JP2012247937A - Information processing unit, log storage control program, and log storage control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To store a log which is effective for a cause analysis of a trouble event.SOLUTION: A detection portion 21 detects a load state of software. When a log of the software occurs, a determination portion 22 determines, in accordance with a type of the log and the load state detected by the detection portion 21, to store, in a first load state, the log regarding a first type into a nonvolatile storage portion 25 and the log regarding a second type into a volatile storage portion 24, and to store, in a second load state in which the load is heavier than in the first load state, the log regarding the first type and the log regarding the second type into the nonvolatile storage portion 25.

Description

  The present invention relates to an information processing apparatus, a log storage control program, and a log storage control method.

  Information processing apparatuses such as computers and communication apparatuses are equipped with various applications and software such as firmware. Also, in the information processing apparatus, various information such as software operation status and error occurrence status is stored as a log for cause analysis when a malfunction event occurs. For example, as a technique for storing a log, there is a technique for fixing a log to be stored in a nonvolatile memory and a log to be stored in a volatile memory in accordance with the content and importance of the log.

  In addition, there is a technique that enables a program state to be restored to a state just before the abnormal termination when the program is abnormally terminated and restarted. In this technique, when the utilization rate of a CPU (Central Processing Unit) becomes low, information in a volatile memory used by a program is copied to a nonvolatile memory as memory copy information.

  Also, there is a technology that enables efficient accumulation of useful information for specifying the cause even if the CPU runs away or stops. In this technique, when a predetermined state that would interfere with the processing of the CPU is detected, a predetermined signal that the CPU exchanges with the peripheral circuit is acquired and recorded in the nonvolatile memory.

JP 2007-122357 A JP 2004-185318 A

  By the way, when a malfunction event occurs in the information processing apparatus, the apparatus is often restarted after that, and in such a case, only the log stored in the nonvolatile memory can be obtained. In many cases, the cause analysis of the failure event requires a log stored in the volatile memory, and the conventional technology for saving the log cannot save a log effective for the cause analysis of the failure event.

  Although a method of storing all logs in a nonvolatile memory is also conceivable, there are physical capacity problems and processing capacity problems, which are not realistic solutions. Also, in the technology of copying information in the volatile memory to the non-volatile memory when the CPU usage rate becomes low, if the power of the device is suddenly turned off while the CPU load is high, the technology is volatile. Information in memory cannot be stored in non-volatile memory. Also, with this technology, all the information in the volatile memory is copied to the nonvolatile memory, so that logs effective for analysis cannot be centrally stored. In addition, when a predetermined state that hinders the processing of the CPU is detected, the technology for recording a predetermined signal that the CPU exchanges with the peripheral circuit in the nonvolatile memory is such that the power supply of the apparatus is turned on in a state other than the predetermined state. If it turns off suddenly, the log cannot be saved. Also, with this technique, only a log relating to a predetermined signal can be saved.

  The disclosed technology has been made in view of the above, and an object thereof is to provide an information processing apparatus, a log storage control program, and a log storage control method capable of storing a log effective for cause analysis of a failure event.

  In one aspect, an information processing apparatus disclosed in the present application includes a volatile storage unit, a nonvolatile storage unit, a detection unit, a determination unit, and a storage unit. The detection unit detects a load state of the software. When a software log is generated, the determination unit determines whether the log according to the first type is stored in the nonvolatile storage unit in the first load state according to the type of the log and the load state detected by the detection unit. It is determined that the log relating to the second type is stored in the volatile storage unit. Further, the determination unit determines that the log according to the first type and the log according to the second type are stored in the nonvolatile storage unit in the second load state where the load is higher than the first load state. . The storage unit stores the log in either the volatile storage unit or the nonvolatile storage unit according to the determination result of the determination unit.

  According to one aspect of the log storage control method disclosed in the present application, it is possible to save a log effective for cause analysis of a failure event.

FIG. 1 is a diagram illustrating an example of the configuration of the information processing apparatus. FIG. 2 is a diagram illustrating an example of the configuration of the image distribution system. FIG. 3 is a diagram illustrating an example of the configuration of the CPU unit. FIG. 4 is a diagram illustrating an example of a data configuration of log storage destination information. FIG. 5 is a diagram illustrating an example of a functional configuration of a CPU that executes a log storage control program. FIG. 6 is a diagram illustrating an example of the relationship between the number of messages waiting to be processed by a program and the load state. FIG. 7 is a diagram illustrating a procedure of log storage control processing. FIG. 8 is a diagram illustrating an example of the relationship between the CPU load and the software load state. FIG. 9 is a diagram illustrating an example of a relationship between a CPU load factor and a log storage destination. FIG. 10 is a diagram illustrating an example of the relationship between the memory usage rate and the software load state. FIG. 11 is a diagram illustrating an example of the relationship between the number of logs generated during a predetermined period and the software load state. FIG. 12 is a diagram illustrating an example of a relationship between a change in CPU load and an increase rate of the load. FIG. 13 is a diagram illustrating an example of a CPU unit provided with a temperature sensor. FIG. 14 is a diagram illustrating a computer that executes a log storage control program.

