Priority-based network fault analysis

• US 20060104208A1
• Filed: 11/12/2004
• Published: 05/18/2006
• Est. Priority Date: 11/12/2004
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Natarajan, Srikanth, Wong, Kam Chooi, Walker, Anthony Paul Michael, Smith, Darren D., Huonder, Russell John, Rhodes, David M.
• Granted Patent: US 8,068,432 B2
• Time in Patent Office: Days
• US Class Current: 370/242
• CPC Class Codes: H04L 41/0631 using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
• CPC Class Codes: H04L 41/0677 Localisation of faults
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route

Abstract

A method and system for reducing fault analysis times in a network having a plurality of pollable objects. Poll responses are received from pollable objects, priorities are assigned to each received poll response in accordance with one or more of a received poll response value, a pollable object status and a type of pollable object, and the assigned priorities are processed and to determine any fault locations, wherein a pollable object with a higher priority is processed before a second pollable object with a lower priority. A status manager is operable to receive poll responses from one or more of the pollable objects, a status analyzer is operable to assign, in conjunction with one or more analyzers, priorities to each received poll response in accordance with the received poll response value, a pollable object status and a type of pollable object, and analyzers coupled the status analyzer use the assigned priorities and the corresponding received poll responses to determine a fault location.

Claims

1. A method for reducing fault analysis times in a network comprising a plurality of pollable objects, comprising:

receiving one or more poll responses from one or more of the plurality of pollable objects;

assigning priorities to each received poll response of the one or more received poll responses in accordance with one or more of a received poll response value, a pollable object status and a type of pollable object; and

processing the assigned priorities and the corresponding received poll responses to determine any fault locations, wherein for the plurality of pollable objects a poll result of a pollable object with a higher priority is processed before a second poll result of a second pollable object with a lower priority.

2. The method of claim 1, wherein the fault analysis times are one or more of:

mean time from failure to report of root cause;

mean time from failure to fix of root cause; and

mean time from failure to processing of all impacted devices.

3. The method of claim 1, further comprising using more than one thread to poll the plurality of pollable objects.

4. The method of claim 1, further comprising listing and sorting by priority the received poll responses with assigned priorities.

5. The method of claim 1, further comprising prior to receiving poll responses, polling the one or more pollable objects.

6. The method of claim 1, further comprising using one or more threads to analyze the received poll responses.

7. The method of claim 1, wherein an UP pollable object status and a DOWN pollable object status has a higher priority than a NORESPONSE pollable object status.

8. The method of claim 1, further comprising assigning the plurality of pollable objects a default priority, wherein the default priority is set in accordance with one or more of the pollable object status and a poll response value.

9. The method of claim 8, further comprising setting the default priority to one or more of:

High, when the pollable object status is DOWN;

Medium, when the pollable object status is one of UP and MARGINAL;

Low, when the pollable object status is one of UNDETERMINED and NORESPONSE; and

Very Low, when the pollable object status is VALIDATE.

10. The method of claim 8, further comprising overriding the default priority using one or more of the pollable object type and the pollable object status.

11. The method of claim 10, further comprising overriding the default priority to one of:

High, when the pollable object is one of an Egress router and a gateway router and the pollable object status is one of DOWN, UP, UNDETERMINED, and UNRESPONSIVE; and

Medium High, when the pollable object is one of a node, interface and IP address and the pollable object status is UP and a previous pollable object status value is one of DOWN and MARGINAL.

12. The method of claim 1, further comprising assigning pollable objects in a Fault Area a higher priority than pollable objects in a Far-From-Fault Area.

13. A system operable to reduce fault analysis times of a network comprising a plurality of pollable objects, the system comprising:

a status manager operable to receive poll responses from one or more of the plurality of pollable objects;

a status analyzer operable to assign, in conjunction with one or more analyzers, priorities to each received poll response of the received poll responses in accordance with one or more of the received poll response value, a pollable object status and a type of pollable object; and

one or more analyzers coupled the status analyzer, said one or more analyzers operable to use the assigned priorities and the corresponding received poll responses to determine a fault location, wherein for the plurality of pollable objects a poll response for a pollable object with a higher priority is processed before a second poll response for a second pollable object with a lower priority.

14. The system of claim 13, wherein the type of pollable object comprises one of a gateway device, egress device, end node, address, interface, network electronic component and node.

15. The system of claim 13, wherein the pollable object status is one of UP, DOWN, MARGINAL, UNDETERMINED, NORESPONSE, and VALIDATE.

