JP2001265623A - Monitoring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a monitoring device capable of automatically optimizing the monitoring of plural system equipments and detecting a fault which can not detected by the system equipment to be monitored itself. SOLUTION: A state detection part 2 refers to a schedule table 3, requests a fault report to the monitoring object system equipment 12, acquires the report and samples the state information of the system equipment 12 further. An analysis part 1 recognizes the operation state of the system equipment on the basis of information inside a sample information table 4 where past information regarding the system equipment 12 is recorded, the acquired fault report and the sampled state information and updates the contents of the schedule table 3 where a parameter for monitoring is set on the basis of a recognized result. Also, the analysis part 1 compares a hardware constitution record recorded inside the sample information table 4 and hardware constitution reported by the system equipment 12 and detects the fault on the basis of a compared result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a monitoring apparatus for monitoring the operation state of system equipment constituting a computer system, and in particular, to obtain reports from system equipment to be monitored according to a schedule set by a user. The present invention relates to a monitoring device that monitors a failure by monitoring the failure.

[0002]

2. Description of the Related Art In general, there is a monitoring device as a device for continuously monitoring on behalf of a monitoring person that a plurality of system devices installed in a computer center are always operating normally. Some monitoring devices are set with a plurality of parameters for designating when and what kind of monitoring should be performed on the system equipment, and can perform monitoring in accordance with the parameters. At this time, the system device to be monitored is, for example, a host computer or an auxiliary storage device that constitutes a computer system. Some of these system devices have a function of checking an internal configuration or an operation state by itself by using a part of the processing capacity and reporting a detected failure when requested from outside.

[0003]

SUMMARY OF THE INVENTION Conventional monitoring devices include:
There are the following issues. (1) Monitoring target load Since a plurality of system devices to be monitored by the monitoring device are devices that are engaged in time-critical processing in the computer system, by monitoring, the system device has an internal configuration and operation. Since the status is checked by itself and the detected fault is reported, the load on the system equipment should be increased and the performance should not be reduced. However, even for system devices of the same type, the load greatly changes due to factors such as the hardware configuration, operation schedule, and information to be processed. For this reason, it was not possible to perform detailed monitoring according to the load state of the operating system equipment only by using the preset parameters. That is, for example, the method of monitoring when the load of the system device is 80% and the method of monitoring when the load of the system device is 40% should be different. could not. (2) Failures that cannot be detected by system devices themselves There are some failures that cannot be detected even by system devices that can report their own operation status. For example, the hardware configuration detected when the power is turned on is held as the basic configuration,
When a monitoring request is received, a system device that investigates the operation state of hardware included in the basic configuration cannot detect a failure with respect to hardware that cannot be detected since power-on. An object of the present invention is to provide a monitoring device that can automatically optimize monitoring of a plurality of system devices having the same type but different operating states, and can detect a failure that the system device to be monitored cannot detect by itself. Is to do.

[0004]

In order to achieve the above object, according to the present invention, a failure report is requested from each of a plurality of system devices according to a monitoring setting in a schedule table,
A monitoring device that monitors using a result received from the system device, a status detection unit that requests a fault report from the system device, obtains a fault report, and further obtains status information of the system device from the system device; A collection information table for recording past information on the device, and a combination of the obtained failure report, the collected status information, and past information on the system device to grasp the operation state of the system device, An analysis unit having means for updating is provided. Further, each of the system devices has a hardware configuration record, and the analysis unit detects a failure by comparing the hardware configuration reported by the system device with the hardware configuration in the hardware configuration record. Is provided.

[0005]

DESCRIPTION OF THE PREFERRED EMBODIMENTS When a monitoring device makes an inquiry to a system device to be monitored, a fault report and a hardware configuration are reported from the system device. At the same time, by evaluating the response time from issuing an inquiry to the system device to returning a response and the obtained hardware configuration against past records,
It is possible to obtain more information than the system equipment reports directly. At this time, the analysis unit performs fault analysis using the obtained information, and at the same time, performs a comparative analysis of the hardware configuration and busy rate of the system equipment with the stored past state and reference value, and performs comparison. Based on the results of the analysis,
Automatically optimize your monitoring schedule.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram illustrating a schematic configuration of the monitoring apparatus according to the embodiment. This monitoring device includes software components on six control programs including an analysis unit 1, a state detection unit 2, a schedule table 3, a collection information database 4, an input / output processing unit 5, and an alarm unit 9, a display unit 6, an operation unit It comprises four hardware components: a unit 7, an auxiliary memory 8, and an alarm device 10. The schedule table 3 includes, for each of the plurality of system devices 12 to be monitored,
Monitoring settings 1 to n indicating the monitoring priority, the monitoring execution interval, the items to be monitored, and the like are stored in advance for each date and time when monitoring is performed. The state detection unit 2 refers to the setting contents at the current time from the schedule table 3 and selects the system device whose monitoring start time has been reached from the system devices. At this time, if a plurality of system devices meet the condition, the system device with the highest priority is selected. The state detection unit 2 collects the requested information for the selected system device 12 and simultaneously measures the busy rate of the system device itself. Thereafter, the collection information held by the state detection unit 2 is sent to the analysis unit 1.