  Hereinafter, embodiments of an information processing apparatus, a log storage control program, and a log storage control method disclosed in the present application will be described in detail with reference to the drawings. Each embodiment does not limit the disclosed technology.

  An information processing apparatus according to the first embodiment will be described. FIG. 1 is a diagram illustrating an example of the configuration of the information processing apparatus. As illustrated in FIG. 1, the information processing apparatus 10 includes a detection unit 21, a determination unit 22, a storage unit 23, a volatile storage unit 24, and a nonvolatile storage unit 25. The information processing device 10 is a device on which at least software is installed, and is, for example, a computer, a communication device, or the like. The volatile storage unit 24 is a non-volatile semiconductor memory that can rewrite data, such as a flash memory or a non-volatile static random access memory (NVSRAM). The nonvolatile storage unit 25 is, for example, a nonvolatile semiconductor memory that can rewrite data, such as a flash memory or an NVSRAM.

  The detection unit 21 detects the load state of the software 26 being executed in the information processing apparatus 10. The software 26 is, for example, a program such as an application or firmware executed by a CPU (Central Processing Unit) included in the information processing apparatus 10. Further, the software 26 generates various information such as an operation state and an error occurrence state as a log. This load state is, for example, a steady state in which software processing is steadily operating within an assumed range, an overload state in which the software process is operating beyond the assumed range, or a temporary process that greatly exceeds the assumed range. There is an abnormal condition that causes a complete stop or an uncontrollable complete stop. The detection unit 21 loads the arithmetic unit used for executing the software 26, the rate of increase of the load, the number of queued messages waiting for processing by the software 26, the number of logs generated within a predetermined period, and the use of the memory used for executing the software 26 The load state is detected based on the rate, the memory temperature, and the like.

  When a log of the software 26 is generated, the determination unit 22 determines a log storage destination according to the type of the log and the load state detected by the detection unit 21. The determination unit 22 determines that, in the first load state, the log related to the first type is stored in the nonvolatile storage unit 25 and the log related to the second type is stored in the volatile storage unit 24. In addition, in the second load state where the load is higher than the first load state, the determination unit 22 indicates that the log related to the first type and the log related to the second type are stored in the nonvolatile storage unit 25. judge. For example, in the steady state, the determination unit 22 determines that a log related to an error such as a memory protection violation log is the storage destination of the nonvolatile storage unit 25, and determines that a processing history log such as a trace log is a storage destination in the volatile storage unit 24. To do. Further, the determination unit 22 determines that the log relating to the error and the log of the processing history are stored in the nonvolatile storage unit 25 in the transient state or the abnormal state. The storage unit 23 stores the log in either the volatile storage unit 24 or the nonvolatile storage unit 25 according to the determination result of the determination unit 22.

  As described above, in the first load state, the information processing apparatus 10 stores the log according to the first type in the nonvolatile storage unit 25 and the log according to the second type in the volatile storage unit 24. Store. Further, the information processing apparatus 10 stores the log related to the first type and the log related to the second type in the nonvolatile storage unit 25 in the second load state where the load is higher than the first load state. That is, the information processing apparatus 10 stores many types of logs in the nonvolatile storage unit 25 when the load is high. Therefore, according to the information processing apparatus 10 according to the present embodiment, it is possible to save a log effective for the cause analysis of the failure event. That is, if the information processing apparatus 10 transitions from a steady state to a transient state or an abnormal state and the power is turned off, the log stored in the volatile storage unit 24 is deleted. On the other hand, the log stored in the nonvolatile storage unit 25 is saved. For example, in addition to a log relating to an error, a log of a processing history in an overload state or an abnormal state is stored. This facilitates the cause analysis of the failure event.

  Next, Example 2 will be described. In the second embodiment, an image distribution system 30 that distributes moving images and the like via a communication line such as a network will be described. FIG. 2 is a diagram illustrating an example of the configuration of the image distribution system. The image distribution system 30 has an interface slot 31 in which various units 33 can be mounted according to the functions to be mounted. The image distribution system 30 also has a power supply unit 32 below the interface slot 31. The power supply unit 32 supplies power to each unit mounted in the interface slot 31.

  In the image distribution system 30, as various units 33, two CPU units 33a, one clock unit 33b, and three image units 33c are mounted. The CPU unit 33a controls the entire apparatus. The clock unit 33 b provides a clock indicating the operation timing of each unit 33. The image unit 33c performs image transmission processing of a moving image to be distributed. Each unit 33 is an example of the information processing apparatus 10 described in the first embodiment.

  Here, the configuration of the CPU unit 33a will be described. FIG. 3 is a diagram illustrating an example of the configuration of the CPU unit. The CPU unit 33a includes a program storage memory 40, a nonvolatile memory 41, a volatile memory 42, a storage location information storage memory 43, a CPU 44, a RAM 45, a serial port 46, and a LAN (Local Area Network) port 47. And have. The program storage memory 40, the nonvolatile memory 41, the volatile memory 42, the storage location information storage memory 43, the CPU 44, the RAM 45, the serial port 46, and the LAN port 47 are connected by a bus 48. Exchange information.

  The program storage memory 40 is a non-volatile semiconductor memory element such as a ROM (Read Only Memory) or a flash memory. Instead of the program storage memory 40, a nonvolatile storage device such as a hard disk may be used. The program storage memory 40 stores various software programs 60 such as an OS (operating system) and firmware. The various software programs 60, when executed by the CPU 44, generate various information such as software operating states and error occurrence statuses as logs in order to analyze the cause when a malfunction event occurs. The generated log includes type information indicating a log type such as a code indicating an operation state and an error code. The program storage memory 40 stores a log storage control program 61 that controls the storage location of the software log.

  The non-volatile memory 41 is a non-volatile semiconductor memory capable of rewriting data, such as a flash memory or NVSRAM. Instead of the nonvolatile memory 41, a nonvolatile storage device that can rewrite data, such as a hard disk or an optical disk, may be used. The non-volatile memory 41 is used as a storage area for logs that need to be retained even when the image distribution system 30 is powered off, and retains the stored logs.

  The volatile memory 42 is a volatile semiconductor memory capable of rewriting data such as a RAM (Random Access Memory). The nonvolatile memory 41 is used as a storage area for temporarily storing a log, and temporarily holds the log.

  The storage destination information storage memory 43 is a non-volatile semiconductor memory capable of rewriting data, such as a flash memory or NVSRAM. Instead of the nonvolatile memory, a nonvolatile storage device that can rewrite data such as a hard disk or an optical disk may be used. In the storage location information storage memory 43, for each software, information indicating whether the log generated by the software is stored in either the volatile memory 42 or the nonvolatile memory 41 (hereinafter also referred to as “log storage location information”). .) Is stored. This log storage location information stores whether the log is stored in either the volatile memory 42 or the nonvolatile memory 41 for each type of software log and each software load state.

  FIG. 4 is a diagram illustrating an example of a data configuration of log storage destination information. In the log storage destination information, for each log type, either the volatile memory 42 or the non-volatile memory 41 is stored in each of the software load states of “normal state”, “transient state”, and “abnormal state”. Is memorized. The “normal state” is a state in which the apparatus is operated within an assumed range, for example. The “abnormal state” is, for example, a state in which there is a problem with the device, the operation is hindered, and there is a high possibility of the device being down or rebooted. The “excessive state” is, for example, a state that does not reach an abnormal state but is operated beyond an assumed range.

  By the way, the log has different effectiveness for the cause analysis of the failure event depending on the type. Therefore, in this embodiment, the importance of how effective the cause analysis is for each type of log is defined as a three-level log rank of “high”, “medium”, and “low”. For example, a fatal error log defines the log rank as “high”. A log effective for cause analysis of a specific situation has a log rank of “medium”. The trace and history log for debugging defines the log rank as “low”. In the log storage location information according to the present embodiment, the storage location is set so that the type of log with a high log rank is stored in the nonvolatile memory 41 in any load state. Further, the log storage destination information is set such that a log of a low log rank is stored in the volatile memory 42 when the load state is low, and is stored in the nonvolatile memory 41 when the load state is high. Is done.

  In the example of FIG. 4, the log rank of the memory protection violation log is “high”, and the nonvolatile memory 41 is set as the storage destination in any of the normal state, the excessive state, and the abnormal state. In the example of FIG. 4, the process queue information and process information have a log rank of “medium”, the volatile memory 42 is set as the storage destination in the normal state, and the nonvolatile memory 41 in the excessive state and abnormal state. Set as storage location. In the example of FIG. 4, the log rank of the program processing trace log is “low”, the volatile memory 42 is set as the storage destination in the normal state and the excessive state, and the nonvolatile memory 41 is stored in the abnormal state. Is set.

  Returning to the description of FIG. 3, the RAM 45 is a volatile semiconductor memory capable of rewriting data. The RAM 45 is used as a work area when the CPU 44 executes various programs.

  The CPU 44 reads and executes various programs 60 and a log storage control program 61 stored in the program storage memory 40. When the program 60 is executed, it generates a log such as an operation state and an error occurrence state. The log storage control program 61 controls whether the log is stored in the volatile memory 42 or the nonvolatile memory 41 in accordance with the type of log generated by the program 60 and the load state.

  The serial port 46 and the LAN port 47 are connected to a terminal device 50 such as a personal computer via a communication cable 49. For example, the administrator of the image distribution system 30 transmits a log collection command from the terminal device 50 connected to the serial port 46 or the LAN port 47 when performing the cause analysis of the malfunction event. When the CPU 44 receives the log collection command, the log stored in the volatile memory 42 and the nonvolatile memory 41 is filed and transferred to the terminal device 50. The administrator analyzes the cause of the failure event by analyzing the transferred log file. The other units 33b and c have the same configuration as the unit 33a.

