JPH04241657A - Method and circuit for outputting fault alarm of peripheral controller - Google Patents

Abstract

PURPOSE:To perform preventive maintenance by temporarily storing the content of an error after the content is analyzed and edited and using the stored content for comparison and collation with the content of another error which occurs thereafter and, when a count value is smaller than a fixed value, predicting the occurrence of a fault. CONSTITUTION:When an error occurs, an error analyzing section 5 generates a status byte by analyzing and editing the content of the error, but does not make storing operations and inputs the new status byte to an error comparing section 7, since a flag 10 provided in an error storing section 6 indicates a fully occupied state of the section 6. Upon receiving the status byte from the analyzing section 5, the comparing section 7 receives a corresponding status byte from the storing section 6 and compares both bytes with each other. When both bytes coincide with each other, a fault information output section 9 reads the count value of a counter section 8 and, when the count value is smaller than a prefixed value, outputs a fault alarm to a central processing unit 1.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fault alarm output method and circuit for a peripheral control device.

0002

[Prior Art] Conventional methods and circuits for outputting fault alarms for peripheral control devices include counting the number of times this fault occurs when a fault occurs in a peripheral control device or a peripheral device, and counting the number of faults that occur in a peripheral control device or peripheral device. Sometimes special reports are made to the central processing unit.

0003

[Problems to be Solved by the Invention] The conventional fault alarm output method and circuit for a peripheral control device described above count the number of faults, and when this counted value exceeds a predetermined counted value, a special warning is issued. Since the report is sent to the central processing unit, it is not possible to know the frequency of occurrence of failures relative to the number of commands from the central processing unit. Although it is occurring,
The problem is that failures are detected late when recovery is done by hardware retry or software retry, or when the number of failures does not immediately exceed a predetermined value in a system that is booted infrequently. There is a point.

An object of the present invention is to develop a system that is superficially normal due to a failure in which the same error is repeated intermittently, but is recovered by hardware retry or software retry, and Peripheral control that can be applied to systems where the number of failures does not quickly exceed a predetermined value because the number of failures is small, and can detect signs of failure before a major failure occurs and take preventive maintenance measures. An object of the present invention is to provide a fault alarm output method and circuit for a device.

[0005]

[Means for Solving the Problems] A failure alarm output method for a peripheral control device according to the present invention is a failure alarm output method for a peripheral control device that is connected to a central processing unit and controls a peripheral device. At the end of the connection control process for the peripheral device received, the content of the error that occurred in the peripheral device and the peripheral control device during the connection control process is analyzed and edited, the error content is temporarily stored, and Analyzes and edits new error contents that occur during connection control processing according to instructions from the central processing unit, compares and verifies the error contents with the temporarily stored error contents, and if the same error is detected, the central processing unit The number of commands received by the processing unit is counted until the same error as the temporarily stored error content is detected again, and if this counted value is within a predetermined value, a failure is predicted and the central processing unit is This configuration outputs a failure alarm.

[0006] The failure alarm output method for a peripheral control device of the present invention analyzes and edits the content of a new error that has occurred during connection control processing according to a command from the central processing unit, and compares the content with the previously temporarily stored error content. When the same error is detected, the number of subsequent commands received by the central processing unit is counted until the same error as the temporarily stored error content is detected again, and this counted value is kept at a predetermined constant value. If the value exceeds the value, it is determined to be normal, and the number of times this normality is detected is counted. If normality is detected a predetermined number of times, the temporarily stored error content is erased, and when a new error is detected, the process is performed again. The information may be temporarily stored and the number of times the command from the central processing unit is received may be counted.