[0007] If the analysis unit 1 detects a failure based on the information obtained from the state detection unit 2, the failure is generated via the alarm unit 9, the alarm device 10, and the communication line 11 to the management center where the supervisor is stationed. Report. The collection information database 4 holds operating history information corresponding to each of the monitored system devices 12, and the busy rate is statisticized for each date and time. And records 1 to n of the system devices are held. The analysis unit 1 collects the collection information sent from the state detection unit 2 and the collection information database 4
Is compared with the history information stored in the schedule table 3, and if it is determined that the monitoring setting needs to be optimized, the schedule table 3 is optimized. After the optimization of the schedule table 3 is completed, the analysis unit 1 updates the collection information database 4 using the information obtained by the monitoring, and uses it for the next and subsequent monitoring.

The display unit 6 and the operation unit 7 are composed of, for example, a CRT display device and a keyboard.
, The initial setting of the schedule table 3 is input. At the same time, the display unit 6 and the operation unit 7 can refer to the contents of the schedule table 3 and the monitoring record 4 through the input / output processing unit 5. The auxiliary storage device 8 can input and output the contents of the schedule table 3 and the collection information database 4 held in the monitoring device.

Next, the software components 1, 2,.
The detailed operations of 3, 4, 5, 9 will be described according to the flowcharts of FIGS. 2A to 8. FIGS. 2A and 2B constitute a single figure in both figures, and are detailed diagrams in which a portion related to a flow of information generated at the time of monitoring is extracted from the schematic diagram of FIG. FIG. 3 is a schematic flowchart showing the flow of the monitoring process. FIG. 4 shows parameters input by the user at the time of initial setting. FIG. 5 is a detailed flowchart for selecting a monitoring target system device. FIG. 6A and FIG. 6B are flow charts that constitute one figure in both figures and show detailed processing from collection of information to detection of a failure. 7A and FIG. 7B show 1 in both figures.
5 is a detailed flowchart of optimizing a schedule table based on the result of analysis, which constitutes two figures. FIG. 8 is a detailed flowchart of updating the collection information database based on the analysis result. The computer system of this embodiment includes an information processing device such as a computer.

[0010] In the flowchart of FIG. 3, the monitoring device is input only once at the time of installation in S (step) 1 through the operation unit 7 or the auxiliary storage device 8 from the user through the operation unit 7 or the auxiliary storage device 8. FIG. 4 shows parameters input by the user and how those parameters are reflected in the schedule table 3 and the collection information table 4. At this time, the input parameter by the user is not a strict monitoring setting for each system device, but a value that serves as an index when the monitoring device creates a schedule.

In FIG. 4, there are two types of settings, “monitoring item” and “general device monitoring interval”, as prescribed values according to the type of system equipment 12. "Monitoring Items" refers to which collected information to detect system equipment failures.
The analysis unit 1 is programmed in advance with an item indicating a value for determining a failure and a specified value. “Monitoring item” is reflected in “Monitoring item of system device” in the collection information database 4. "General monitoring interval" indicates how often system equipment is monitored,
Similarly to the “monitoring item”, a value considered from the function of the system device is programmed in the analysis unit 1 in advance. The “general monitoring interval” is reflected in the “basic monitoring interval” in the collection information database 4. In FIG. 4, three input items by the user are “priority of system device”, “priority of each monitoring item”, and “limit of monitoring setting”.
There are different settings. The “priority of system devices” indicates which system device is important for the computer system when there are a plurality of devices of the same type in the computer system. The “system device priority” is reflected in the “basic priority” in the collection information database 4. The “priority for each monitoring item” indicates how important a failure due to the specified monitoring item is in monitoring the operating computer system. "Priority of each monitoring item"
This is reflected in “monitoring frequency” of “system device monitoring item” in the collection information database 4. The “limit of monitoring setting” corresponds to a schedule for operating the computer system when performing continuous monitoring for 24 hours, and is used for subdividing the monitoring setting. For example, 00:00 to 5:59
In a computer system with a schedule of power off until 6:00 to 11:59, accounting processing from 2:00 to 23:59, and batch processing from 12:00 to 23:59, part 3
Scheduling is performed with one as a break. The “limit of monitoring setting” is reflected in the “limit of monitoring setting” of the “history of busy rate by date and time” in the collection information database 4. 2A and 2B show an example of the contents of the collection information database 4 and an example of the contents of the schedule table 3.

