JPH0468450A - Fault data logging system - Google Patents

Abstract

PURPOSE:To specify the cause of generating the fault of a processor without fail by storing a data required for analyzing the fault, holding the data and analyzing the cause of generating the fault at another device. CONSTITUTION:While operating a processor 10, a data showing the operational state is written in a storing means 12 of a substrate 20 and when the fault is generated, the substrate 20 is detached from the processor 10. The data stored in the storing means 12 of the detached substrate 20 is held by a holding means 14 and further, the read request of the stored data to be applied from the outside is accepted by an accepting means 16. According to the accepted read request, the stored data is read from the storing means 12 by an output means 18 and outputted to the outside. Thus, the operational state when generating the fault is read out at the other device by using the detached substrate 20 so as to specify the cause.

Description

[Detailed description of the invention] [Table of contents] Overview Industrial field of application Conventional technology Problems to be solved by the invention Effects of the invention Effects of the invention [Summary Regarding the system for logging failure data of processing equipment; The purpose is to provide a method that makes it possible to reliably identify the cause of the occurrence of a problem. A mounting board of the processing device is provided with a reception means for accepting a request for successive storage data, and an output means for reading out the storage data from the storage means and outputting it to the outside in accordance with the received request for successive storage data.

[Field of Industrial Application] The present invention relates to a system for logging failure data of a processing device.

Processing devices are used for various purposes and are used in different operating environments, so intermittent failures of unknown cause often occur depending on the operating environment.

When such a failure occurs, it is necessary to quickly identify and eliminate the cause of the failure.

[Prior Art] Intermittent failures of unknown cause often occur in financial computer systems.

In that case, the top priority is to continue business operations, so the mounting board of the processing equipment that is thought to be the cause of the failure should be removed.
Replaced with another board.

This reduces system downtime due to failure.

The removed board is then placed in a different device from the system in which the failure occurred.A test is then conducted to reproduce the operating conditions at the time of the failure and identify the cause.

[Problems to be Solved by the Invention] However, since the actual operating environment is different between the processing device in which the fault occurred and the device that performs the test to reproduce the operating state at the time of the fault, it is difficult to reproduce the operating state at the time of the fault. becomes difficult.

Therefore, it is not possible to determine the cause of the failure, and therefore, in many cases, it is impossible to identify the cause of the failure.

The present invention has been made in view of the above-mentioned conventional circumstances, and
The purpose is to provide a method that makes it possible to reliably identify the cause of a failure in a processing device.

"Means for Solving the Problems" In order to achieve the above object, the method according to the present invention is the first method.
The method shown in the figure is adopted.

A substrate 20 is mounted on the processing apparatus 10 shown in FIG.

Then, for the substrate 20, as shown in FIG.
Data indicating the operating state of the tag is written in the storage means 12.
The data stored in the storage means 12 is held in the holding means 14.

Furthermore, when a read request for stored data given from the outside is accepted by the receiving means 16, the stored data is read from the storage means 12 by the output means 18 and outputted to the outside in accordance with the received read request.

[Function] In the present invention, data indicating the operating state of the processing device 10 is written into the storage means 12 of the board 20 while the processing device 10 is in operation. When a failure occurs, the board 20 is removed from the processing device 1110.

The data stored in the storage means 12 of the removed board 20 is then held in the holding means 14.

Therefore, using the removed board 20, the operating state at the time of the failure can be read out using another device, and the cause can be identified.

[Embodiment] Hereinafter, a preferred embodiment of the system according to the present invention will be described based on the drawings.

The board 20 in FIG. 2 is mounted on the processing device 10 as shown in FIG.
Circuit 34. Counter 36. Or Gate 38. Resistance 40.
Resistor 42, memory 44. Battery 46. An OR section 48 is mounted.

Then, the board 20 is connected to the outside through a connector 50.

The memory 44 of the board 20 also includes a logic processing circuit 32, a counter 36 . Or Gate 38. A connector 50 is connected, whereby data indicating various operating states of the processing device 10 (parity error bus no response, bus acquisition over, etc.) is written from the logic processing circuit 32 or the like to the memory 44 and stored therein.

The power necessary to hold the data stored in the memory 44 is supplied to the memory 44 from an external normal power source via the OR section 48.

In FIG. 4, the work procedure when a failure occurs is explained using a flowchart. When a failure occurs in the processing device 10 (step 401), the board 2 that is considered to be the cause of the failure
0 is removed from the processing device 10 as shown in FIG. 3 (step 403).

Bidding for a new board in place of the removed processing device 10
Processing continues (step 417).

Then, the removed substrate 20 is set in the analysis device 52 (or the same device) provided at a location different from the processing device 10, as shown in FIG. 3 (step 405).

A request to read the stored data from the memory 44 of the set board 20 is made from the analysis device 52.
4, the request is accepted.

In accordance with the received read request, the stored data is output from the memory 44 to the analysis device 52 via the connector 50 (step 407).

Further, the analysis device 52 analyzes the cause of the failure based on the output data (step 409), and the analysis result is output from the analysis device 152 (step 41).
1).

Then, based on this analysis result, the analysis personnel performs work to identify the cause of the failure in the processing device 10 (step 413).

If the cause of the failure is identified, the failed portion is removed from the board 20 (step 415).

As described above, in this embodiment, data indicating the operating state of the processing device 10 is stored in the memory 44 of the substrate 20 while the processing device 10 is operating.
It is written to and retained in memory.

Then, in the event of a failure, the processing device 10 transfers the substrate 2 to the processing device 10.
0 is removed, the data stored and held in the memory 44 of the removed board 20 is held by the power supply from the battery 46.

Therefore, the data necessary to reproduce the operating state of the processing device 10 at the time of failure is retained, making it possible to perform failure analysis on another device. It becomes possible to eliminate the cause of the occurrence.

This is particularly effective when reducing system down time is the top priority.

In this embodiment, the bus operating state of the processing device 10 is stored and held in the memory 44. Running address data of the program and specific address data in the main memory of the processing device 10 can also be stored and held.

Furthermore, instead of reading data from a special connector 50, it is also easily possible to read data from the system bus by software or hardware settings.

[Effects of the Invention] As explained above, according to the present invention, the data necessary for failure analysis is stored and retained, so the cause of the failure can be analyzed with another device, and therefore the cause of the failure in the processing device can be determined with certainty. can be specified.

[Brief explanation of drawings]

Fig. 1 is an explanation of the principle of the invention. Fig. 2 is an explanation of the structure of the substrate in the embodiment. Fig. 3 is an explanation of the operation of the embodiment. Fig. 4 is a flowchart explaining the working procedure of the embodiment.
It is. 10 ・ ・ 20 ・ ・ 32 ・ ・ 34 ・ ・ 38 ・ ・ 40 ・ ・ 42 ・ 44 ・ ・ 46 ・ ・ 48 ・ ・ 50 ・ ・ Processing equipment substrate, logic processing circuit, delay circuit, counter, OR gate・Resistance 0 resistance ・Memory ・Battery ・OR section ・Connector

Claims (1)

[Claims] A storage means (12) into which data indicating the operating state of the processing device (10) is written, and a holding means (1) that holds the data stored in the storage means (12).
4), a reception means (16) for accepting a request to read stored data given from the outside, and a storage means (16) for reading stored data according to the received read request.
Output means (18) for reading from 12) and outputting it to the outside
A failure data logging method characterized in that: and are provided on a mounting board (20) of a processing device (10).