JPS6313560Y2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a device for recording failure conditions in order to investigate the cause of intermittent hardware failures in a computer system.

In conventional computer systems, when a hardware failure occurs, if the environmental conditions change, such as when the commercial power supply is restored, the failure can be recovered without any special measures, or it may not be possible to recover from the failure. do. Systems with complex operations are often affected by environmental conditions such as voltage, temperature, humidity, and static electricity, but in the past, it was almost impossible to accurately understand these effects and correlate them with the occurrence of failures. Furthermore, when warning of the occurrence of a failure, the warning was only a buzzer to notify of a commercial power outage or an abnormal rise in temperature. Therefore, unless the appropriate sensor is activated when a failure occurs,
No attempt was made to record the circumstances before and after the incident and investigate later. When a general user of a computer system informs the manufacturer that a failure has occurred during use, there may be no abnormalities at all, and the cause of the failure is often unknown.

The purpose of this invention is to provide a sensor for monitoring environmental conditions,
An object of the present invention is to provide a device for recording failure conditions for investigating the cause of intermittent hardware failures in a computer system.

Embodiments of the present invention shown in the drawings will be described below. Figure 1 shows an embodiment of the present invention as a block configuration diagram, where PCS is a computer system, CPU is a target processing device, ED is an operational failure detection device,
SNS1, SNS2...SNSn is a sensor, CC is an environmental condition monitoring circuit, SPC is an auxiliary processing device, LGM is a data storage medium, DBS is a data bus, and SPC
and LGM form a storage control device with a data storage section. Here, the signal detected by the sensor SNS is constantly monitored by the environmental condition monitoring circuit CC, which will be described later, and the signal detected by the operation failure detection device ED is stored in the data storage medium LGM under the control of the storage unit-equipped control device.

FIG. 2 shows a detailed diagram of the environmental condition monitoring circuit CC.
Prepare a plurality of each for RF as a reference value, CMP as a comparison circuit, and HLD as a holding circuit. The comparison circuit CMP detects that the sensor SNS output exceeds the reference value RF.
When detected, it sends a signal to the auxiliary processing unit SPC via the signal holding circuit HLD. SW indicates a selection switch, and RSS indicates a reset signal from the auxiliary processing unit SPC. In addition to being operated by the operator, the selection switch SW can also be commanded by the auxiliary processing device SPC to switch the value of the reference value RF. Figure 3 shows an operation time chart of the circuit in Figure 2,
Each interval of m-1 to m+1 and n-1 to n+1 corresponds to the interval of the reset signal described above. holding circuit
For example, the existence of a certain output of HLD in m-1 to m means that the sensor SNS has detected a certain abnormality.
If the operational failure detection device ED does not detect an abnormality at that time, but detects it from m to m+1, it determines that the operational failure of the processing unit CPU is not related to changes in environmental conditions, and
Nothing is stored in LGM. If time interval n~n+
As shown in Figure 1, when the output of the holding circuit HLD and the output of the failure detection device ED coexist, it is considered that an operational failure has occurred in the processing unit CPU related to the environmental conditions, and the situation of the holding circuit HLD and the failure detection device ED are considered to have occurred. and the current status are stored in the data storage medium LGM. Specifically, when the detection device ED detects an operational failure in the processing unit CPU, it performs a series of processes to collect storage information, checks the status of the sensor holding circuit HLD, and collects the storage information from the same period. Store using the same area as the storage information described above.

It should be noted that, compared to the reset intervals such as n-1 to n+1 in FIG. 3, when the failure time continues, the holding circuit HLD is unnecessary.

In this way, according to the present invention, it is possible to easily store only the data that has a temporal correlation between the sensor output and the occurrence of an operational failure in the monitored processing device, so that the operator/maintenance person can It can also be effectively used as data to find the cause of failures that occur when you are not present.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a detailed diagram of the environmental condition monitoring circuit in FIG. 1, and FIG. 3 is an operation time chart in FIG. 2. PCS...computer system, CPU...processing device,
ED...Operational failure detection device, SNS...Sensor, CC
...Environmental condition monitoring circuit, SPC...Auxiliary processing device, LGM...Data storage medium, DBS...Reference value, CMP...Comparison circuit, HLD...Holding circuit,
RSS...Reset circuit.

Claims (1)

[Scope of utility model registration request] A plurality of sensors that monitor and control a monitored device in a computer system while changing a reference value, a circuit that receives and processes the sensor output, an operation failure detection device for the monitored device, and a control device with a data storage unit. and when the control device with data storage unit determines that a failure of the monitored device and the occurrence of the sensor output are temporally related, the sensor output and the failure status of the monitored device are stored in the data storage unit. A failure recording device for a computer system characterized by storing information.