In S2, the state detection unit 2 refers to the schedule table 3 at regular intervals and selects a system device to be monitored at this time from the system devices 12 to be monitored. FIG. 5 shows details of a procedure for selecting a system device to be monitored. At S21, state detection unit 2 captures the current time. After that, in S22, the state detection unit 2 refers to the “scheduled monitoring of system devices” in the schedule table 3 from each “scheduled monitoring of the next time” and sets to monitor at the current time. Search for system devices that are running. At this time, if a plurality of system devices are set to be monitored, in S23, the system devices having the highest priority are compared by comparing the "priorities" of the "scheduled monitoring of system devices". Search and select the system device to monitor.

In S3, the state detection unit 2 performs a failure detection process on the selected system device. 6A and 6
B shows details of the procedure for detecting a failure in a system device. FIG. 6A
Also, in S31 of FIG. 6B, the state detection unit 2 sets the “monitoring setting of the monitored system device” in the schedule table 3.
Refer to the "monitoring item" part of the above. The monitoring setting of the monitoring target system device indicates the item of the monitoring process to be performed according to the time and the monitoring frequency by using the “limit of monitoring setting”. S
At 32, the state detection unit 2 determines a monitoring item to request a report to the system device from the “item name” and “monitoring frequency” in the schedule table 3, and
Issues a report request according to the monitoring item. This is also shown in FIGS. 2A and 2B. In S33, the state detection unit 2 receives “hardware configuration information recognized by the system device” and “failure report” for the requested monitoring item from the system device. At the same time, the time from issuing the request to receiving the report is measured, and the busy rate of the system device is calculated from the elapsed time. This is also shown in FIGS. 2A and 2B. Thereafter, the information obtained by the monitoring is sent from the state detection unit 2 to the analysis unit 1. In S34, the analysis unit 1
Determines whether the report received from the state detection unit 2 includes a “failure report”. If a "failure report" is detected, a report is sent to a monitoring person stationed at the management center through the alarm unit 9, the alarm device 10, and the communication line 11. Thereafter, the supervisor receiving the alarm collects detailed fault information from the display unit 6, the operation unit 7, the auxiliary storage device 8, and the like through the input / output processing unit 5. In S35, the analysis unit 1 refers to “recording of the hardware configuration held by the monitoring device” for the history information of the monitoring target system device from the collection information database 4. This information is a record of the hardware configuration of the system device collected from the initial settings or past monitoring. In S36, the analysis unit 1 compares the “hardware configuration information recognized by the system device” collected in the current monitoring and the “recording of the hardware configuration held by the monitoring device” read in S33. Showing each other. At this time, even though the information is recorded in “Record of hardware configuration held by monitoring device”, it is not reported in “Hardware configuration information recognized by system equipment” and a failure report is also made. If no hardware is detected, it is considered faulty hardware that the system equipment could not be detected, and the analysis unit 1 sends a report to the supervisor as in the case of receiving a “fault report” from the system equipment. Do. In S37, if a failure is detected in this monitoring and a warning is issued, information to be used for analysis by a monitor is collected for the system equipment. At this time,
A report request is issued to the system device for all monitoring items regardless of the monitoring frequency.