The fault alarm output circuit for a peripheral control device of the present invention is a fault alarm output circuit for a peripheral control device that is connected to a central processing unit and controls a peripheral device, and receives a command from the central processing unit to control the peripheral device. an error analysis unit that analyzes and edits the content of an error occurring in the peripheral control device and the peripheral device during the connection control process at the end of the connection control process; and temporarily stores the error content analyzed by the error analysis unit. The error storage section and the error analysis section analyze and edit error contents newly generated during connection control processing from the next time onward, and compare and output the error contents with the error contents temporarily stored in the error storage section. reads the current count value of the error comparison section to be executed and the counter section that has a function of receiving a coincidence signal output when the error comparison section detects the same error and counting the number of times the command from the central processing unit is received. After that, the counted value of the counter unit is reset to zero, and if the read counted value is within a predetermined constant value, a failure is predicted to occur and a failure alarm is output to the central processing unit, and the read-in count value is returned to zero. If the counted value exceeds a predetermined certain value, it is determined to be normal, and the number of times this normality is detected is counted internally, and when normality is detected a predetermined number of times, an erase signal is output to the error storage section and the The configuration includes a failure alarm output unit that erases temporarily stored error contents.

[0008]

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

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

The central processing unit 1 is connected to a peripheral control device 3 that controls peripheral devices 2 . The failure alarm output circuit 4 in the peripheral control device 3 receives commands from the central processing unit 1, and when the connection control process for the peripheral device 2 is completed, the failure alarm output circuit 4 in the peripheral control device 3 receives the commands from the central processing unit 1, and when the connection control process for the peripheral device 2 is completed, the failure alarm output circuit 4 receives the command from the central processing unit 1. An error analysis unit 5 that analyzes and edits the content of the error that has occurred; an error storage unit 6 that temporarily stores the error content analyzed by the error analysis unit 5; The error comparison unit 7 analyzes and edits the error content that has occurred, and compares and outputs the error content with the error content temporarily stored in the error storage unit 6, and the error comparison unit 7 detects the same error. A counter unit 8 has a function of receiving a coincidence signal output when the central processing unit 1 receives a command, and counting the number of times the command is received from the central processing unit 1. After reading the current count value of the counter unit 8, the count value of the counter unit 8 is is reset to zero, and if the read count value is within a predetermined certain value, it predicts the occurrence of a failure and outputs a failure alarm to the central processing unit 1, and if the read count value exceeds a predetermined certain value. If this happens, it is determined to be normal, and the number of times this normality is detected is counted internally, and if normality is detected a predetermined number of times, an erase signal is output to the error storage section 6 to erase the previously temporarily stored error content. A fault alarm output unit 9 is provided.

Next, the operation will be explained.

[0012] If an error occurs in the peripheral device 2 and the peripheral control device 3 during connection control processing, the error analysis unit 5
This error content is analyzed and edited, a status byte is created, and the status byte is stored in the error storage unit 6. At this time,
A flag 10 provided in the error storage section 6 indicates a free space, and when a status byte is stored, a flag 10 indicates a blockage. External storage is possible in the error storage unit 6 only when the flag 10 indicates empty space.

Next, when an error occurs in the peripheral device 2 and the peripheral control device 3 during connection control processing, the error analysis section 5 analyzes and edits the error contents and creates a status byte, but the error memory is Since the flag 10 provided in the section 6 indicates a blockage, this new status byte is input to the error comparison section 7 without performing a storage operation. Upon receiving the status byte from the error analysis unit 5, the error comparison unit 7 receives the corresponding status byte from the error storage unit 6, compares these status bytes to confirm a match, and determines whether these status bytes are the same. If so, the matching signal is sent to the fault alarm output section 9.
Output to. The fault alarm output unit 9 that has received the matching signal,
After reading the count value at this point from the counter unit 8, the count value of the counter unit 8 is reset to zero, and if the read count value is within a predetermined value, a failure is predicted and the central processing unit Outputs a failure alarm to 1. Further, if the read count value exceeds a predetermined constant value, it is determined to be normal, and the number of times this normality is detected is counted by a counter provided inside. The number of times this counting unit counts is a predetermined number within 10 times, and if normality is detected this number of times, it is determined that there is no failure with the error stored in this error storage unit 6, and a new error is detected. In order to observe this, an erase signal is output to the error storage section 6 to erase the previously temporarily stored error contents. At this time, the flag 10 provided in the error storage section 6 also indicates that it is empty.

