Method for controlling a policy

• US 7,571,351 B2
• Filed: 02/15/2006
• Issued: 08/04/2009
• Est. Priority Date: 11/28/2005
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Abe, Takashi, Fujino, Shuji, Miyazaki, Daisuke
• Primary Examiner(s): Maskulinski; Michael C
• Application Number:
  US11/354,116
• Publication Number:
  US 20070174708A1
• Time in Patent Office: 1,266 Days
• Field of Search: None
• US Class Current: 714/25
• CPC Class Codes: G06F 11/2252 using fault dictionaries

Abstract

The method includes confirmation policy processing for acquiring confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information when a confirmation item value of a confirmation object satisfies a condition in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition; execution policy processing for executing contents of a measure against the presumed cause read from execution policy information; evaluation policy processing for acquiring evaluation item information to evaluate a measure result, judging an effect of the executed measure based on whether an evaluation item value of an evaluation object is in a stable operation range, and storing a judging result; and controlling an autonomous policy cycle for conducting the confirmation, execution and evaluation policy processing.

Claims

1. A policy control method responsive to occurrence of a fault or abnormality in a computer to distinguish a fault occurrence place and a cause and take a measure by using policies, the policy control method comprising:

conducting confirmation policy processing for acquiring confirmation item information from a storage apparatus, the confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading out presumed cause information from the storage apparatus when a confirmation item value of a confirmation subject object satisfies a condition in the confirmation item information, the presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition, and judging whether a probability of the presumed cause is greater than a probability of a cause for automatic execution;

conducting execution policy processing for causing a processing apparatus to execute contents of a measure read out from execution policy information, the execution policy information indicating a measure against the acquired presumed cause;

conducting evaluation policy processing for acquiring evaluation item information from the storage apparatus to evaluate a result of the measure, judging an effect of the executed measure on the basis of whether an evaluation item value of an evaluation subject object is in a stable operation range indicated by the evaluation item information, and storing a result of the judging in the storage apparatus; and

controlling an autonomous policy cycle for causing the processing apparatus to conduct the confirmation policy processing, the execution policy processing and the evaluation policy processing in order.

2. The policy control method according to claim 1, wherein the confirmation item information, the presumed cause information, the execution policy information and the evaluation item information are stored in a management database area in the storage apparatus, and managed.

3. The policy control method according to claim 2, wherein, if the measure is judged to be effective, then the number of times of pattern validity in the confirmation item information is stored in the management database area, the confirmation item information is acquired from the storage apparatus in a decreasing order of the number of times of pattern validity, and the confirmation policy processing is conducted.

4. The policy control method according to claim 3, wherein operation know-how is accumulated by storing information according to know-how for a fault place in the confirmation item information, the presumed cause information, the execution policy information and the evaluation item information in the management database area.

5. The policy control method according to claim 2, wherein, if the confirmation item value does not satisfy the condition, then the number of times of pattern non-matching in the confirmation item information is stored in the management database area, and insufficiency of pattern kinds in the confirmation item information is output to an output apparatus to give warning.

6. The policy control method according to claim 2, wherein operation know-how is accumulated by storing information according to know-how for a fault place in the confirmation item information, the presumed cause information, the execution policy information and the evaluation item information in the management database area.

7. The policy control method according to claim 2, wherein:

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted,

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted, and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

8. The policy control method according to claim 1, wherein, if the measure is judged to be effective, then the number of times of pattern validity in the confirmation item information is stored in a management database area, the confirmation item information is acquired from the storage apparatus in a decreasing order of the number of times of pattern validity, and the confirmation policy processing is conducted.

9. The policy control method according to claim 8, wherein, if the confirmation item value does not satisfy the condition, then the number of times of pattern non-matching in the confirmation item information is stored in the management database area, and insufficiency of pattern kinds in the confirmation item information is output to an output apparatus to give warning.

10. The policy control method according to claim 8, wherein operation know-how is accumulated by storing information according to know-how for a fault place in the confirmation item information, the presumed cause information, the execution policy information and the evaluation item information in the management database area.

11. The policy control method according to claim 8, wherein:

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted,

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted, and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

12. The policy control method according to claim 1, wherein, if the confirmation item value does not satisfy the condition, then the number of times of pattern non-matching in the confirmation item information is stored in the management database area, and insufficiency of pattern kinds in the confirmation item information is output to an output apparatus to give warning.

13. The policy control method according to claim 12, wherein:

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted,

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted, and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

14. The policy control method according to claim 1, wherein operation know-how is accumulated by storing information according to know-how for a fault place in the confirmation item information, the presumed cause information, the execution policy information and the evaluation item information in a management database area.

