JPH1091543A - Fault information recording method and information processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the efficiency of fault analysis by adopting minute hardware internal information of a device where a fault occurs by means of individual interfaces as log information in a central processing unit and a diagnosis controller. SOLUTION: A shift register 102 is connected to a fault interruption register holding means 105 through a fault interruption path 103 and the fault interruption register holding means 105 suppresses the occurrence of fault interruption again during the adoption of fault information. When fault interruption occurs, a fault interruption register 104 generates fault interruption and that the fault occurs in the central processing unit 100 or the diagnosis controller 200 is reported to a log adoption control part 106. The log adoption control part 106 controls the adoption of log information by the central processing unit 100. Log information of the central processing unit 100 is stored in a log information storage part for central processing unit 107 and log information of the diagnosis controller 200 is stored in a log information storage part for diagnosis control 108.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a countermeasure for a fault that occurs in an information processing system, and more particularly to a process for recording fault information performed when a system fails due to a fault.

[0002]

2. Description of the Related Art Since current information processing systems have a close relationship with society, it is necessary to avoid interruption of the system operation due to a failure as much as possible. Quick recovery and investigation of the cause are required.

When a system failure occurs due to a failure, the failure information, which is the contents of the main storage device at the time of the failure, is stored in an external storage device in order to analyze the cause of the failure later. ing.

Conventionally, as this type of fault information collection, for example, as disclosed in Japanese Patent Application Laid-Open No. Hei 5-108419 (hereinafter referred to as Reference 1), the fault information collection means is made redundant in the diagnostic apparatus to obtain the fault information. The collection process was going to be performed reliably.

In Japanese Patent Application Laid-Open No. 63-12036 (hereinafter referred to as Reference 2), the internal memory of the maintenance processor is provided independently of the central processing unit failure log collecting means for collecting a log of the central processing unit when a failure occurs. By providing a maintenance diagnostic device log collecting means for collecting a state, an environment at the time of occurrence of a failure including the inside of the maintenance processor can be reproduced.

[0006]

In Reference 1, since the failure information collecting means is made redundant only in the diagnostic device, if a common unit failure or a diagnostic interface failure occurs in the diagnostic device, the failure information collection fails. There is a problem of doing.

In Reference 2, since the fault information collecting means is not redundant, there is a problem that if a fault in the maintenance diagnostic device or a diagnostic interface fault occurs, fault information of the central processing unit and the maintenance diagnostic device cannot be collected. .

An object of the present invention is to enable collection of fault information even when a common unit fault or a diagnostic interface fault in a diagnostic device occurs.

[0009]

In order to achieve the above object, according to the present invention, a fault which occurs in both a central processing unit and a diagnostic control unit is detected, and when a fault is detected, the central processing unit and the diagnostic control are detected. Cause an interrupt on both of the devices,
The hardware internal information of the failed device is recorded as log information by both the central processing unit and the diagnostic control device.

[0010]

FIG. 1 is a block diagram showing an embodiment of the present invention. The embodiment shown in FIG.
00 and the diagnostic control device 200.
First, each unit of the central processing unit 100 will be described.

The central processing unit 100 includes a central processing unit 10
0, a fault information register 101 for storing detailed fault information, and a shift register 10 for shifting out fault information of the central processing unit 100.
2

The shift register 102 is connected to the fault interrupt path 10
3 is connected to the fault interrupt register 104,
The fault interrupt register holding means 105 suppresses the occurrence of another fault interrupt during the execution of collecting fault information.

When a fault interrupt occurs, the fault interrupt register 104 generates a fault interrupt and reports to the log collection control unit 106 that a fault has occurred in the central processing unit 100 or the diagnostic control unit 200.

The log collection control unit 106 is a central processing unit 10
0, the log information of the central processing unit 100 is stored in the log information storage unit 107 for the central processing unit, and the log information of the diagnostic control unit 200 is stored in the log information storage unit 108 for the diagnostic control unit. Is done.

The failure information register 101 can be read by the log collection control unit 106. Central processing unit 100
, A bit indicating the type of the fault is set on the fault information register 101 and the shift register 102 by hardware. However, the shift register 10
During the shift operation of 2, the failure information is set in the shift register after the shift operation is completed.

Next, each part of the diagnostic control device 200 will be described.

The diagnostic control device 200 includes the diagnostic control device 20
0, a fault information register 201 for storing detailed fault information, and a shift register 20 for shifting out fault information of the diagnostic control device 200.
2

The shift register 202 is connected to the fault interrupt path 20.
3 to the fault interrupt register 204,
The fault interrupt register holding means 205 suppresses the occurrence of another fault interrupt during the execution of collecting fault information.