[0014]

[Effects of the Invention] As explained above, the present invention provides the following advantages: When the peripheral device connection control processing, which is performed by receiving a command from the central processing unit, is completed, the Analyze and edit the error details,
The contents of the error are temporarily stored, and new error contents that occur during connection control processing according to instructions from the central processing unit are analyzed and edited from the next time onward. Comparison and verification are performed with the previously temporarily stored error contents, and the same If an error is detected,
The number of subsequent commands received by the central processing unit is counted until the same error as the previously temporarily stored error is detected again, and if this count is within a predetermined value, a failure is predicted. A system that outputs a failure alarm to the central processing unit, causing a failure in which the same error is repeated intermittently, but appears to be normal because it is recovered by hardware or software retry. It can be applied to systems where the number of failures does not immediately exceed a predetermined value because the number of startups is small, and it is possible to detect signs of failure and take preventive maintenance measures before a major failure occurs. There is an effect.

[Brief explanation of the drawing]

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

[Explanation of symbols]

1 Central processing unit 2 Peripheral devices 3 Peripheral control device 4 Fault alarm output circuit 5 Error analysis section 6 Error storage section 7 Error comparison section 8 Counter section 9 Fault alarm output section

Claims (3)

[Claims] 1. A failure alarm output method for a peripheral control device that is connected to a central processing unit and controls a peripheral device, in which a command from the central processing unit is received, and when the connection control process for the peripheral device is completed, the connection control process is performed. The content of errors that occur in the peripheral device and the peripheral control device is analyzed and edited, the content of the error is temporarily stored, and the content of new errors that occur during connection control processing based on instructions from the central processing unit from the next time onwards is analyzed and edited. is analyzed and edited, and compared and verified with the temporarily stored error content. If the same error is detected, the number of subsequent commands received from the central processing unit is determined to be the same as the temporarily stored error content. A failure alarm output of a peripheral control device, characterized in that it counts until an error is detected again, and if the counted value is within a predetermined value, it predicts the occurrence of a failure and outputs a failure alarm to the central processing unit. Method. [Claim 2] Analyzing and editing the content of a new error that has occurred during connection control processing according to a command from the central processing unit, and comparing the content with the previously temporarily stored error content;
If the same error is detected, the number of subsequent commands received by the central processing unit is counted until the same error as the temporarily stored error content is detected again, and this counted value exceeds a predetermined certain value. If a predetermined number of times normality is detected, the temporarily stored error content is erased, and when a new error is detected, the content is temporarily stored again. 2. A failure alarm output method for a peripheral control device according to claim 1, further comprising counting the number of times the central processing unit receives a command. 3. In a failure alarm output circuit of a peripheral control device connected to a central processing unit and controlling a peripheral device, a command from the central processing unit is received, and at the end of connection control processing of the peripheral device, the fault alarm output circuit is connected to the central processing unit and controls a peripheral device. an error analysis unit that analyzes and edits the error content that has occurred in the peripheral control device and the peripheral device; an error storage unit that temporarily stores the error content analyzed by the error analysis unit; an error comparison section that analyzes and edits newly generated error contents during subsequent connection control processing and compares and outputs the error contents with the error contents temporarily stored in the error storage section; After reading the current count value of a counter unit having a function of receiving a coincidence signal outputted when the same error is detected and counting the number of times the command is received from the central processing unit, the count value of the counter unit is zeroed. When the read count value is within a predetermined constant value, a failure is predicted to occur and a failure alarm is output to the central processing unit, and the read count value becomes equal to or higher than a predetermined constant value. If it is determined to be normal, the number of times this normality is detected is counted internally, and if normality is detected a predetermined number of times, an erase signal is output to the error storage section to erase the temporarily stored error content. A failure alarm output circuit for a peripheral control device, comprising: