JP2017045168A - Fault processing device, system, fault management device, method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To collect fault information necessary for fault analysis and to allow the processing amount and traffic for fault processing to be restrained.SOLUTION: A fault processing device detects fault interruption indicating the occurrence of a fault and timer interruption indicating the expiration of a time set to a timer, sets a mask time corresponding to an elapsed time from first occurrence clock time of a fault to occurrence clock time this time to the timer by masking the fault interruption when detecting the fault interruption, and cancels the mask of the fault interruption when detecting the timer interruption.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a failure processing device, a system, a failure management device, a method, and a program for performing failure processing of an arithmetic processing device.

  The arithmetic processing device monitors a certain minor fault that can be continuously executed for arithmetic processing, such as occurrence of a correctable error. By monitoring this minor failure, parts can be replaced proactively before a major failure occurs that prevents continued processing, and failure trends and occurrence mechanisms can be analyzed from the collected failure information statistical data. It is possible.

  For monitoring a minor fault, it is conceivable to detect the occurrence of a minor fault, collect detailed fault information, or send fault information to a fault management device. If these fault processes are performed each time a minor fault occurs, the processing amount of the fault process increases when minor faults occur frequently, which may affect other arithmetic processes. Further, when the failure information of a plurality of arithmetic processing devices is centrally managed by the arithmetic management device, the amount of communication from the arithmetic processing device to the failure management device may increase.

  On the other hand, for example, in the method described in Patent Document 1, the notification gate is opened at a predetermined interval for a predetermined time, and the failure notification signal is input to the interrupt generation circuit only when the notification gate is open. I am doing so. As a result, the occurrence of a fault interrupt for notifying that a fault has occurred is set at regular intervals. As described above, by setting the occurrence of the failure interrupt at a constant interval, it is possible to reduce the processing amount of the failure processing and the communication amount to the failure management device even when the failure frequently occurs.

Japanese Patent Laid-Open No. 10-247148

  However, if the occurrence of a fault interrupt is set at a fixed interval as in Patent Document 1, when this interval is long, the number of times of information collection of light faults will be reduced and the fault analysis will be reduced when a light fault progresses to a heavy fault in a short time. Insufficient failure information is available. Conversely, if the interval is shortened, the processing amount of fault processing and the communication amount to the fault management device increase.

  An object of the present invention is to provide a fault processing apparatus, system, fault management apparatus, method, and program capable of collecting fault information necessary for fault analysis and suppressing the processing amount and communication amount of fault processing. It is to provide.

  In order to solve the above-described problem, the failure processing apparatus of the present invention includes a failure interrupt that indicates the occurrence of a failure, and an interrupt detection unit that detects a timer interrupt that indicates that a time set in the timer has elapsed, When the fault interrupt is detected, the fault interrupt is masked, and when the timer interrupt is detected, an interrupt control unit for canceling the mask of the fault interrupt, and when the fault interrupt is detected, the first occurrence of the fault A timer control unit that sets a mask time corresponding to an elapsed time from the time to the current occurrence time in a timer.

  In addition, the failure management apparatus of the present invention transmits the first occurrence time of the failure when the occurrence of the failure is not the first time in response to reception of the current occurrence time of the failure, and the occurrence of the failure is the first time. In this case, the present invention includes a communication unit that stores the current occurrence time in the failure storage unit as the first occurrence time and transmits the first occurrence time of the failure.

  The fault processing method of the present invention detects a fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that the time set in the timer has elapsed, and when the fault interrupt is detected, the fault interrupt is detected. A mask time corresponding to an elapsed time from the first occurrence time of the failure to the current occurrence time is set in a timer, and the mask of the failure interrupt is canceled when the timer interrupt is detected. .

  In addition, the failure management method of the present invention transmits the first occurrence time of the failure when the occurrence of the failure is not the first time in response to reception of the current occurrence time of the failure, and the occurrence of the failure is the first time. In this case, the current occurrence time is stored in a failure storage unit as the first occurrence time, and the first occurrence time of the failure is transmitted.

  In addition, the fault processing program of the present invention provides a computer with an interrupt detection function for detecting a fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that a time set in the timer has elapsed, and the fault interrupt. When detected, the fault interrupt is masked, and when the timer interrupt is detected, the interrupt control function for canceling the mask of the fault interrupt, and when the fault interrupt is detected, the current interrupt time is detected from the first occurrence time of the fault. And a timer control function for setting a mask time corresponding to an elapsed time until the occurrence time in a timer.

  The fault processing apparatus, system, fault management apparatus, method, and program of the present invention can collect fault information necessary for fault analysis and suppress the processing amount and communication amount of fault processing.

It is a figure which shows the structural example of the failure processing apparatus of 1st embodiment of this invention. It is a figure which shows the operation example of the failure processing apparatus of 1st embodiment of this invention. It is a figure which shows the structural example of the failure processing apparatus of 2nd embodiment of this invention. It is a figure which shows the structural example of the failure processing apparatus of 2nd embodiment of this invention. It is a figure which shows the operation example of the main memory part of 2nd embodiment of this invention. It is a figure which shows the operation example of the main memory part of 2nd embodiment of this invention. It is a figure which shows the operation example of the failure processing apparatus of 2nd embodiment of this invention. It is a figure which shows the specific example of the memory content of the table memory | storage part of 2nd embodiment of this invention. It is a figure which shows the specific example of the generation | occurrence | production interval of the failure interrupt of 2nd embodiment of this invention. It is a figure which shows the structural example of the failure processing apparatus of 3rd embodiment of this invention. It is a figure which shows the structural example of the failure processing apparatus of 3rd embodiment of this invention. It is a figure which shows the structural example of the failure management apparatus of 3rd embodiment of this invention. It is a figure which shows the structural example of the failure management apparatus of 3rd embodiment of this invention. It is a figure which shows the operation example of the failure processing apparatus of 3rd embodiment of this invention. It is a figure which shows the operation example of the failure management apparatus of 3rd embodiment of this invention. It is a figure which shows the operation example of the failure management apparatus of 3rd embodiment of this invention. It is a figure which shows the structural example of the failure processing apparatus of 4th embodiment of this invention. It is a figure which shows the structural example of the failure management apparatus of 4th embodiment of this invention. It is a figure which shows the operation example of the failure processing apparatus of 4th embodiment of this invention. It is a figure which shows the operation example of the failure management apparatus of 4th embodiment of this invention.

[First embodiment]
A first embodiment of the present invention will be described.

  FIG. 1 shows a configuration example of the failure processing apparatus 10 of the present embodiment.

  The failure processing device 10 includes an interrupt detection unit 11, an interrupt control unit 12, and a timer control unit 13.

  The interrupt detection unit 11 is a part that detects a fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that the time set in the timer has elapsed.

  The interrupt control unit 12 is a part that masks a fault interrupt when a fault interrupt is detected, and cancels the mask of the fault interrupt when a timer interrupt is detected.

  The timer control unit 13 is a part that sets a mask time corresponding to an elapsed time from the first occurrence time of the failure to the current occurrence time in the timer when a failure interrupt is detected.

  By configuring the failure processing apparatus 10 in this way, it is possible to collect failure information necessary for failure analysis and to suppress the processing amount and communication amount of failure processing.

  Next, FIG. 2 shows an example of the operation of the failure processing apparatus 10 of the present embodiment.

  First, the interrupt detection unit 11 of the failure processing apparatus 10 detects a failure interrupt and a timer interrupt.

  When the interrupt detection unit 11 detects a fault interrupt (step S101), the interrupt control unit 12 masks the fault interrupt (step S102). Further, the timer control unit 13 sets a mask time corresponding to the elapsed time from the first occurrence time of the failure to the current occurrence time in the timer (step S103).

  When the interrupt detection unit 11 detects a timer interrupt (step S104), the interrupt control unit 12 cancels the masking of the fault interrupt (step S105).

  By operating the failure processing apparatus 10 in this way, it is possible to collect failure information necessary for failure analysis and to suppress the processing amount and communication amount of failure processing.

  As described above, in the first embodiment of the present invention, the mask time corresponding to the elapsed time from the first occurrence time of the failure to the current occurrence time is set in the timer. As a result, the occurrence interval of the failure interrupt can be changed according to the elapsed time from the first occurrence time of the failure. By changing the mask time according to the elapsed time, it is possible to shorten the mask time for the elapsed time for which more fault information is to be collected, and to increase the mask time for the elapsed time for suppressing the processing amount and the communication amount. For this reason, it is possible to collect failure information necessary for failure analysis, and to suppress the amount of processing and the amount of communication for failure processing.

[Second Embodiment]
Next, a second embodiment of the present invention will be described.

  In the present embodiment, the failure first-time occurrence time and the correspondence between the elapsed time from the first occurrence time to the current occurrence time and the mask time are stored in the failure processing apparatus.

  FIG. 3 shows a configuration example of the failure processing apparatus 20 of the present embodiment. The failure processing device 20 includes an interrupt detection unit 11, an interrupt control unit 12, a timer control unit 13, a table storage unit 25, and a failure storage unit 26. A table storage unit 25 and a failure storage unit 26 are added to the failure processing apparatus 10.

  The interrupt detection unit 11 is a part that detects a failure interrupt from the monitoring target 50 and a timer interrupt from the timer 14. Since one or more monitoring targets 50 can exist, the interrupt detection unit 11 determines which monitoring target 50 is the fault interrupt.

  The interrupt control unit 12 masks the fault interrupt of the monitoring target 50 that has output the fault interrupt when a fault interrupt is detected, and cancels the mask of the fault interrupt when a timer interrupt from the timer 14 is detected. .

  The timer control unit 13 is a part that sets a mask time corresponding to an elapsed time from the first occurrence time of the failure to the current occurrence time in the timer when a failure interrupt is detected.

  The table storage unit 25 is a part that stores a correspondence relationship between the elapsed time from the first occurrence time of the failure to the current occurrence time and the mask time for masking the failure interrupt.

  The failure storage unit 26 is a part that stores the first occurrence time of a failure interrupt and other failure information.

  FIG. 4 shows a configuration example when the failure processing apparatus 20 of the present embodiment is applied to the arithmetic processing apparatus 100. The arithmetic processing apparatus 100 includes a DGP (Diagnostic Processor) 101, an arithmetic unit 102, and a main storage unit 103. One or more arithmetic units 102 and main storage units 103 exist in the arithmetic processing unit 100.

  The DGP 101 corresponds to the failure processing apparatus 10 in FIG. The calculation unit 102 and the main storage unit 103 correspond to the monitoring target 50. The DGP 101 receives a light failure interrupt from the computing unit 102 and the main storage unit 103 that are monitoring targets. In addition, a light failure interrupt mask control is performed on the monitoring unit 102 and the main storage unit 103.

  By configuring the failure processing device 20 in this way, it is possible to collect failure information necessary for failure analysis and to reduce the processing amount and communication amount of failure processing.

  Next, an example of the operation of the main storage unit 103 will be described as an example of the operation of the monitoring target 50. The monitoring target 50 detects a failure and outputs a failure interrupt to the failure processing apparatus 20. The following operation example of the main storage unit 103 is an example of the failure interrupt output method, and does not limit the failure detection method, interrupt output method, and the like of the monitoring target 50.

  5 and 6 show an example of the operation of the main storage unit 103. FIG. 5A shows an example of the operation at the time of writing, FIG. 5B shows an example of the operation at the time of reading, and FIG. 6 shows an example of the interrupt output operation.

  When writing from the calculation unit 102, first, the main storage unit 103 generates a check bit for the write data (step S201 in FIG. 5). Then, write data and check bits are written into the memory (step S202).

  When reading from the computing unit 102, first, the main storage unit 103 reads data including check bits from the memory (step S203). Then, a syndrome is generated from the read data and check bits, and error detection is performed (step S204). If there is an error (step S205), the data error is corrected (step S206). Then, detailed information on the failure, for example, the occurrence time of the failure, the cause of the failure (the read data has an error, etc.), etc. are stored (step S207).

  Next, it is determined whether the error is minor or serious (step S208). For example, a light failure is assumed when the data error can be corrected, and a serious failure is found when the error cannot be corrected. If the fault is minor, a minor fault flag is set (step S209). If the fault is serious, the major fault flag is set (step S210).

  The example of the fault interrupt output operation in FIG. 6 is common to the case of a minor fault and the case of a major fault. When the light failure (serious failure) flag is set, the main storage unit 103 issues a light failure (serious failure) interrupt when the mask of the light failure (serious failure) interrupt is released (step S301 in FIG. 6). Output (step S302).

  When the light failure (serious failure) interrupt mask is released, if the light failure (serious failure) flag is set (step S303), a light failure (serious failure) interrupt is output (step S304).

  Although the example of the operation of the main storage unit 103 has been described above, the failure interrupt output from the monitoring target 50 can be caused by a failure other than that caused by error detection. Further, the monitoring target 50 is not limited to the main storage unit 103, and each unit in the failure processing apparatus 20 and devices other than the failure processing apparatus 20 can be set as monitoring targets.

  Next, an example of the operation of the failure processing apparatus 20 of this embodiment will be described with reference to FIG.

  In the present embodiment, the interrupt detection unit 11 detects a fault interrupt and a timer interrupt of the monitoring target 50. For example, when the main storage unit 103 is the monitoring target 50, it is conceivable that the interrupt interval is controlled for a light fault interrupt, and an interrupt is output whenever a heavy fault interrupt occurs. In this case, the light fault interrupt corresponds to the fault interrupt of FIG.

  First, the interrupt detection unit 11 detects a failure interrupt and a timer interrupt. If the interrupt detection unit 11 detects a fault interrupt (step S401), the interrupt control unit 12 masks the fault interrupt (step S402). More specifically, a setting (mask) for prohibiting the output of the fault interrupt is performed on the monitoring target 50 that has output the fault interrupt.

  When there are a plurality of monitoring targets 50, the interrupt detection unit 11 specifies from which monitoring target 50 the detected fault interrupt is the fault interrupt when the fault interrupt is detected. When a fault interrupt from each monitoring target 50 is input to an independent interrupt input terminal, it is possible to identify the monitoring target 50 that has output the interrupt depending on which input terminal the interrupt is input to. When fault interrupts of a plurality of monitoring targets 50 are input to one input terminal, for example, the fault flag of the monitoring target 50 is read, and if the flag is set, it is determined that the interrupt is output. , Etc. can be specified. In the case of the above-described example of the main storage unit 103, it is possible to read a light (heavy) trouble flag and determine that a light (heavy) trouble has occurred if the flag is set. Note that the fault interrupt of the monitoring target 50 for which the mask is set is not detected.

  Next, the interrupt detection unit 11 acquires detailed information on a failure necessary for failure analysis or the like from the monitoring target 50 that has output the failure interrupt (step S403). For example, the main storage unit 103 reads the detailed information stored in step S207. The acquired detailed information is stored in the failure storage unit 26. Further, the light failure flag and the detailed information stored in the main storage unit 103 are cleared.

  In the present embodiment, at least for each monitoring target 50, the time when the failure first occurred is stored in the failure storage unit 26. The occurrence time may be acquired from the monitoring target 50, or may be the time when the interrupt detection unit 11 detects a failure interrupt. Further, the first occurrence time may be stored for each cause of failure (correctable error or the like).

  Next, the timer control unit 13 acquires the first occurrence time of the failure from the failure storage unit 26 for the monitoring target 50 that has output the failure interrupt (step S404). When the first occurrence time is stored for each cause of failure, the first occurrence time is acquired for the current failure content.

  Then, the timer control unit 13 calculates the elapsed time from the first occurrence time to the current failure occurrence time (step S405). If this is the first time, the elapsed time is zero.

  Next, the timer control unit 13 acquires from the table storage unit 25 the mask time corresponding to the elapsed time for masking the failure interrupt. The table storage unit 25 stores a mask time corresponding to the elapsed time.

  FIG. 8 shows an example of the correspondence relationship between the elapsed time and the mask time. In this example, the mask time is set longer as the elapsed time from the first occurrence time of the failure becomes longer.

  FIG. 9 shows the occurrence intervals of fault interrupts when the correspondence between the elapsed time and the mask time is that shown in FIG. By increasing the mask time as the elapsed time from the first occurrence time becomes longer, failure information is acquired at a short interval in the early stage of failure occurrence, which is important for failure analysis, and the failure information acquisition interval is increased with time. Thus, it is possible to reduce the fault handling load.

  Then, the timer control unit 13 sets the mask time acquired from the table storage unit 25 in the timer 14 (step S406). The timer outputs a timer interrupt to the interrupt detection unit 11 when the set mask time has elapsed.

  When the interrupt detection unit 11 detects a timer interrupt (step S407), the interrupt control unit 12 cancels the fault interrupt mask of the monitoring target 50 corresponding to the timer interrupt (step S408).

  By operating the failure processing apparatus 20 in this way, it is possible to collect failure information necessary for failure analysis, and to suppress the processing amount and communication amount of failure processing.

  As described above, in the second embodiment of the present invention, as in the first embodiment, the mask time corresponding to the elapsed time from the first occurrence time of the failure to the current occurrence time is set in the timer. . As a result, the occurrence interval of the failure interrupt can be changed according to the elapsed time from the first occurrence time of the failure. By changing the mask time according to the elapsed time, it is possible to shorten the mask time for the elapsed time for which more fault information is to be collected, and to increase the mask time for the elapsed time for suppressing the processing amount and the communication amount. For this reason, it is possible to collect failure information necessary for failure analysis, and to suppress the amount of processing and the amount of communication for failure processing.

  In addition, the longer the elapsed time, the longer the mask time, so that failure information is acquired at short intervals in the early stage of failure occurrence, which is important for failure analysis, and the failure information acquisition interval increases with time. It becomes possible to reduce processing load.

[Third embodiment]
Next, a third embodiment of the present invention will be described.

  In this embodiment, an example in which the failure occurrence time is managed by the failure management apparatus will be described.

  FIG. 10 shows a configuration example of the failure processing apparatus 30 of the present embodiment. For the failure processing apparatus 20, the failure storage unit 26 is deleted and a communication unit 37 is added.

  The communication unit 37 is a part that communicates with the failure management apparatus 70. Detailed information on the failure is transmitted to the failure management device 70, and the first occurrence time or elapsed time of the failure is received from the failure management device 70. As the detailed information, at least the information for specifying the monitoring target 50 in which the failure has occurred and the time at which the failure has occurred are transmitted. Also, the cause of failure and information necessary for failure analysis are transmitted.

  When receiving the first occurrence time from the failure management device 70, the timer control unit 13 calculates the elapsed time from the first occurrence time and the current occurrence time.

  FIG. 11 shows a configuration example when the failure processing apparatus 30 of the present embodiment is applied to an arithmetic processing apparatus. A configuration example of the arithmetic processing devices 100A and 100X is the same as that of the arithmetic processing device 100 in FIG. The DGP 101 corresponds to the failure processing apparatus 30. The calculation unit 102 and the main storage unit 103 correspond to the monitoring target 50.

  The SVP (Service Processor) 200 corresponds to the failure management device 70. In the SVP 200, fault information from one or more DGPs 101 is collectively managed.

  FIG. 12 shows a configuration example of the failure management device 70. The failure management device 70 includes a failure storage unit 71 and a communication unit 72. The communication unit 72 is a part that communicates with the failure processing apparatus 30. Detailed information and occurrence time of the failure are received from the failure processing device 30 and stored in the failure storage unit 71. In addition, the first occurrence time of the failure is transmitted to the failure processing apparatus 30.

  The failure management apparatus can be configured as the failure management apparatus 80 in FIG. A calculation unit 83 is added to the failure management apparatus 70 of FIG. The calculation unit 83 is a part that calculates the elapsed time from the first occurrence time of the failure to the current failure occurrence time.

  By configuring the failure processing device 30 and the failure management device 70 (80) in this way, it is possible to collect failure information necessary for failure analysis and to reduce the amount of processing and communication of failure processing. .

  Next, operation examples of the failure processing device 30 and the failure management device 70 (80) of this embodiment will be described with reference to FIGS.

  First, the interrupt detection unit 11 detects a fault interrupt (step S501 in FIG. 14), and masks the fault interrupt of the monitoring target 50 that has output the fault interrupt (step S502). Further, the detailed information on the failure is acquired from the monitoring target 50 (step S503).

Next, the communication unit 37 transmits the detailed information and the occurrence time of the failure to the failure management device 70 (80). The failure management device 70 (80) receives the failure occurrence time (step S601 in FIG. 15 and step S701 in FIG. 16), and stores the occurrence time when the failure occurs for the first time. When the failure management device is the failure management device 70, the failure management device 70 acquires the first occurrence time of the failure from the failure storage unit 71 and transmits it to the failure processing device 30 (step S602). When the failure management device is the failure management device 80, the calculation unit 83 of the failure management device 80 calculates the elapsed time from the failure first occurrence time to the current failure occurrence time (step S702), and calculates the elapsed time. It transmits to the failure processing apparatus 30 (step S703). Then, the communication unit 37 acquires the first occurrence time or elapsed time from the failure management device (70 or 80). (Step S504)
Next, the timer control unit 13 acquires a mask time corresponding to the elapsed time from the table storage unit 25 and sets it in the timer 14 (step S505). When the first occurrence time is received from the failure management device 70, the timer control unit 13 calculates the elapsed time.

  When the mask time has elapsed, the timer 14 outputs a timer interrupt to the interrupt detection unit 11, and the interrupt detection unit 11 detects the timer interrupt (step S506).

  Then, the interrupt control unit 12 cancels the masking of the fault interrupt (Step S507).

  By operating the failure processing device 30 and the failure management device 70 (80) in this way, it is possible to collect failure information necessary for failure analysis and to suppress the processing amount and communication amount of failure processing. .

  As described above, in the third embodiment of the present invention, as in the first and second embodiments, the mask time corresponding to the elapsed time from the first occurrence time of the failure to the current occurrence time is set to the timer. Set to. As a result, the occurrence interval of the failure interrupt can be changed according to the elapsed time from the first occurrence time of the failure. By changing the mask time according to the elapsed time, it is possible to shorten the mask time for the elapsed time for which more fault information is to be collected, and to increase the mask time for the elapsed time for suppressing the processing amount and the communication amount. For this reason, it is possible to collect failure information necessary for failure analysis, and to suppress the amount of processing and the amount of communication for failure processing.

  In addition, the longer the elapsed time, the longer the mask time, so that failure information is acquired at short intervals in the early stage of failure occurrence, which is important for failure analysis, and the failure information acquisition interval increases with time. It becomes possible to reduce processing load.

  Furthermore, in the present embodiment, since the failure information is managed by the failure management device, it is not necessary to store the failure information in the failure processing device, and the failure processing device can be made smaller.

[Fourth embodiment]
Next, a fourth embodiment of the present invention will be described.

  In the present embodiment, an example will be described in which the failure management device manages the first occurrence time of a failure and stores the correspondence between the elapsed time from the failure first occurrence time and the mask time in the failure management device.

  FIG. 17 shows a configuration example of the failure processing apparatus 40 of the present embodiment. The table storage unit 25 is deleted from the failure processing apparatus 30. The communication unit 37 transmits the current occurrence time of the failure to the failure management device 90 and acquires the mask time from the failure management device 90. The timer control unit 13 sets the mask time received from the failure management device 90 in the timer.

  FIG. 18 shows a configuration example of the failure management apparatus 90 of this embodiment. A table storage unit 94 is added to the failure management device 80. The table storage unit 94 is a part that stores the correspondence between the elapsed time from the first occurrence time of the failure and the mask time. Note that this correspondence can be different for each monitoring target 50.

  The calculation unit 83 calculates the elapsed time from the first occurrence time of the failure to the current occurrence time, and acquires the mask time corresponding to the elapsed time from the table storage unit 94. The communication unit 72 transmits the mask time to the failure management apparatus 90.

  By configuring the failure processing device 40 and the failure management device 90 in this way, it is possible to collect failure information necessary for failure analysis and to suppress the processing amount and communication amount of failure processing.

  Next, operation examples of the failure processing device 40 and the failure management device 90 of this embodiment will be described with reference to FIGS. 19 and 20.

  First, the interrupt detection unit 11 detects a fault interrupt (step S801), and masks the fault interrupt of the monitoring target 50 that has output the fault interrupt (step S802). Also, detailed information on the failure is acquired from the monitoring target 50 (step S803).

  Next, the communication unit 37 transmits a failure occurrence time to the failure management apparatus 90. The failure management device 90 receives the failure occurrence time (step S901), and stores the first occurrence time in the failure storage unit 71 when the failure occurs for the first time. Then, the first occurrence time is acquired from the failure storage unit 71, and the elapsed time from the first occurrence time to the current occurrence time is calculated (step S902). Further, the mask time corresponding to the elapsed time is acquired from the table storage unit 94. The communication unit 72 transmits the mask time to the failure processing apparatus 40 (step S903). Then, the communication unit 37 of the failure processing device 40 receives the mask time from the failure management device 90 (step S804).

  Next, the timer control unit 13 sets the received mask time in the timer 14 (step S805).

  When the mask time has elapsed, the timer 14 outputs a timer interrupt to the interrupt detection unit 11, and the interrupt detection unit 11 detects the timer interrupt (step S806).

  Then, the interrupt control unit 12 cancels the masking of the failure interrupt (Step S807).

  By operating the failure processing device 40 and the failure management device 90 in this way, it is possible to collect failure information necessary for failure analysis and to suppress the processing amount and communication amount of failure processing.

  As described above, in the fourth embodiment of the present invention, as in the first to third embodiments, the mask time corresponding to the elapsed time from the first occurrence time of the failure to the current occurrence time is set to the timer. Set to. As a result, the occurrence interval of the failure interrupt can be changed according to the elapsed time from the first occurrence time of the failure. By changing the mask time according to the elapsed time, it is possible to shorten the mask time for the elapsed time for which more fault information is to be collected, and to increase the mask time for the elapsed time for suppressing the processing amount and the communication amount. For this reason, it is possible to collect failure information necessary for failure analysis, and to suppress the amount of processing and the amount of communication for failure processing.

  In addition, the longer the elapsed time, the longer the mask time, so that failure information is acquired at short intervals in the early stage of failure occurrence, which is important for failure analysis, and the failure information acquisition interval increases with time. It becomes possible to reduce processing load.

  Furthermore, in the present embodiment, since the failure information is managed by the failure management device, it is not necessary to store the failure information in the failure processing device, and the failure processing device can be made smaller.

  Each of the above-described embodiments is a preferred embodiment of the present invention, and various modifications can be made without departing from the scope of the present invention.

  A part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.

(Appendix 1)
An interrupt detection unit for detecting a fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that the time set in the timer has elapsed;
An interrupt control unit that masks the fault interrupt when detecting the fault interrupt, and cancels the mask of the fault interrupt when the timer interrupt is detected;
A failure processing apparatus comprising: a timer control unit that sets a mask time corresponding to an elapsed time from an initial occurrence time of the failure to a current occurrence time when the failure interrupt is detected.

(Appendix 2)
The failure processing apparatus according to appendix 1, wherein the timer control unit increases the mask time as the elapsed time is longer.

(Appendix 3)
The interrupt detection unit acquires a cause of the failure,
The failure processing apparatus according to appendix 1 or appendix 2, wherein the first occurrence time is a time when the failure causing the failure is generated for the first time.

(Appendix 4)
The timer control unit acquires the mask time corresponding to the elapsed time from the correspondence relationship of the table storage unit that stores the correspondence relationship between the elapsed time and the mask time. 4. The failure processing apparatus according to any one of 3.

(Appendix 5)
The interrupt detection unit stores the first occurrence time of the failure in a failure storage unit,
The failure processing apparatus according to appendix 4, wherein the timer control unit calculates the elapsed time from the first occurrence time acquired from the failure storage unit and the current occurrence time.

(Appendix 6)
A communication unit for transmitting the current occurrence time and receiving the first occurrence time when detecting the failure interrupt;
The failure processing apparatus according to appendix 4, wherein the timer control unit calculates the elapsed time from the first occurrence time and the current occurrence time.

(Appendix 7)
The fault processing apparatus according to appendix 4, further comprising: a communication unit that transmits the current occurrence time and receives the elapsed time when the fault interrupt is detected.

(Appendix 8)
The failure processing apparatus according to any one of appendix 1 to appendix 3, further comprising: a communication unit that transmits the current occurrence time and receives the mask time when the failure interrupt is detected.

(Appendix 9)
In response to reception of the current occurrence time of the failure, if the occurrence of the failure is not the first time, the first occurrence time of the failure is transmitted, and if the occurrence of the failure is the first time, the first occurrence time is A failure management apparatus comprising: a communication unit that stores a current occurrence time in a failure storage unit and transmits the first occurrence time of the failure.

(Appendix 10)
The communication unit further receives the cause of the failure,
The failure management apparatus according to appendix 9, wherein the first occurrence time is a time at which the failure causing the failure occurs for the first time.

(Appendix 11)
If the occurrence of the failure is not the first time in response to reception of the current occurrence time of the failure, an elapsed time from the first occurrence time of the failure to the current occurrence time is transmitted, and the occurrence of the failure is the first time. In this case, the current occurrence time is stored in the failure storage unit as the first occurrence time, and the communication unit transmits the elapsed time;
A failure management apparatus comprising: a calculation unit that calculates the elapsed time.

(Appendix 12)
The communication unit further receives the cause of the failure,
The failure management apparatus according to appendix 11, wherein the first occurrence time is a time when the failure causing the failure occurs for the first time.

(Appendix 13)
When the occurrence of the failure is not the first time with respect to reception of the current occurrence time of the failure, a mask time for masking the failure interrupt indicating the occurrence of the failure is transmitted, and when the occurrence of the failure is the first time A communication unit that stores the current occurrence time in the failure storage unit as the first occurrence time, and transmits the mask time;
The elapsed time from the first occurrence time of the failure to the current occurrence time is calculated, and the mask time corresponding to the elapsed time is obtained from a table storage unit that stores the correspondence between the elapsed time and the mask time. A failure management device comprising:

(Appendix 14)
The communication unit further receives the cause of the failure,
The failure management apparatus according to appendix 13, wherein the first occurrence time is a time at which the failure causing the failure occurs for the first time.

(Appendix 15)
The fault handling apparatus according to appendix 6,
The failure management device according to appendix 9 or appendix 10, and
The failure is characterized in that the communication unit of the failure processing device transmits the current occurrence time to the communication unit of the failure management device and receives the first occurrence time from the communication unit of the failure management device. Processing system.

(Appendix 16)
The fault handling apparatus according to appendix 7,
The failure management device according to appendix 11 or appendix 12, and
The failure processing device, wherein the communication unit of the failure processing device transmits the current occurrence time to the communication unit of the failure management device and receives the elapsed time from the communication unit of the failure management device. system.

(Appendix 17)
The failure handling apparatus according to appendix 8,
The failure management device according to appendix 13 or appendix 14, and
The failure processing device, wherein the communication unit of the failure processing device transmits the current occurrence time to the communication unit of the failure management device and receives the mask time from the communication unit of the failure management device. system.

(Appendix 18)
A fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that the time set in the timer has elapsed are detected. When the fault interrupt is detected, the fault interrupt is masked and the fault is first generated. A failure processing method, wherein a mask time corresponding to an elapsed time up to the current occurrence time is set in a timer, and the mask of the failure interrupt is canceled when the timer interrupt is detected.

(Appendix 19)
The failure processing method according to appendix 18, wherein the mask time is lengthened as the elapsed time is longer.

(Appendix 20)
The failure processing method acquires a failure cause of the failure,
The failure processing method according to appendix 18 or appendix 19, wherein the first occurrence time is a time when the failure causing the failure is first generated.

(Appendix 21)
Any one of appendix 18 to appendix 20, wherein the mask time corresponding to the elapsed time is acquired from the correspondence stored in a table storage unit that stores the correspondence between the elapsed time and the mask time. The fault processing method according to the above.

(Appendix 22)
The failure processing method according to appendix 21, wherein the first occurrence time is stored in a failure storage unit, and the elapsed time is calculated from the first occurrence time acquired from the failure storage unit and the current occurrence time. .

(Appendix 23)
When the failure interrupt is detected, the current occurrence time is transmitted, the first occurrence time is received, and the elapsed time is calculated from the first occurrence time and the current occurrence time. The failure handling method described in 1.

(Appendix 24)
The failure processing method according to appendix 21, wherein when the failure interrupt is detected, the current occurrence time is transmitted and the elapsed time is received.

(Appendix 25)
The fault processing method according to any one of appendix 18 to appendix 20, wherein when the fault interrupt is detected, the current occurrence time is transmitted and the mask time is received.

(Appendix 26)
In response to reception of the current occurrence time of the failure, if the occurrence of the failure is not the first time, the first occurrence time of the failure is transmitted, and if the occurrence of the failure is the first time, the first occurrence time is A failure management method comprising storing a current occurrence time in a failure storage unit and transmitting the first occurrence time of the failure.

(Appendix 27)
The failure management method receives a failure cause of the failure,
27. The failure management method according to appendix 26, wherein the first occurrence time is a time when the failure causing the failure occurs for the first time.

(Appendix 28)
For the reception of the current occurrence time of the failure, if the occurrence of the failure is not the first time, calculate the elapsed time from the first occurrence time of the failure to the current occurrence time, and transmit the elapsed time, When the failure occurs for the first time, the failure occurrence method is characterized in that the current occurrence time is stored in a failure storage unit as the first occurrence time, the elapsed time is calculated and the elapsed time is transmitted. .

(Appendix 29)
The failure management method receives a failure cause of the failure,
29. The failure management method according to appendix 28, wherein the first occurrence time is a time when the failure causing the failure occurs for the first time.

(Appendix 30)
If the occurrence of the failure is not the first time with respect to reception of the current occurrence time of the failure, an elapsed time from the initial occurrence time of the failure to the current occurrence time is calculated, and the elapsed time and the mask time When the occurrence of the failure is the first time, the mask time for masking the failure interrupt indicating the occurrence of the failure corresponding to the elapsed time is acquired from the table storage unit storing the correspondence relationship with Stores the current occurrence time in the failure storage unit as the first occurrence time, calculates the elapsed time, obtains the mask time corresponding to the elapsed time from the table storage unit, and transmits the mask time A failure management method characterized by:

(Appendix 31)
The failure management method receives a failure cause of the failure,
The failure management method according to appendix 30, wherein the first occurrence time is a time at which the failure causing the failure occurs for the first time.

(Appendix 32)
On the computer,
An interrupt detection function for detecting a fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that the time set in the timer has elapsed;
An interrupt control function for masking the fault interrupt when detecting the fault interrupt, and canceling the mask of the fault interrupt when detecting the timer interrupt;
And a timer control function for setting, in a timer, a mask time corresponding to an elapsed time from the first occurrence time of the failure to the current occurrence time when the failure interrupt is detected.

(Appendix 33)
The failure processing program according to appendix 32, wherein the timer control function increases the mask time as the elapsed time increases.

(Appendix 34)
The interrupt detection function acquires the cause of the failure,
The failure processing program according to appendix 32 or appendix 33, wherein the first occurrence time is a time when the failure causing the failure occurs for the first time.

(Appendix 35)
From the supplementary note 32, the timer control function acquires the mask time corresponding to the elapsed time from the correspondence relation of a table storage unit that stores the correspondence relation between the elapsed time and the mask time. 34. The failure processing program according to any one of 34.

(Appendix 36)
The failure processing program according to appendix 35, wherein the timer control function calculates the elapsed time from the first occurrence time acquired from the failure storage unit that stores the first occurrence time and the current occurrence time. .

(Appendix 37)
To the computer,
When the failure interrupt is detected, a communication function for transmitting the current occurrence time and receiving the first occurrence time is realized,
The failure processing program according to appendix 35, wherein the timer control function calculates the elapsed time from the first occurrence time and the current occurrence time.

(Appendix 38)
To the computer,
36. The fault processing program according to appendix 35, wherein when the fault interrupt is detected, a communication function for transmitting the current occurrence time and receiving the elapsed time is realized.

(Appendix 39)
To the computer,
The failure processing program according to any one of appendix 32 to appendix 34, wherein when the fault interrupt is detected, a communication function for transmitting the current occurrence time and receiving the mask time is realized.

(Appendix 40)
On the computer,
In response to reception of the current occurrence time of the failure, if the occurrence of the failure is not the first time, the first occurrence time of the failure is transmitted, and if the occurrence of the failure is the first time, the first occurrence time is A failure management program for storing a current occurrence time in a failure storage unit and realizing a communication function for transmitting the first occurrence time of the failure.

(Appendix 41)
The communication function further receives the cause of the failure,
The failure management program according to appendix 40, wherein the first occurrence time is a time at which the failure causing the failure occurs for the first time.

(Appendix 42)
On the computer,
If the occurrence of the failure is not the first time in response to reception of the current occurrence time of the failure, an elapsed time from the first occurrence time of the failure to the current occurrence time is transmitted, and the occurrence of the failure is the first time. In this case, the communication function for storing the current occurrence time in the failure storage unit as the first occurrence time and transmitting the elapsed time;
A failure management program that realizes a calculation function for calculating the elapsed time.

(Appendix 43)
The communication function further receives the cause of the failure,
The failure management program according to appendix 42, wherein the first occurrence time is a time at which the failure causing the failure is first generated.

(Appendix 44)
On the computer,
When the occurrence of the failure is not the first time with respect to reception of the current occurrence time of the failure, a mask time for masking the failure interrupt indicating the occurrence of the failure is transmitted, and when the occurrence of the failure is the first time A communication function for storing the current occurrence time as a first occurrence time in a failure storage unit and transmitting the mask time;
The elapsed time from the first occurrence time of the failure to the current occurrence time is calculated, and the mask time corresponding to the elapsed time is obtained from a table storage unit that stores the correspondence between the elapsed time and the mask time. A fault management program characterized by realizing a calculation function.

(Appendix 45)
The communication function further receives the cause of the failure,
45. The failure management program according to appendix 44, wherein the first occurrence time is a time at which the failure causing the failure is first generated.

10, 20, 30, 40 Fault processing device 11 Interrupt detection unit 12 Interrupt control unit 13 Timer control unit 14 Timer 25 Table storage unit 26 Fault storage unit 37 Communication unit 50 Monitoring target 70, 80, 90 Fault management device 71 Fault Storage unit 72 Communication unit 83 Calculation unit 94 Table storage unit

Claims (10)

An interrupt detection unit for detecting a fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that the time set in the timer has elapsed;
An interrupt control unit that masks the fault interrupt when detecting the fault interrupt, and cancels the mask of the fault interrupt when the timer interrupt is detected;
A failure processing apparatus comprising: a timer control unit that sets a mask time corresponding to an elapsed time from an initial occurrence time of the failure to a current occurrence time when the failure interrupt is detected. The failure processing apparatus according to claim 1, wherein the timer control unit increases the mask time as the elapsed time is longer. The said timer control part acquires the said mask time corresponding to the said elapsed time from the said correspondence of the table memory | storage part which memorize | stores the correspondence between the said elapsed time and the said mask time. The failure processing apparatus according to claim 2. The interrupt detection unit stores the first occurrence time in a failure storage unit,
The failure processing apparatus according to claim 3, wherein the timer control unit calculates the elapsed time from the first occurrence time acquired from the failure storage unit and the current occurrence time. A communication unit for transmitting the current occurrence time and receiving the first occurrence time when detecting the failure interrupt;
The failure processing apparatus according to claim 3, wherein the timer control unit calculates the elapsed time from the first occurrence time and the current occurrence time. In response to reception of the current occurrence time of the failure, if the occurrence of the failure is not the first time, the first occurrence time of the failure is transmitted, and if the occurrence of the failure is the first time, the first occurrence time is A failure management apparatus comprising: a communication unit that stores a current occurrence time in a failure storage unit and transmits the first occurrence time of the failure. A fault handling apparatus according to claim 5;
The failure management device according to claim 6,
The failure is characterized in that the communication unit of the failure processing device transmits the current occurrence time to the communication unit of the failure management device and receives the first occurrence time from the communication unit of the failure management device. Processing system. A fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that the time set in the timer has elapsed are detected. When the fault interrupt is detected, the fault interrupt is masked and the fault is first generated. A failure processing method, wherein a mask time corresponding to an elapsed time up to the current occurrence time is set in a timer, and the mask of the failure interrupt is canceled when the timer interrupt is detected. In response to reception of the current occurrence time of the failure, if the occurrence of the failure is not the first time, the first occurrence time of the failure is transmitted, and if the occurrence of the failure is the first time, the first occurrence time is A failure management method comprising storing a current occurrence time in a failure storage unit and transmitting the first occurrence time of the failure. On the computer,
An interrupt detection function for detecting a fault interrupt indicating the occurrence of a fault and a timer interrupt indicating that the time set in the timer has elapsed;
An interrupt control function for masking the fault interrupt when detecting the fault interrupt, and canceling the mask of the fault interrupt when detecting the timer interrupt;
And a timer control function for setting, in a timer, a mask time corresponding to an elapsed time from the first occurrence time of the failure to the current occurrence time when the failure interrupt is detected.

Images