When a fault interrupt occurs, the fault interrupt register 204 generates a fault interrupt, and reports to the shift-out control unit 206 that a fault has occurred in the central processing unit 100 or the diagnostic control unit 200.

The shift-out control unit 206 controls the collection of log information by the diagnostic control unit 200, and
The log information of 0 is stored in the log information storage unit 207 for the central processing unit, and the log information of the diagnostic control device 200 itself is stored in the log information storage unit 208 for the diagnostic control device.

Log collection control unit 10 of central processing unit 100
6 can read the fault information register 201 using the system bus 300. When a failure occurs in the diagnosis control device 200, a bit indicating a failure type in the failure information register 201 and the shift register 202 is set by hardware. However, during the shift operation of the shift register 202, the shift register 202
Is set to failure information.

The description of each part of the embodiment of FIG. 1 is as above. Next, a log collection process according to the present embodiment will be described. The log collection process is broadly divided into a hardware processing flow and a software processing flow. The hardware processing flow is based on whether the failed device is the central processing unit 100 or not.
It depends on whether the diagnostic control device 200 is used.

First, a hardware processing flow when a failure occurs in the central processing unit 100 will be described with reference to FIG. When a fault occurs in the central processing unit 100, bits indicating the fault type are set in the fault information register 101 and the shift register 102 (a1, a2).

Accordingly, a signal obtained by ORing the parallel outputs of all the bits of the shift register 102 is sent to the fault interrupt registers 104 and 204 via the fault interrupt path 103.
As a result, a bit indicating that a failure has occurred in the central processing unit 100 in the bits on the failure interrupt registers 104 and 204 is set to 1 (a3).

In the process a3, a failure interrupt occurs almost simultaneously in the log collection control unit 106 and the shift-out control unit 206. The failure interruption register 104 is held by the failure interruption register holding means 105, and the failure interruption register 204 Are held by the fault interrupt register holding means 205. As a result, it is possible to prevent a failure interrupt from occurring again in the log collection control unit 106 and the shift-out control unit 206 during the log collection process (a4).

Next, a hardware processing flow when a failure occurs in the diagnosis control device 200 will be described. When a failure occurs in the diagnostic control device 200, the failure information register 20
1 and a bit indicating the fault type are set in the shift register 202 (b1, b2).

Accordingly, a signal obtained by ORing the parallel outputs of all the bits of the shift register 202 is sent to the fault interrupt registers 104 and 204 via the fault interrupt path 203.
As a result, a bit indicating that a failure has occurred in the diagnostic control device 200 in the bits on the failure interrupt registers 104 and 204 is set to 1 (b3).

By the processing of b3, a failure interrupt occurs almost simultaneously in the log collection control unit 106 and the shift-out control unit 206, and the failure interruption register 104 is held by the failure interruption register holding means 105, and the failure interruption register 204 Are held by the fault interrupt register holding means 205. As a result, it is possible to prevent another failure interrupt from occurring in the log collection control unit 106 and the shift-out control unit 206 during the log collection process (b4).

The above is the description of the hardware processing flow. Next, a software processing flow will be described with reference to FIGS.

Failure interruption (a3, b3) when a failure occurs in central processing unit 100 or diagnostic control unit 200
As a result, both the log collection control unit 106 and the shift-out control unit 206 start the log collection control software almost simultaneously (c4, d4).

First, the software processing flow of the log collection control unit 106 will be described. When the log collection control software is started, the log collection control unit 106 reads the failure interrupt register 104 (c5).

If a bit indicating that a failure has occurred in the central processing unit 100 is set, the failure information register 101 of the central processing unit 100 is read out and stored in the log information storage unit 107 for the central processing unit. Yes (c
6).

If a bit indicating that a failure has occurred in the diagnostic control device 200 is set, the system bus 3
The contents of the failure information register 201 are read out using 00, and stored in the diagnostic control device log information storage unit 108 (c7).

After c6 or c7, the log collection control unit 10
6 resets the failure interrupt register 104 to release the hold (c8), and ends the log collection processing performed by the central processing unit 100.

Next, a software processing flow of the shift-out control unit 206 will be described. When the log collection control software is started, the shift-out control unit 206 reads the failure interrupt register 204 (d5).

If a bit indicating that a failure has occurred in the central processing unit 100 is set, the shift register 102 is shifted out to the central processing unit log storage unit 207 as log data using the shift path 109. (D6).

When a bit indicating that a failure has occurred in the diagnostic control device 200 is set, the shift path 20
9, the shift register 202 is shifted out to the diagnostic control apparatus log information storage unit 208 (d7).

After d6 or d7, the shift-out control unit 206 releases the hold by resetting the fault interrupt register 204 (d8), and ends the log collection processing performed by the diagnostic control device 200.

