnostic test after the replacement of the board, and the diagnostic test is not performed. After replacement of the board, the diagnostic test must be performed for all the boards, which increases the time required for the diagnostic test and makes it difficult to search for a fault.

[0004] An object of the present invention is to provide a backup system in the event of a failure.
When replacing a plurality of faulty boards including a CPU mounted board and a non-CPU mounted board, regardless of the mounting order of the CPU mounted board and the CPU non-mounted board, only the micro diagnostic program corresponding to the newly mounted board for replacement. To provide a micro-diagnosis method for an information processing apparatus that executes the method.

[0005]

According to the present invention, there is provided a micro-diagnosis system for an information processing apparatus, comprising: a CPU-mounted board having a CPU and a diagnosis program storage section storing a micro-diagnosis program operated by the CPU; It is configured as a duplex system including an active system unit in operation and a standby system unit in standby, each mounted with a non-CPU mounted board, and issues a diagnostic test of the standby system unit from a higher-level control device. Based on the micro-diagnostic program periodically, based on the diagnosis result is notified to the higher-level control device, and when a failure is detected, a failure-occurring board that needs to be replaced with a maintenance spare board for repairing the failure is specified. Notify the upper-level control device and perform a diagnostic test of the maintenance spare board mounted on the spare system device after the replacement with the maintenance spare board. In the micro-diagnosis method of the information processing device performed by the micro-diagnosis program based on an instruction from a higher-level control device, the CPU-mounted board of the active device detects a mounting state of the CPU-unmounted board of the standby device, The CPU mounting board of the standby device is configured to read the mounting status of the CPU non-mounted board of the standby device detected by the CPU mounting board of the active device from the CPU mounting board of the active device. You.

Further, a micro diagnostic system of an information processing apparatus according to the present invention does not include a CPU mounting substrate having a CPU and a diagnostic program storage section storing a micro diagnostic program operated by the CPU, and does not include the CPU. CPU
Each of the unmounted boards is mounted, and is configured as a duplex system including an active system device in operation and a standby system device in standby, and a diagnostic test of the standby system device is periodically performed based on an instruction from a higher-level control device. Performed by the micro-diagnosis program, and notifies the host controller of the diagnosis result,
When a fault is detected, a faulty board that needs to be replaced with a maintenance spare board for repairing the fault is specified and notified to the higher-level control device, and mounted on the spare system device after replacement with the maintenance spare board. In the micro-diagnosis method of the information processing apparatus for performing the diagnostic test of the maintenance spare board performed by the micro-diagnosis program based on an instruction from the higher-level control device, the CPU-mounted board of the active system apparatus includes:
A spare CPU non-mounted board mounting detecting means for detecting a mounting status of the CPU non-mounted board of the standby apparatus, wherein the CPU mounted board of the standby apparatus is a CPU of the active apparatus.
A standby system for reading the mounting status of the non-CPU mounted substrate of the standby system device detected by the standby CPU non-mounted substrate mounting status detecting unit of the mounting substrate from the standby CPU non-mounted substrate mounting status detecting unit of the active system device It is configured to include a board mounted detection result reading unit without a CPU.

[0007]

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of a micro diagnostic system of an information processing apparatus according to the present invention.

An information processing apparatus according to an embodiment of the present invention shown in FIG. 1 includes an active active device 111 and an active active device 111 when a failure occurs in the active active device 111. And a standby device 110 that operates instead.

The active system device 111 comprises a CPU mounting substrate 91 which is a substrate on which a CPU is mounted, and a non-CPU mounted substrate 101 which is a substrate on which no CPU is mounted. Substrate 90 and C
And a substrate 100 on which no PU is mounted.

Further, the CPU mounting board 90 includes the CPU 1
0, a diagnostic activation monitoring unit 20 for monitoring a micro diagnostic program activation instruction from a higher-level control device (not shown), and a higher-level control for a result of micro-diagnosis performed based on the micro diagnostic program activation instruction from the higher-level control device. A diagnosis result notifying unit 30 for notifying the device and a CP for the standby device 110
A board mounting monitoring unit 40 for monitoring the mounting of the U mounting board 90 itself, a device micro-diagnosis program for testing the operation of the entire standby system device 110 and identifying a faulty board, and a program mounted on the standby system device 110 Program storage unit 50 in which a board type micro diagnostic program corresponding to each board for individually performing a diagnostic test on the CPU mounted board 90 and the CPU non-mounted board 100 is stored.
A self-system board mounting detection result reading unit that reads the mounting status of the non-CPU mounted board 100 of the standby system device 111 detected by the CPU mounted board 91 of the active system device 111 from the CPU mounted board 91 of the active system device 111 (Spare CPU
An unmounted board mounting detection result reading means) 60 and an other-system board mounting detecting unit 70 for detecting the mounting status of the CPU non-mounted board 101 of the active system device 111.

The CPU mounting board 91 has the same configuration as the CPU mounting board 90, and includes a CPU 11, a diagnosis start-up monitoring section 21 for monitoring a micro-diagnosis program starting instruction from a higher-level control device, and a higher-level control device. A diagnostic result notifying unit 31 for notifying a higher-level control device of a micro-diagnosis result resulting from a micro-diagnosis program start instruction from the host, a self-board mounting monitoring unit 41 for monitoring mounting of the CPU mounting board 91 on the active system device 111, A device micro-diagnosis program for testing the operation of the entire active system device 111 and specifying a faulty substrate, and a CPU mounted substrate 91 and a CPU non-mounted substrate 101 mounted on the active system device 111
A diagnostic program storage unit 51 in which a board type micro diagnostic program corresponding to each board for individually performing a diagnostic test is stored, and the CPU of the active system unit 111 detected by the CPU mounting board 90 of the standby system unit 110
The mounting status of the unmounted board 101 is determined by the C
Self-system board mounting detection result reading section 61 read from PU mounting board 90, CPU non-mounted board 100 of standby system device 110
And the other-system board mounting detecting unit (board mounting detecting means without standby CPU) 71 for detecting the mounting of the device.

The non-CPU board 100 of the standby system device 110 is connected to the other-system board mounting detection section 71. When the non-CPU substrate 100 is mounted on the standby system device 110, the other-system board mounting is performed. A voltage signal is output to the detecting unit 71 to notify that the mounting is performed. Also,
The own-system board mounting detection result reading unit 60 of the CPU mounting board 90 of the standby system device 110 is also connected to the other-system board mounting detecting unit 71. It is possible to read the data of the detected mounting status of the CPU-unmounted board 100 of the standby apparatus 110.

Next, the operation will be described with reference to FIGS.

FIG. 2 is a flowchart showing an example of an operation when a diagnostic test of the standby system is performed in response to a diagnostic start instruction periodically performed from the host controller. FIG.
8 is a flowchart illustrating an example of an operation at the time of a diagnostic test performed after a failure is detected in a standby device and a failed board is replaced.

In FIG. 1, the diagnosis start-up monitoring section 20 receives a micro-diagnosis start-up from the host controller (step 201 in FIG. 2) and notifies the CPU 10 of the start-up. The CPU 10 tests the operation of the standby system device 110 from the diagnostic program storage unit, and when the occurrence of a failure is detected, reads out and executes a device micro diagnostic program for specifying a failed board (step 202). The CPU 10 checks the diagnostic result obtained by executing the micro diagnostic program for the device and determines whether there is a diagnostic abnormality (step 203). If there is no abnormality in the diagnosis result, the diagnosis result notification unit 30 is notified that there is no abnormality in the diagnosis result. Diagnosis result notification unit 3
0 notifies the host controller of the diagnosis result without abnormality sent from the CPU 10 as a result of the micro diagnosis based on the micro diagnosis program start instruction from the host controller (step 204). The operation of the device micro-diagnosis periodically performed by the higher-level control device in accordance with the micro-diagnosis program start instruction from the higher-level control device ends if the diagnosis result shows no abnormality. If the diagnosis result is abnormal, CP
U10 creates a status of a diagnosis abnormality of the standby system device 110 and a list of target boards for diagnosis abnormality and sends the list to the diagnosis result notification unit 30, so that the diagnosis result notification unit 30 performs micro diagnosis based on a micro diagnosis program start instruction from the host controller. As a result, the diagnosis result sent from the CPU 10 is notified to the host controller (step 205), and the operation of the micro diagnosis based on the micro diagnosis program start instruction from the host controller is completed.

Next, based on the diagnosis result notified to the higher-level control device, the maintenance staff replaces the faulty board mounted in the standby system with the maintenance spare board. Mounting board 90 of system device 110
If both the CPU and the board 100 without CPU are abnormal as a result of the diagnostic test, each board is replaced with a spare board for maintenance. CP based on the instruction from the host controller
U10 reads the monitoring data of its own board mounting monitoring unit 40 (Step 210 in FIG. 3), and
It is checked whether or not the U-mounted board 90 has been mounted (step 211). If mounting is detected, a board-type micro diagnostic program corresponding to the CPU-mounted board 90 is executed (step 212). Monitoring unit 40
Is reset (step 213). If it is not detected that the CPU mounting board 90 of the standby apparatus 110 has been mounted, the above procedure is skipped and the process proceeds to the next procedure. Next, the CPU 10 notifies the self-system board mounting detection result reading unit 60 of the standby system device 110 and reads the detection data of the other system board mounting detecting unit 71 of the active system device 111 from the self-system board mounting detection result reading unit 60. (Step 21)
4) Check whether the non-CPU mounted board 100 is mounted on the standby system device 110. When the CPU-unmounted board 100 of the standby system device 110 is mounted, it outputs a voltage signal to the active system device 1.
11 is output to the other-system-board-mount detecting section 71, so that the other-system-board-mount detecting section 71 knows that the non-CPU board 100 of the standby apparatus 110 has been mounted (step 215).
The CPU 10 includes a board 1 on which the CPU
When it detects that 00 is mounted, the mounted CPU
The board type micro diagnostic program corresponding to the unmounted board 100 is executed (step 216), and after execution, the self-system board mounting detection result reading unit 60 of the standby system device 110 and the other system board mounting detection unit 71 of the active system device 111 are executed. The data is reset (step 217), and the operation of the micro-diagnosis for performing a diagnostic test on whether or not each board newly mounted for replacement operates normally is completed.