  Next, a functional configuration when the log storage control program 61 is executed will be described with reference to FIG. FIG. 5 is a diagram illustrating an example of a functional configuration of a CPU that executes a log storage control program. In the example of FIG. 5, in addition to the log storage control program 61, as the program 60, a fault detection program 60 a, various service programs 60 b, an image processing program 60 c, and a control program 60 d such as an OS are CPU 44. Has been executed by. The programs 60a, 60b, and 60c exchange messages by inter-process communication. The control program 60d manages interprocess communication between the programs 60a, 60b, and 60c. The RAM 45 is provided with a message queue area that stores messages transmitted for each of the programs 60a, 60b, and 60c, and manages messages waiting to be processed in a FIFO (First In First Out) format.

  The log storage control program 61 includes a monitoring unit 70, a determination unit 71, and a storage unit 72.

  The monitoring unit 70 monitors the number of messages waiting to be processed for each of the programs 60a, 60b, and 60c as the load state of each software. For example, the monitoring unit 70 monitors the number of messages waiting to be processed for each of the programs 60a, 60b, and 60c using a message management interface such as an API (Application Programming Interface) provided by the control program 60d.

  Logs generated by the programs 60a, 60b, and 60c are transmitted to the determination unit 71. When the log is transmitted, the determination unit 71 refers to the number of messages waiting for processing for each of the programs 60a, 60b, and 60c monitored by the monitoring unit 70, and assumes that the load state increases as the number of messages waiting for processing increases. Identify load conditions. For example, when the number of messages waiting to be processed by any program 60 is within a predetermined allowable range, the determination unit 71 specifies that the load state is the normal state. In addition, the determination unit 71 specifies that the load state is a transient state when the number of messages waiting to be processed by any one of the programs 60 is greater than the allowable range and within a predetermined threshold value that is greater than the allowable range. Further, the determination unit 71 specifies that the load state is an abnormal state when the number of messages waiting to be processed by any program 60 is greater than the threshold value. FIG. 6 is a diagram illustrating an example of the relationship between the number of messages waiting to be processed by a program and the load state. In the example of FIG. 6, the allowable range is up to four messages waiting to be processed. As shown in FIG. 6, when the number of messages waiting for processing of the programs 60a and 60c is two and the number of messages waiting for processing of the program 60b is three, the load state is specified as the normal state. In addition, when the number of messages waiting for processing of the programs 60a and 60c is two and the number of messages waiting for processing of the program 60b is increased to six, the load state is identified as an excessive state. Furthermore, when the number of messages waiting for processing of the programs 60a and 60c is two and the number of messages waiting for processing of the program 60b is larger than the threshold value, the load state is identified as an abnormal state. The threshold value may be determined based on the allowable range, for example, twice the allowable range. Further, the load state may be specified as an abnormal state when the number of messages waiting to be processed overflows the buffer of the message queue area.

  The determination unit 71 refers to the log storage destination information of the program that generated the log among the log storage destination information of each program stored in the nonvolatile memory 41, and the input log is stored in the volatile memory 42 or the nonvolatile memory. 41 to determine whether to store the data. For example, the determination unit 71 determines the storage location of the input log by reading the storage location corresponding to the type indicated by the type information included in the log input from the log storage location information and the specified load state. .

  Thereby, for example, when the log storage destination information of the program that generated the log has the contents shown in FIG. 4, the memory protection violation log is stored in the nonvolatile memory 41 in any of the normal state, the excessive state, and the abnormal state. It is determined. Further, the process queue information and the process information are determined as the storage destination in the volatile memory 42 in the normal state, and the nonvolatile memory 41 is determined as the storage destination in the excessive state and the abnormal state. In the trace log of the program processing, the volatile memory 42 is determined as the storage destination in the normal state and the excessive state, and the nonvolatile memory 41 is determined as the storage destination in the abnormal state.

  The storage unit 72 stores the transmitted log in either the volatile memory 42 or the nonvolatile memory 41 in accordance with the determination result of the determination unit 71. The storage unit 72 stores the new log by overwriting the oldest log when the storage areas of the logs stored in the volatile memory 42 and the non-volatile memory 41 respectively reach predetermined amounts. Accordingly, it is possible to prevent the log capacity from exceeding the capacity of the volatile memory 42 and the non-volatile memory 41 and the log cannot be stored.

  As described above, when the load state is high, by storing many types of logs in the nonvolatile memory 41, it is possible to save logs effective for the cause analysis of the failure event.

  When the load state is low, a log with a high log rank is stored in the non-volatile memory 41, and a log with a low log rank is stored in the volatile memory 42, so that only the log with high importance is stored in the non-volatile memory 41. Is remembered. Thereby, since only the log with a high log rank is stored, the non-volatile memory 41 can save many logs with a high log rank. Further, when the image distribution system 30 is not frozen or restarted, it is possible to acquire logs from a high log rank to a low log by reading the logs from the nonvolatile memory 41 and the volatile memory 42.