At S4, the schedule table 3 is optimized. 7A and 7B show details of the schedule table optimization. In S41 of FIGS. 7A and 7B,
The analysis unit 1 compares the busy rate of the monitoring target system in the current monitoring with the “average busy rate” in the collection information database 4 at the same time in the past. In S42, this busy rate is the “average busy rate” of the system.
If the size is equal to or larger than a predetermined value determined in advance by the type of the device, for example, the current busy rate becomes 35
%, The average busy rate is 30%, and if the prescribed value predetermined by the type of device is 2%, 35% is (30%
+ 2%), the analysis unit 1 determines that the system device is overloaded, and increases the “monitoring interval” and “next monitoring schedule” in the schedule table 3 to “priority” and the monitoring item. By lowering the “monitoring frequency”, the setting is performed so as to reduce the load of monitoring. In S43, the analysis unit 1 retrieves “priority”, “monitoring interval”,
The value of “monitoring frequency” of each monitoring item at the time of user initial setting is read. In S44, if no schedule resetting factor is detected in the current monitoring and the busy rate does not significantly exceed the “average busy rate”, the system device is considered to be in a steady state. At this time, if the “priority”, the “monitoring interval”, and the “monitoring frequency” of the monitoring item in the schedule table 3 are different from the values at the time of the user initial setting, the analysis unit 1 sets the schedule table 3 to the user. Reset according to the initial settings.

At S5, the collection information database is updated. FIG. 8 shows details of updating the collection information database 4. In S51 of FIG. 8, the analysis unit 1 updates the “average busy rate” using the busy rate obtained by the current monitoring. In S52, if a failure is detected in the current monitoring, 1 is added to the "frequency of failure occurrence". In step S53, when the “number of failure occurrences” exceeds the value specified by the type of the system device, the “basic priority” and the “basic monitoring interval” are increased to increase the number of system devices in which failures frequently occur. Monitoring will be prioritized. In S54, if a change is detected in the hardware configuration of the system device in this monitoring, the “recording of the hardware configuration of the system device held by the monitoring device” in the collection information database 4 is performed.
Update according to the current configuration.

Thus, the monitoring of one monitored system device is completed. Thereafter, the schedule table 3 optimized in S4 and the collection information database updated in S5 are immediately reflected, and monitoring is repeated, thereby performing monitoring optimized for the computer system. As described above, according to this embodiment, optimal monitoring of a plurality of monitored system devices at that time is performed without monitoring personnel intervention in accordance with the operating state of the computer system or the occurrence of a failure. In this way, it is possible to schedule optimized monitoring for system devices that are processing intensive and load, so after sufficient monitoring,
It is possible to prevent performance degradation due to overload. In addition, it is possible to perform intensive monitoring on a system device which has a failure in the past and is expected to recur. Further, a hardware failure that cannot be detected by the monitored system device can be detected.

[0017]

According to the present invention, optimum monitoring of a plurality of monitored system devices at that time can be realized without intervention of a supervisor according to the operating status of the computer system or the occurrence of a failure. it can.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of a monitoring device according to an embodiment of the present invention.

FIG. 2A is a part of a detailed block diagram in which a part related to a flow of information generated at the time of monitoring is extracted from the schematic diagram of FIG. 1;

FIG. 2B is another part of the detailed block diagram in which a part related to a flow of information generated at the time of monitoring is extracted from the schematic diagram of FIG. 1;

FIG. 3 is a schematic flowchart showing the flow of a monitoring process.

FIG. 4 is a development diagram in a collection information database of parameters set by a user at the time of initial setting.

FIG. 5 is a detailed flowchart for selecting a monitoring target system device.

FIG. 6A is a part of a flowchart showing a detailed process from collection of information to detection of a failure;

FIG. 6B is another part of the flowchart showing the detailed processing from collection of information to detection of a failure.

FIG. 7A is a part of a detailed flowchart of schedule table optimization based on the result of analysis;

FIG. 7B is another part of the detailed flowchart of schedule table optimization based on the result of analysis.

FIG. 8 is a detailed flowchart of updating a collection information database based on a result of analysis.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Analysis part 2 State detection part 3 Schedule table 4 Sampling information database 5 Input / output processing part 6 Display part 7 Operation part 8 Auxiliary storage device 9 Alarm part 10 Alarm device 11 Communication line 12 Monitored system equipment

Claims (2)

[Claims] 1. A failure report is requested to each of a plurality of system devices according to a monitoring setting in a schedule table.
A monitoring device that monitors using the results received from the system equipment, requests a failure report from the system equipment, acquires a failure report,
Further, a state detection unit that collects state information of the system device from the system device, a collection information table that records past information about the system device, and the acquired failure report, the collected state information, and past information about the system device. A monitoring unit comprising means for grasping the operation state of the system device by combining the above and updating the monitor setting of the schedule table based on the grasp result. 2. The monitoring device according to claim 1, further comprising a hardware configuration record for each system device, wherein the analysis unit includes a hardware configuration reported by the system device and a hardware configuration in the hardware configuration record. A monitoring device comprising: means for detecting a failure by comparing the configuration with a configuration.