15. The policy control method according to claim 14, wherein:

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted,

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted, and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

16. The policy control method according to claim 14, wherein configuration information of each computer is acquired from an agent processing part of the computer in the system to generate the system configuration information.

17. The policy control method according to claim 1, wherein:

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted,

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted, and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

18. The policy control method according to claim 17, wherein business information of each computer is acquired from an agent processing part of the computer in the system to generate the business configuration information.

19. The policy control method according to claim 1, wherein reliability of the cause is altered according to whether the evaluation item value lies between an upper limit and a lower limit of stable operation or not.

20. A policy control apparatus responsive to occurrence of a fault or abnormality in a computer to distinguish a fault occurrence place and a cause and take a measure by using policies, the policy control apparatus comprising:

a confirmation policy processing part used to conduct confirmation policy processing for acquiring confirmation item information from a storage apparatus, the confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading out presumed cause information from the storage apparatus when a confirmation item value of a confirmation subject object satisfies a condition in the confirmation item information, the presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition, and judging whether a probability of the presumed cause is greater than a probability of a cause for automatic execution;

an execution policy processing part used to conduct execution policy processing for causing a processing apparatus to execute contents of a measure read out from execution policy information, the execution policy information indicating a measure against the acquired presumed cause;

an evaluation policy processing part used to conduct evaluation policy processing for acquiring evaluation item information from the storage apparatus to evaluate a result of the measure, judging an effect of the executed measure on the basis of whether an evaluation item value of an evaluation subject object is in a stable operation range indicated by the evaluation item information, and storing a result of the judging in the storage apparatus; and

an autonomous policy processing part used to control an autonomous policy cycle for causing the processing apparatus to conduct the confirmation policy processing, the execution policy processing and the evaluation policy processing in order.

21. A computer readable storage medium embodying program carried out by a computer to execute a policy control method responsive to occurrence of a fault or abnormality in a computer to distinguish a fault occurrence place and a cause and take a measure by using policies, the policy control method comprising:

conducting confirmation policy processing for acquiring confirmation item information from a storage apparatus, the confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading out presumed cause information from the storage apparatus when a confirmation item value of a confirmation subject object satisfies a condition in the confirmation item information, the presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition, and judging whether a probability of the presumed cause is greater than a probability of a cause for automatic execution;

conducting execution policy processing for causing a processing apparatus to execute contents of a measure read out from execution policy information, the execution policy information indicating a measure against the acquired presumed cause;

conducting evaluation policy processing for acquiring evaluation item information from the storage apparatus to evaluate a result of the measure, judging an effect of the executed measure on the basis of whether an evaluation item value of an evaluation subject object is in a stable operation range indicated by the evaluation item information, and storing a result of the judging in the storage apparatus; and

controlling an autonomous policy cycle for causing the processing apparatus to conduct the confirmation policy processing, the execution policy processing and the evaluation policy processing in order.

22. A policy control method responsive to occurrence of a fault or abnormality in a computer to distinguish a fault occurrence place and a cause and take a measure by using policies, the policy control method comprising:

conducting confirmation policy processing for acquiring confirmation item information from a storage apparatus, the confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading out presumed cause information from the storage apparatus when a confirmation item value of a confirmation subject object satisfies a condition in the confirmation item information, the presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition;

conducting execution policy processing for causing a processing apparatus to execute contents of a measure read out from execution policy information, the execution policy information indicating a measure against the acquired presumed cause;

conducting evaluation policy processing for acquiring evaluation item information from the storage apparatus to evaluate a result of the measure, judging an effect of the executed measure on the basis of whether an evaluation item value of an evaluation subject object is in a stable operation range indicated by the evaluation item information, and storing a result of the judging in the storage apparatus;

controlling an autonomous policy cycle for causing the processing apparatus to conduct the confirmation policy processing, the execution policy processing and the evaluation policy processing in order; and

wherein;

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted,

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted, and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

23. A policy control method responsive to occurrence of a fault or abnormality in a computer to distinguish a fault occurrence place and a cause and take a measure by using policies, the policy control method comprising:

conducting confirmation policy processing for acquiring confirmation item information from a storage apparatus, the confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading out presumed cause information from the storage apparatus when a confirmation item value of a confirmation subject object satisfies a condition in the confirmation item information, the presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition;

conducting execution policy processing for causing a processing apparatus to execute contents of a measure read out from execution policy information, the execution policy information indicating a measure against the acquired presumed cause;

conducting evaluation policy processing for acquiring evaluation item information from the storage apparatus to evaluate a result of the measure, judging an effect of the executed measure on the basis of whether an evaluation item value of an evaluation subject object is in a stable operation range indicated by the evaluation item information, and storing a result of the judging in the storage apparatus;

controlling an autonomous policy cycle for causing the processing apparatus to conduct the confirmation policy processing, the execution policy processing and the evaluation policy processing in order;

wherein the confirmation item information, the presumed cause information, the execution policy information and the evaluation item information are stored in a management database area in the storage apparatus, and managed;