Although the embodiment in which the log information storage means is provided inside the central processing unit 100 and the diagnosis control apparatus 200 has been described here, as another embodiment of the present invention, the log information storage means is provided in the central processing unit. In some embodiments, the processing device 100 and the diagnosis control device 200 are provided outside as independent devices.

[0040]

As described above, according to the present invention,
If a failure occurs in the central processing unit or the diagnostic control unit in the information processing system, both the central processing unit and the diagnostic control unit collect detailed hardware internal information of the failed unit as log information through separate interfaces. I do. For this reason, even if one of the central processing unit and the diagnostic control unit cannot execute the log collection process, the other device can collect the log data,
This has the effect of improving the efficiency of failure analysis.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention.

FIG. 2 is a central processing unit 100 and a diagnostic control device 2 of FIG.
6 is a hardware processing flow performed in 00.

FIG. 3 is a software processing flow performed by the log collection control unit 106 following the hardware processing flow of FIG. 2;

FIG. 4 is a software processing flow performed by a shift-out control unit 206 following the hardware processing flow of FIG. 2;

[Description of Signs] 100 Central processing unit 101, 201 Fault information register 102, 202 Shift register 103, 203 Fault interrupt path 104, 204 Fault interrupt register 105, 205 Fault interrupt register holding means 106 Log collection control unit 107, 207 Central processing unit log information storage unit 108, 208 Diagnosis control unit log information storage unit 109, 209 Shift path 200 Diagnosis control unit 206 Shift-out control unit 300 System bus

Claims (6)

[Claims] 1. Fault information for recording information on a fault occurring in an information processing device comprising both a central processing unit and a diagnostic control device connected to the central processing unit and detecting a fault occurring in the central processing unit. In the recording method, a fault detection step of detecting a fault that occurs independently in the central processing unit and the diagnostic control device; and a fault is detected by the fault detection stage in the central processing device and the diagnostic control device. Not only the device in which the fault has occurred, but also a fault interrupt stage in which an interrupt is generated in the other device; and after the interrupt in the fault interrupt stage, the fault occurs in the central processing unit and the diagnostic control device. A log information recording step of recording the hardware internal information of the device as log information by both the central processing unit and the diagnostic control device. Characteristic failure information recording method. 2. An information processing apparatus comprising both a central processing unit and a diagnostic control unit connected to the central processing unit and detecting a failure occurring in the central processing unit, wherein the central processing unit detects a failure of the central processing unit. A processing device failure detection unit, a diagnosis control device failure detection unit that detects a failure of the diagnosis control device, and a failure detected by any of the central processing unit failure detection unit and the diagnosis control device failure detection unit. Fault interrupt means for generating an interrupt in both the central processing unit and the diagnostic control device; and means activated by the fault interrupt unit, wherein the device in which a fault is detected is the central processing unit and the diagnostic control. A log collection control unit for identifying which of the devices is the device and collecting hardware internal information of the failed device as log information; The information processing apparatus characterized in that it comprises a log information storage means for storing the log information collected by. 3. The information processing apparatus according to claim 2, wherein
The information processing apparatus according to claim 1, wherein the log collection control unit includes a first log collection control unit provided in the central processing unit and a second log collection control unit provided in the diagnosis control unit. 4. The information processing apparatus according to claim 2, wherein the log information storage unit has a part provided in the central processing unit and a part provided in the diagnostic control unit. The central processing unit includes a first log information storage unit that stores log information of the central processing unit, and a second log information storage unit that stores log information of the diagnostic control device. The diagnostic control device comprises: a third log information storage unit that stores log information of the central processing unit; and a fourth log information storage that stores log information of the diagnostic control device. And an information processing apparatus. 5. A storage device used by being connected to both of a first and a second information processing device connected to each other, wherein a failure detection means for detecting a failure occurring in the first and the second information processing device. A fault interrupting means for generating an interrupt in both the first and second information processing devices when a fault is detected by the fault detecting means; and a means activated by the fault interrupting means. A log collection control unit for identifying which of the first and second information processing apparatuses is the apparatus in which the failure is detected, and collecting hardware internal information of the failed apparatus as log information; And a log information storage unit for storing log information collected by the log collection control unit. 6. The storage device according to claim 5, wherein said log information storage means comprises a portion connected to said first information processing device and a portion connected to said second information processing device. The portion connected to the first information processing device is the first information processing device.
A first log information storage unit for storing log information of the information processing device, and a second log information storage unit for storing log information of the second information processing device. The part connected to the device is the first
A storage device comprising: a third log information storage unit that stores log information of the information processing device; and a fourth log information storage unit that stores log information of the second information processing device.