  Next, the flow of processing of the CPU unit 33a according to the present embodiment will be described. FIG. 7 is a flowchart showing the procedure of the log storage control process. This log storage control process is always executed after the CPU unit 33a is activated.

  As illustrated in FIG. 7, the determination unit 71 determines whether or not a log has been received (step S101). If the log has not been received (No at Step S101), the determination unit 71 determines whether an end of processing has been instructed from the OS or the like (Step S102). When the process end is instructed (Yes in step S102), the determination unit 71 ends the process. On the other hand, when the process end is not instructed (No at Step S102), the determination unit 71 proceeds to Step S101.

  When receiving the log (Yes at Step S101), the determination unit 71 identifies the load state with reference to the number of messages waiting for processing for each of the programs 60a, 60b, and 60c monitored by the monitoring unit 70 (Step S103). Then, the determination unit 71 determines the log storage destination by reading the log storage destination according to the type of log input from the log storage destination information and the specified load state (step S104). The storage unit 72 determines whether or not the log storage destination is the nonvolatile memory 41 (step S105). When the storage destination of the log is the nonvolatile memory 41 (Yes at Step S105), the storage unit 72 stores the log in the nonvolatile memory 41 (Step S106), and then proceeds to Step S101. On the other hand, if the storage location of the log is not the nonvolatile memory 41 (No at Step S105), the storage unit 72 stores the log in the volatile memory 42 (Step S107), and then proceeds to Step S101.

  Thus, for example, when a memory protection violation log is detected, the memory protection violation log is stored in the nonvolatile memory 41 regardless of whether the load state is a normal state, an excessive state, or an abnormal state. On the other hand, when the process queue information and the process information log are detected and the load state is the normal state, the process queue information and the process information log are stored in the volatile memory 42. Further, when the load state is an excessive state or an abnormal state, the process queue information and the process information log are stored in the nonvolatile memory 41. On the other hand, when the load state is a normal state and an excessive state when the trace log is detected, the trace log is stored in the volatile memory 42. When the load state is an abnormal state, the trace log is stored in the nonvolatile memory 41.

  As described above, the CPU unit 33a according to the present embodiment detects a software load state. When the load state is the normal state, the CPU unit 33a according to the present embodiment stores the memory protection violation log in the nonvolatile memory 41, and the process queue information, the process information log, and the trace log are the volatile memory 42. To store it in Further, the CPU unit 33a according to the present embodiment stores the process queue information and the process information log and the memory protection violation log in the nonvolatile memory 41 when the load state is excessive, and the trace log is stored in the volatile memory 42. To store it in Further, the CPU unit 33a according to the present embodiment determines that the process queue information, the process information log, the memory protection violation log, and the trace log are stored in the nonvolatile memory 41 when the load state is abnormal. The CPU unit 33a according to the present embodiment stores the log in either the volatile memory 42 or the nonvolatile memory 41 according to the determination result. In this way, in the normal state with a high load state, by storing the memory protection violation log with a high log rank in the nonvolatile memory 41, a large number of logs with a high log rank can be stored in the nonvolatile memory 41. Further, when the CPU unit 33a is not frozen or restarted, it is possible to acquire logs from a high log rank to a low log by reading the logs from the nonvolatile memory 41 and the volatile memory 42. Further, as the load state increases, a log with a lower log rank is stored in the nonvolatile memory 41, so that even when the CPU unit 33a is frozen or restarted, many types of logs can be saved. Analysis becomes easier. In particular, a malfunction event is likely to occur when the load state is high. For this reason, when the load state is high, the number of log types stored in the non-volatile memory 41 for many types of logs is increased, and many types of logs are saved, thereby facilitating the analysis of the cause of the failure event.

  Further, as described above, the CPU unit 33 a according to the present embodiment stores either the volatile memory 42 or the nonvolatile memory 41 in the storage destination information storage memory 43 for each log type and each software load state. Information indicating whether to store the log is stored. Then, when a software log is generated, the CPU unit 33a according to the present embodiment refers to information stored in the storage location information storage memory 43, and the log is stored in either the volatile memory 42 or the nonvolatile memory 41. It is determined whether or not to store. As described above, by storing the storage destination for each log type and each software load state, it is possible to finely control the log storage destination for each log type and each software load state. Further, by changing the information stored in the storage destination information storage memory 43, the log storage destination can be changed for each log type and each software load state.

  Although the embodiments related to the disclosed apparatus have been described so far, the disclosed technology may be implemented in various different forms other than the above-described embodiments. Therefore, another embodiment included in the present invention will be described below.

  For example, in the above embodiment, the monitoring unit 70 monitors the number of messages waiting to be processed for each program 60 as the software load state, and the determination unit 71 detects that the load is higher as the number of messages is larger. However, the disclosed apparatus is not limited to this. For example, the CPU load of the CPU 44 may be obtained, and it may be detected that the load is higher as the CPU load is higher. FIG. 8 is a diagram illustrating an example of the relationship between the CPU load and the software load state. In the example of FIG. 8, the horizontal axis indicates time, and the vertical axis indicates CPU load. In the example of FIG. 8, a range where the CPU load is less than 70% is defined as a normal state, a range where the CPU load is 70% or more and less than 90% is defined as an excessive state, and a range where the CPU load is 90% or more is defined as an abnormal state. It has established. In this case, the monitoring unit 70 monitors the CPU load of the CPU 44. The determination unit 71 detects a load state from the CPU load. FIG. 9 is a diagram illustrating an example of a relationship between a CPU load factor and a log storage destination. In the example of FIG. 9, logs generated by the programs are indicated by a to e. Logs a to e are assumed to have different types. In the example of FIG. 9, when the CPU load is in a normal state of less than 70%, the logs a and b are stored in the nonvolatile memory 41 and the logs c to e are stored in the volatile memory 42. In the example of FIG. 9, when the CPU load is in an excessive state in the range of 70% or more and less than 90%, the logs a to c are stored in the nonvolatile memory 41 and the logs d and e are stored in the volatile memory 42. Store. In the example of FIG. 9, the logs a to e are stored in the nonvolatile memory 41 when the CPU load is in an abnormal state in the range of 90% or more. As described above, when the load state is high, by storing many types of logs in the nonvolatile memory 41, it is possible to save logs effective for the cause analysis of the failure event.

  In addition, the RAM 45 has a high memory usage rate when a problem such as a memory leak occurs or when many programs are executed and the processing load is high. Therefore, for example, the memory usage rate of the RAM 45 may be obtained as the software load state, and the higher the memory usage rate, the higher the load may be detected. FIG. 10 is a diagram illustrating an example of the relationship between the memory usage rate and the software load state. In the example of FIG. 10, a range where the memory usage rate is less than 50% is defined as a normal state, a range where the memory usage rate is 50% or more and less than 80% is defined as an excessive state, and a range where the memory usage rate is 80% or more. It is defined as an abnormal condition. In this case, the monitoring unit 70 monitors the memory usage rate of the RAM 45. The determination unit 71 detects the load state from the memory usage rate.

  Further, the program 60 increases the number of logs generated when a problem occurs. Therefore, for example, the number of logs generated during a predetermined period may be obtained as the software load state, and it may be detected that the load is higher as the number of logs is higher. FIG. 11 is a diagram illustrating an example of the relationship between the number of logs generated during a predetermined period and the software load state. In the example of FIG. 11, the horizontal axis indicates time, and the vertical axis indicates the number of logs generated every 10 minutes. In the example of FIG. 11, a range where the number of logs generated in 10 minutes is less than 50 is defined as a normal state, and a range where the number of logs generated in 10 minutes is 50 or more and less than 150 is defined as an excessive state. A range in which the number of generated logs is 150 or more is defined as an abnormal state. In this case, the monitoring unit 70 monitors the number of logs generated during a predetermined period. The determination unit 71 detects the load state from the number of generated logs.

  In the information processing apparatus such as the CPU unit 33a, the CPU load of the CPU 44 increases rapidly when the load increases. Therefore, for example, an increase rate of the CPU load of the CPU 44 may be obtained as the software load state, and it may be detected that the load is higher as the increase rate is higher. FIG. 12 is a diagram illustrating an example of a change in CPU load and an increase rate of the load. In the example of FIG. 12, the horizontal axis indicates time, and the vertical axis indicates CPU load. In the example of FIG. 12, the CPU load increase rate is obtained by calculating B / A from the predetermined period A and the CPU load increase amount B during the predetermined period A. In this case, the monitoring unit 70 monitors the increase rate of the CPU load of the CPU 44 every predetermined period. The determination unit 71 detects the load state from the increasing rate of the CPU load.

  Further, the RAM 45 has a high temperature when the internal data is frequently accessed and the processing load is high. Therefore, for example, a temperature sensor 90 that detects the temperature of the RAM 45 is provided, the monitoring unit 70 monitors the temperature of the RAM 45 as a software load state, and the determination unit 71 detects that the load is higher as the temperature is higher. Also good. FIG. 13 is a diagram illustrating an example of a CPU unit provided with a temperature sensor. In FIG. 13, the same parts as those of the CPU unit of the second embodiment are denoted by the same reference numerals as those in FIG. 3, and the description thereof is omitted. In the example of FIG. 13, a temperature sensor 90 is provided in the RAM 45.

  Further, for example, as the software load state, the load of the CPU 44, the increase rate of the load, the queue number of messages waiting to be processed by the software, the number of logs generated within a predetermined period, the memory usage rate of the RAM 45, the memory temperature of the RAM 45 Any one or more may be detected. In this case, the load state as a whole may be specified from the plurality of detected load states. For example, the highest load among the plurality of detected load states may be specified as the entire load state. Further, for example, the most load state may be specified as the entire load state. Thus, by detecting the load state from various things such as the load of the CPU 44 and the increase rate of the load, changes in the load state due to various trouble events can be grasped.

  In the above-described embodiment, the software load state is divided into two stages of “first load state” and “second load state”, or “normal state”, “transient state”, and “abnormal state”. Although the case where it discriminate | determined in the stage was illustrated, the apparatus of an indication is not limited to this. For example, the load state may be discriminated into four or more stages.

  Further, in the above-described second embodiment, the case where the log importance is set to the three-level log rank of “high”, “medium”, and “low” is illustrated, but the disclosed apparatus is not limited to this. For example, the log rank may be determined in a state of two levels or four or more levels.

  In addition, the processing at each step of each processing described in the second embodiment can be arbitrarily finely divided or combined according to various loads and usage conditions. Also, the steps can be omitted. For example, when a plurality of detections such as the load of the CPU 44 and the increase rate of the load are performed as the software load state, a step of determining the load state from each detection step and each detection result is added.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific state of distribution / integration of each device is not limited to the one shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, in the second embodiment, the case where the nonvolatile storage unit for storing the log and the nonvolatile storage unit for storing the log storage destination information are provided separately from the nonvolatile memory 41 and the storage destination information storage memory 43 are illustrated. However, the disclosed apparatus is not limited to this. For example, the log and the log storage location information may be stored in one nonvolatile storage unit. In the second embodiment, the case where the volatile storage unit that stores the log and the volatile storage unit that the CPU 44 uses as the work area is provided separately from the volatile memory 42 and the RAM 45 is described. It is not limited to this. For example, one volatile storage unit may be used as a log storage area and the CPU 44 as a work area.

[Log storage control program]
Various processes of the information processing apparatus described in the above embodiments can also be realized by executing a program prepared in advance on a computer system such as a personal computer or a workstation. In the following, an example of a computer that executes a log storage control program having the same function as that of the information processing apparatus described in the above embodiment will be described with reference to FIG. FIG. 14 is a diagram illustrating a computer that executes a log storage control program.

  As illustrated in FIG. 14, the computer 300 includes a central processing unit (CPU) 310, a read only memory (ROM) 320, a hard disk drive (HDD) 330, and a random access memory (RAM) 340. These units 300 to 340 are connected via a bus 400.

  The ROM 320 has the same functions as those of the detection unit 21, the determination unit 22, the storage unit 23, or the monitoring unit 70, the determination unit 71, and the storage unit 72 described in the second embodiment. A log storage control program 320a to be exhibited is stored in advance. That is, the ROM 320 stores a log storage control program 320a as shown in FIG. Note that the log storage control program 320a may be separated as appropriate.

  Then, the CPU 310 reads the log storage control program 320a from the ROM 320 and executes it.

  The HDD 330 is provided with log storage location information 330a. The log storage location information 330a corresponds to the log storage location information stored in the storage location information storage memory 43 shown in FIG.

  Then, the CPU 310 reads the log storage location information 330a and stores it in the RAM 340. The CPU 310 executes the log storage control program 320a using the association information stored in the RAM 340. Each data stored in the RAM 340 does not always need to be stored in the RAM 340, and only the data necessary for the process may be stored in the RAM 340.

  Note that the log storage control program 320a is not necessarily stored in the HDD 330 from the beginning.

  For example, the program is stored in a “portable physical medium” such as a flexible disk (FD), a CD-ROM, a DVD disk, a magneto-optical disk, or an IC card inserted into the computer 300. Then, the computer 300 may read and execute the program from these.

  Furthermore, the program is stored in “another computer (or server)” connected to the computer 300 via a public line, the Internet, a LAN, a WAN, or the like. Then, the computer 300 may read and execute the program from these.

  The following supplementary notes are further disclosed regarding the embodiment described above and its modifications.

(Supplementary note 1) a volatile storage unit;
A non-volatile storage unit;
A detection unit for detecting the load state of the software;
When the log of the software is generated, the log according to the first type is stored in the nonvolatile storage unit in the first load state according to the type of the log and the load state detected by the detection unit. Then, it is determined that the log according to the second type is stored in the volatile storage unit, and in the second load state where the load is higher than the first load state, the log according to the first type and the first A determination unit that determines to store the log according to the second type in the nonvolatile storage unit;
An information processing apparatus comprising: a storage unit that stores the log in either the volatile storage unit or the non-volatile storage unit according to a determination result of the determination unit.

(Supplementary Note 2) A storage location information storage unit storing information indicating whether to store a log in either the volatile storage unit or the nonvolatile storage unit for each type of the log and each load state of the software In addition,
When the log of the software is generated, the determination unit refers to information stored in the storage location information storage unit, and stores the log in either the volatile storage unit or the nonvolatile storage unit The information processing apparatus according to appendix 1, characterized by:

(Additional remark 3) The said detection part is the load of the calculating part used for execution of the said software, the increase rate of the said load, the queue number of the messages waiting for the process by the said software, the number of logs generated within a predetermined period, execution of the said software The information processing apparatus according to appendix 1 or 2, wherein any one or more of a memory usage rate and a memory temperature used in the processing are detected.

