Method and apparatus for adaptive declarative monitoring

• US 8,769,346 B2
• Filed: 11/20/2008
• Issued: 07/01/2014
• Est. Priority Date: 11/21/2007
• Status: Active Grant

• Current Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Original Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Inventors: Sijelmassi, Rachid
• Primary Examiner(s): Riad, Amine
• Application Number:
  US12/274,930
• Publication Number:
  US 20090177929A1
• Time in Patent Office: 2,049 Days
• Field of Search: 714/47
• US Class Current: 714/47
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/875 Monitoring of systems including the internet

Abstract

A method of and apparatus for monitoring a computer system includes defining a monitoring policy for the computer system. At least one computer is employed to determine a status of a state of the computer system relative to the monitoring policy. At least one computer is employed to determine a condition of at least one monitored element to be monitored in the computer system based on the status of the state of the computer system. Furthermore, at least one computer is employed to monitor the condition of the at least one monitored element in the computer system, based on the monitoring policy. At least one computer is employed to perform an action in response to the condition assuming a predetermined status.

Claims

1. A method of monitoring a computer system comprising;

defining a monitoring policy for the computer system, the monitoring policy comprising at least one self-healing action rule;

selecting at least one monitored element from among a plurality of elements of the computer system that is necessary to be monitored to determine an operating condition of the at least one monitored element, wherein the selecting is responsive to data structures containing information for applying the monitoring policy;

monitoring, with the computer system, the operating condition of the at least one monitored element based on the monitoring policy; and

performing, with the computer system, at least one action that executes the at least one self-healing action rule to modify at least one operation of the computer system in response to the operating condition having a predetermined status, the at least one action causing the operating condition to no longer have the predetermined status,

wherein the monitoring comprises selectively starting or stopping monitoring of the at least one monitored element responsive to the at least one self-healing action rule and the operating condition.

2. The method of claim 1, wherein the defining of the monitoring policy includes defining Action Rules, states and contexts, the determining of the operating condition is based on the context, and the performing is based on the Action Rules.

3. The method of claim 2, wherein the context comprises data including rules, data collected from monitored elements, thresholds, and/or computer names.

4. The method of claim 1, wherein the performing includes setting or resetting a state, or enabling or disabling the performing.

5. The method of claim 4, wherein the state is defined by the operating condition being monitored or an action of a user.

6. The method of claim 1, wherein the defining of the monitoring policy is performed using natural language, and wherein the performing further comprises:

translating the monitoring policy into an extensible markup language (XML) format from the natural language; and

generating at least two sets of rules for each state of a monitored element responsive to the monitoring policy stored in the XML format.

7. The method of claim 6, wherein the at least two sets of rules comprises:

sending a request to a monitoring agent to start monitoring a referenced monitoring element responsive to a determination that the referenced monitoring element is not monitored and the monitoring policy is activated; and

sending a request to the monitoring agent to stop monitoring the referenced monitoring element responsive to a determination that the referenced monitoring element is monitored and the monitoring policy is de-activated.

8. The method of claim 6, wherein the at least two sets of rules comprises:

sending a request to a monitoring agent to start monitoring a referenced monitoring element responsive to a determination that the referenced monitoring element is not monitored, the monitoring policy is activated and a Boolean expression of a state of the referenced monitoring element is met;

sending a request to the monitoring agent to stop monitoring the referenced monitoring element responsive to a determination that the referenced monitoring element is monitored, the monitoring policy is de-activated and a Boolean expression of a state of the referenced monitoring element is not met; and

executing actions specified by the monitoring policy responsive to a determination that the referenced monitoring element is monitored, the monitoring policy is activated and a Boolean expression of a state of the referenced monitoring element is true.

9. The method of claim 8, further comprising:

returning information to a policy manager responsive to a determination that the operating condition has been met.

10. The method of claim 1, wherein the operating condition of the at least one monitored element is a utilization of the at least one monitored element.

11. The method of claim 10, wherein the monitoring includes checking the utilization of the at least one monitored element and comparing the utilization against a predetermined threshold.

12. The method of claim 1, wherein the at least one monitored element is a web tier of an application running on the computer system or a web server, and executing the at least one self-healing action rule comprises:

determining whether the web tier or web server is saturated, wherein the web tier or web server is saturated when a number of connections meets a connection threshold during a threshold period of time;

unprovisioning a predetermined portion of the web tier or web server and/or accepting new connections responsive to a determination that the web tier or web server is not saturated; and

running a remediation workflow to provision a new virtual machine responsive to a determination that the web tier or web server is saturated.