In the above description, only the operation for the diagnostic test of the standby system device 110 has been described.
When 1 becomes a failure and is switched to the standby system device 110, the standby system device 110 that has been in the standby state enters the operating state. Depending on the operation state of the redundant system, even after the failure of the active device 111 shown in FIG.
1 may be in a standby state. In this case, the higher-level control device issues a request to the active system device 111 in the standby state.
As in the case of the standby system device 110, a diagnostic test instruction is periodically issued. As shown in FIG. 1, since the standby system device 110 and the active system device 111 have the same configuration, when the standby system device 110 enters the operating state and the active system device 111 is in the standby state, the host control device When the micro-diagnosis start-up is performed on the active system device 111, the operation of the diagnostic test of the active system device 111 is started.
Is performed in the same manner as in the case of the diagnostic test described above, but the description is omitted.

[0019]

As described above, according to the micro-diagnosis method of the information processing apparatus of the present invention, a failure is detected in a standby spare apparatus while an active apparatus is operating, and the failed board is maintained. When the spare board is replaced with
A means for detecting the mounting status of the board on which the PU is not mounted is provided on the CPU mounting board of the active apparatus, and the detected CP of the standby apparatus is detected.
By providing means for reading the mounting status of the U-unmounted board from the CPU-mounted board of the active system on the CPU-mounted board of the standby system, a plurality of faulty boards including the standby-system CPU-mounted board and the CPU-unmounted board can be replaced. When the CPU
Regardless of the mounting order of the mounting board and the board without CPU,
Only the micro diagnostic program corresponding to the newly mounted board for replacement can be executed,
This has the effect of shortening the time required for the diagnostic test time and facilitating the search for a fault.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a micro diagnostic system of an information processing apparatus according to the present invention.

FIG. 2 is a flowchart showing an example of an operation at the time of starting a diagnosis from a higher-level control device that is periodically performed in a micro-diagnosis method of the information processing apparatus of the present invention.

FIG. 3 is a flowchart showing an example of an operation at the time of a diagnostic test performed after replacing a faulty board in the micro diagnostic method of the information processing apparatus of the present invention.

[Explanation of symbols]

 10, 11 CPU 20, 21 Diagnosis start monitoring unit 30, 31 Diagnosis result notification unit 40, 41 Own board mounting monitoring unit 50, 51 Diagnosis program storage unit 60, 61 Own system board mounting detection result reading unit 70, 71 Other system board Mounting detection unit 90, 91 CPU mounting board 100, 101 CPU non-mounting board 110 Stand-by device 111 Working device

Claims (2)

(57) [Claims] 1. A CPU-mounted board having a CPU and a diagnostic program storage section storing a micro-diagnostic program operated by the CPU;
It is configured as a duplex system including an active system unit in operation and a standby system unit in standby, each mounting a board without a PU mounted thereon, and performs a diagnostic test of the standby system unit periodically based on an instruction from a higher-level control device. In accordance with the micro-diagnostic program, the diagnosis result is notified to the higher-level control device, and when a fault is detected, a faulty substrate that needs to be replaced with a maintenance spare substrate for repairing the fault is specified and the higher-level control device is specified. And performing a diagnostic test of the spare maintenance board mounted on the spare device based on an instruction from the higher-level control device using the micro diagnostic program after the replacement with the spare spare board. In the diagnosis method, the CPU-mounted board of the active system device detects the mounting status of the CPU-unmounted board of the standby system device, and CPU mounting substrate location is, the implementation status of the CPU not installed board of the current system the standby system device detected by the CPU mounting substrate of the device, C of the active system device
A micro-diagnosis method for an information processing device characterized by reading from a PU mounting board. 2. A CPU-mounted board having a CPU and a diagnostic program storage section storing a micro-diagnostic program operated by the CPU;
It is configured as a duplex system including an active system unit in operation and a standby system unit in standby, each mounting a board without a PU mounted thereon, and performs a diagnostic test of the standby system unit periodically based on an instruction from a higher-level control device. In accordance with the micro-diagnostic program, the diagnosis result is notified to the higher-level control device, and when a fault is detected, a faulty substrate that needs to be replaced with a maintenance spare substrate for repairing the fault is specified and the higher-level control device is specified. And performing a diagnostic test of the spare maintenance board mounted on the spare device based on an instruction from the higher-level control device using the micro diagnostic program after the replacement with the spare spare board. In the diagnostic system, the CPU-mounted board of the active device detects a mounting status of the CPU-unmounted board of the standby device. Comprising a non-mounting substrate mounting detecting means, CPU mounting substrate of the standby system device, C of the active system device
The standby state for reading the mounting status of the non-CPU mounted board of the standby apparatus detected by the standby CPU non-mounted board mounting state detecting means of the PU mounting board from the standby CPU non-mounted board mounting state detecting means of the active apparatus. A micro-diagnosis method for an information processing apparatus, comprising: a board mounted detection result reading unit without a system CPU.