(Appendix 4)
Detects software load conditions,
When the log of the software is generated, the log according to the first type is stored in the nonvolatile storage unit in the first load state according to the type of the log and the detected load state, and the second It is determined that the log related to the type is stored in the volatile storage unit, and in the second load state where the load is higher than the first load state, the log related to the first type and the log related to the second type Is stored in the nonvolatile storage unit,
A log storage control program for executing each process of storing the log in either the volatile storage unit or the non-volatile storage unit according to a determination result.

(Additional remark 5) When the said log of software generate | occur | produces, the said determination process stores the said volatile memory | storage part or each for every kind of said log memorize | stored in the storage location information memory | storage part and the said software load state, or It is determined whether to store the log in either the volatile storage unit or the non-volatile storage unit with reference to information indicating whether the log is stored in any of the non-volatile storage unit The log storage control program according to appendix 4.

(Supplementary Note 6) The processing to be detected includes the load of the arithmetic unit used for executing the software, the rate of increase of the load, the number of queued messages waiting to be processed by the software, the number of logs generated within a predetermined period, 6. The log storage control program according to appendix 4 or 5, wherein one or more of a memory usage rate used for execution and the memory temperature are detected.

(Appendix 7) The computer
Detects software load conditions,
When the log of the software is generated, the log according to the first type is stored in the nonvolatile storage unit in the first load state according to the type of the log and the detected load state, and the second It is determined that the log related to the type is stored in the volatile storage unit, and in the second load state where the load is higher than the first load state, the log related to the first type and the log related to the second type Is stored in the nonvolatile storage unit,
A log storage control method comprising: executing each process of storing the log in either the volatile storage unit or the non-volatile storage unit according to a determination result.

(Supplementary Note 8) When the software log is generated, the determination process is performed for each type of log and each load state of the software stored in the storage location information storage unit. It is determined whether to store the log in either the volatile storage unit or the non-volatile storage unit with reference to information indicating whether the log is stored in any of the non-volatile storage unit The log storage control method according to appendix 7.

(Supplementary note 9) The processing to be detected includes the load of the arithmetic unit used for executing the software, the rate of increase of the load, the number of queued messages waiting to be processed by the software, the number of logs generated within a predetermined period, 9. The log storage control method according to appendix 7 or 8, wherein any one or more of a memory usage rate used for execution and the memory temperature are detected.

DESCRIPTION OF SYMBOLS 10 Information processing apparatus 21 Detection part 22 Judgment part 23 Storage part 24 Volatile memory part 25 Nonvolatile memory part 26 Software 30 Image distribution system 33a CPU unit 40 Program storage memory 41 Nonvolatile memory 42 Volatile memory 43 Storage location information storage memory 61 log storage control program 70 monitoring unit 71 determination unit 72 storage unit 90 temperature sensor

Claims (5)

A volatile storage unit;
A non-volatile storage unit;
A detection unit for detecting the load state of the software;
When the log of the software is generated, the log according to the first type is stored in the nonvolatile storage unit in the first load state according to the type of the log and the load state detected by the detection unit. Then, it is determined that the log according to the second type is stored in the volatile storage unit, and in the second load state where the load is higher than the first load state, the log according to the first type and the first A determination unit that determines to store the log according to the second type in the nonvolatile storage unit;
An information processing apparatus comprising: a storage unit that stores the log in either the volatile storage unit or the non-volatile storage unit according to a determination result of the determination unit. A storage location information storage unit that stores information indicating whether to store the log in the volatile storage unit or the non-volatile storage unit for each type of log and for each load state of the software;
When the log of the software is generated, the determination unit refers to information stored in the storage location information storage unit, and stores the log in either the volatile storage unit or the nonvolatile storage unit The information processing apparatus according to claim 1, further comprising:   The detection unit includes a load on the arithmetic unit used for executing the software, an increase rate of the load, a queue number of messages waiting to be processed by the software, a number of logs generated within a predetermined period, and a memory used for executing the software The information processing apparatus according to claim 1, wherein one or more of a rate and a memory temperature are detected. On the computer,
Detects software load conditions,
When the log of the software is generated, the log according to the first type is stored in the nonvolatile storage unit in the first load state according to the type of the log and the detected load state, and the second It is determined that the log related to the type is stored in the volatile storage unit, and in the second load state where the load is higher than the first load state, the log related to the first type and the log related to the second type Is stored in the nonvolatile storage unit,
A log storage control program for executing each process of storing the log in either the volatile storage unit or the non-volatile storage unit according to a determination result. Computer
Detects software load conditions,
When the log of the software is generated, the log according to the first type is stored in the nonvolatile storage unit in the first load state according to the type of the log and the detected load state, and the second It is determined that the log related to the type is stored in the volatile storage unit, and in the second load state where the load is higher than the first load state, the log related to the first type and the log related to the second type Is stored in the nonvolatile storage unit,
A log storage control method comprising: executing each process of storing the log in either the volatile storage unit or the non-volatile storage unit according to a determination result.

Images