13. The method of claim 1, wherein defining the monitoring policy includes consolidating monitoring requirements of a plurality of active management policies.

14. The method of claim 1, wherein the operating condition is a threshold, a Boolean condition, and/or a variable.

15. The method of claim 1, wherein the monitored element is a hardware component, a software component, and/or a combination of a hardware component and a software component.

16. A computer system comprising:

a policy builder module of the computer system constructed and arranged to create a monitoring policy for the computer system, the monitoring policy comprising at least one self-healing action rule;

a policy manager module of the computer system constructed and arranged to determine an operating condition to be monitored in the computer system based on the monitoring policy;

a rules engine module of the computer system constructed and arranged to select at least one monitored element from among a plurality of elements of the computer system that is necessary to be monitored to determine the operating condition responsive to data structures containing information for applying the monitoring policy; and

a monitoring agent module of the computer system constructed and arranged to monitor the operating condition of the at least one monitored element, wherein the monitoring agent is further arranged to selectively start or stop monitoring of the at least one monitored element responsive to the at least one self-healing action rule and the operating condition,

wherein the rules engine module of the computer system, responsive to the monitoring of the monitoring agent module, is constructed and arranged to perform at least one action that executes the at least one self-healing action rule to modify at least one operation of the computer system in response to the operating condition monitored by the monitoring agent module having a predetermined status, the at least one action causing the operating condition to no longer have the predetermined status.

17. The computer system of claim 16, wherein the policy manager module is constructed and arranged to create Action Rules, states and contexts as the monitoring policy, wherein the operating condition determined by the policy manager is based on the context and the rules engine performs the action based on the Action Rules.

18. The computer system of claim 17, wherein the context comprises data including rules, data collected from monitored elements, thresholds, and/or computer names.

19. The computer system of claim 16, wherein action performed by the rules engine module includes setting or resetting a state, or enabling or disabling the performing.

20. The computer system of claim 19, wherein the policy manager module determines a status of the state based on the operating condition being monitored or an action of a user.

21. The computer system of claim 16, wherein the policy builder module creates the monitoring policy based on natural language input to the policy builder module in natural languages, and wherein the policy manager module:

translates the monitoring policy into an extensible markup language (XML) format from the natural language; and

generates at least two sets of rules for each state of a monitored element responsive to the monitoring policy stored in the XML format.

22. The computer system of claim 21, wherein the at least two sets of rules comprises:

sending a request to a monitoring agent to start monitoring a referenced monitoring element responsive to a determination that the referenced monitoring element is not monitored and the monitoring policy is activated; and

sending a request to the monitoring agent to stop monitoring the referenced monitoring element responsive to a determination that the referenced monitoring element is monitored and the monitoring policy is de-activated.

23. The computer system of claim 21, wherein the at least two sets of rules comprises:

sending a request to a monitoring agent to start monitoring a referenced monitoring element responsive to a determination that the referenced monitoring element is not monitored, the monitoring policy is activated and a Boolean expression of a state of the referenced monitoring element is met;

sending a request to the monitoring agent to stop monitoring the referenced monitoring element responsive to a determination that the referenced monitoring element is monitored, the monitoring policy is de-activated and a Boolean expression of a state of the referenced monitoring element is not met; and

executing actions specified by the monitoring policy responsive to a determination that the referenced monitoring element is monitored, the monitoring policy is activated and a Boolean expression of a state of the referenced monitoring element is true.

24. The computer system of claim 16, wherein the at least one monitored element is a web tier of an application running on the computer system, and wherein the rules engine module:

determines whether the web tier or web server is saturated, wherein the web tier or web server is saturated when a number of connections meets a connection threshold during a threshold period of time;

unprovisions a predetermined portion of the web tier or web server and/or accepting new connections responsive to a determination that the web tier or web server is not saturated; and

runs a remediation workflow to provision a new virtual machine responsive to a determination that the web tier or web server is saturated.

25. The computer system of claim 16, wherein the monitoring agent module checks the utilization of the at least one monitored element and compares the utilization against a predetermined threshold.

26. The computer system of claim 16, wherein creating the monitoring policy includes consolidating monitoring requirements of a plurality of active management policies.

27. The computer system of claim 16, wherein the operating condition is a threshold, a Boolean condition, and/or a variable.

28. The computer system of claim 16, wherein the monitored element is a hardware component, a software component, and/or a combination of a hardware component and a software component.