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted;

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted; and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

24. A policy control method responsive to occurrence of a fault or abnormality in a computer to distinguish a fault occurrence place and a cause and take a measure by using policies, the policy control method comprising:

conducting confirmation policy processing for acquiring confirmation item information from a storage apparatus, the confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading out presumed cause information from the storage apparatus when a confirmation item value of a confirmation subject object satisfies a condition in the confirmation item information, the presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition;

conducting execution policy processing for causing a processing apparatus to execute contents of a measure read out from execution policy information, the execution policy information indicating a measure against the acquired presumed cause;

conducting evaluation policy processing for acquiring evaluation item information from the storage apparatus to evaluate a result of the measure, judging an effect of the executed measure on the basis of whether an evaluation item value of an evaluation subject object is in a stable operation range indicated by the evaluation item information, and storing a result of the judging in the storage apparatus;

controlling an autonomous policy cycle for causing the processing apparatus to conduct the confirmation policy processing, the execution policy processing and the evaluation policy processing in order;

wherein, if the measure is judged to be effective, then the number of times of pattern validity in the confirmation item information is stored in a management database area, the confirmation item information is acquired from the storage apparatus in a decreasing order of the number of times of pattern validity, and the confirmation policy processing is conducted; and

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted;

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted; and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

25. A policy control method responsive to occurrence of a fault or abnormality in a computer to distinguish a fault occurrence place and a cause and take a measure by using policies, the policy control method comprising:

conducting confirmation policy processing for acquiring confirmation item information from a storage apparatus, the confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading out presumed cause information from the storage apparatus when a confirmation item value of a confirmation subject object satisfies a condition in the confirmation item information, the presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition;

conducting execution policy processing for causing a processing apparatus to execute contents of a measure read out from execution policy information, the execution policy information indicating a measure against the acquired presumed cause;

conducting evaluation policy processing for acquiring evaluation item information from the storage apparatus to evaluate a result of the measure, judging an effect of the executed measure on the basis of whether an evaluation item value of an evaluation subject object is in a stable operation range indicated by the evaluation item information, and storing a result of the judging in the storage apparatus;

controlling an autonomous policy cycle for causing the processing apparatus to conduct the confirmation policy processing, the execution policy processing and the evaluation policy processing in order;

wherein, if the confirmation item value does not satisfy the condition, then the number of times of pattern non-matching in the confirmation item information is stored in the management database area, and insufficiency of pattern kinds in the confirmation item information is output to an output apparatus to give warning; and

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted;

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted; and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.

26. A policy control method responsive to occurrence of a fault or abnormality in a computer to distinguish a fault occurrence place and a cause and take a measure by using policies, the policy control method comprising:

conducting confirmation policy processing for acquiring confirmation item information from a storage apparatus, the confirmation item information indicating a condition every confirmation item to distinguish a cause at time of fault occurrence, reading out presumed cause information from the storage apparatus when a confirmation item value of a confirmation subject object satisfies a condition in the confirmation item information, the presumed cause information indicating a presumed cause corresponding to each pattern in the confirmation item information, and acquiring a presumed cause corresponding to the confirmation item information of the condition;

conducting execution policy processing for causing a processing apparatus to execute contents of a measure read out from execution policy information, the execution policy information indicating a measure against the acquired presumed cause;

conducting evaluation policy processing for acquiring evaluation item information from the storage apparatus to evaluate a result of the measure, judging an effect of the executed measure on the basis of whether an evaluation item value of an evaluation subject object is in a stable operation range indicated by the evaluation item information, and storing a result of the judging in the storage apparatus;

controlling an autonomous policy cycle for causing the processing apparatus to conduct the confirmation policy processing, the execution policy processing and the evaluation policy processing in order;

wherein operation know-how is accumulated by storing information according to know-how for a fault place in the confirmation item information, the presumed cause information, the execution policy information and the evaluation item information in a management database area; and

system configuration information indicating a computer configuration in a system is read out from the storage apparatus, and an address of a computer that has issued a fault event or a load increase event is extracted;

business configuration information indicating a business configuration in the system is read out from the storage apparatus, and a business application classification and a business object corresponding to the extracted address are extracted; and

the autonomous policy cycle is conducted on the business object of the extracted business application classification.