16. The system of claim 13, wherein the received poll response value is one of reachable, unreachable.

17. The system of claim 13, wherein pollable objects in a Fault Area are operable to be processed before pollable objects in a Far-From-Fault Area.

18. The system of claim 13, wherein the fault analysis times are one or more of:

mean time from failure to report of root cause;

mean time from failure to fix of root cause; and

mean time from failure to processing of all impacted devices.

19. The system of claim 13, wherein more than one thread produced by a poller is used to poll the plurality of nodes.

20. The system of claim 13, wherein the status analyzer lists and sorts by priority the received poll responses with assigned priorities.

21. The system of claim 13, wherein one or more analyzers of the status analyzer are operable to process the assigned priorities and the received poll responses to determine the fault location.

22. The system of claim 21, wherein the status analyzer is operable to select which analyzer of the one or more analyzers is used to process the assigned priorities and the received poll responses to determine the fault location.

23. The system of claim 22, wherein the analyzer assigns the assigned priority.

24. The system of claim 21, wherein more than one thread is used by the one or more analyzers.

25. A method for analyzing a network comprising a plurality of pollable objects, the method comprising:

initializing one or more analyzer threads;

scheduling for polling one or more pollable objects;

one or more polling threads polling the one or more pollable objects;

placing one or more received poll responses in a queue wherein the received poll responses are ordered in the queue based upon priority, said priority set in accordance with one or more of the received poll response value, a pollable object status and a type of pollable object; and

the one or more analyzer threads processing the queue to analyze one or more network faults, wherein for the plurality of pollable objects a poll result of a pollable object with a higher priority is processed before a second poll result of a second pollable object with a lower priority.

26. The method of claim 25, further comprising using one or more threads to process the assigned priorities and the received poll responses.

27. The method of claim 25, further comprising assigning the plurality of pollable objects a default priority, wherein the default priority is set in accordance with one or more of a pollable object status and a poll response value.

28. The method of claim 27, further comprising overriding the default priority using one or more of the pollable object type and the pollable object status.

29. The method of claim 28, further comprising setting the default priority to one or more of:

High, when the pollable object status is DOWN;

Medium, when the pollable object status is one of UP and MARGINAL;

Low, when the pollable object status is one of UNDETERMINED and NORESPONSE; and

Very Low, when the pollable object status is VALIDATE.

30. The method of claim 29, further comprising overriding the default priority to one of:

High, when the pollable object is one of an Egress router and a gateway router and the pollable object status is one of DOWN, UP, UNDETERMINED, and UNRESPONSIVE; and

Medium High, when the pollable object is one of a node, interface and IP address and the pollable object status is UP and a previous pollable object status value is one of DOWN and MARGINAL.

31. The method of claim 25, further comprising assigning pollable objects in a Fault Area a higher priority than pollable objects in a Far-From-Fault Area.

32. A system operable to analyze one or more pollable objects of a network, comprising:

one or more analyzer threads operable to schedule one or more pollable objects to be polled;

one or more polling threads operable to poll the one or more pollable objects;

a status manager operable to receive poll responses from one or more of the plurality of pollable objects;

a status analyzer operable to assign, in conjunction with one or more analyzers, a priority to each received poll response, said priority set in accordance with one or more of the received poll response value, a pollable object status and a type of pollable object, wherein the status analyzer is operable to place the received poll responses in a queue sorted by priority; and

one or more analyzers coupled to the status analyzer and operable to process the queue to analyze any network faults, wherein for the plurality of pollable objects a pollable object with a high priority is processed before a second pollable object with a low priority.

33. The system of claim 32, wherein more than one thread is operable to be used to poll the plurality of nodes.

34. The system of claim 32, wherein one or more analyzers are coupled with the status analyzer, said one or more analyzers operable to process the assigned priorities and the received poll responses to determine the fault location.

35. The system of claim 34, wherein the status analyzer is operable to select which analyzer of the one or more analyzers is used to process the assigned priorities and the received poll responses to determine a fault location.

36. The system of claim 35, wherein the status analyzer is operable to assign a default priority, wherein the default priority is set in accordance with one or more of a pollable object status and a poll response value.

37. The system of claim 34, wherein an analyzer of one or more analyzers is operable to override the default priority using one or more of the pollable object type and the pollable object status.

38. The system of claim 32, wherein more than one thread is usable by the one or more analyzers.

39. The system of claim 32, wherein an analyzer that analyzes a pollable object marks one or more of the pollable object, neighbor nodes of the pollable object, and devices connected to the pollable object busy until the analyzer has finished the analysis.

40. The system of claim 32, wherein if a pollable object that comprises an assigned priority is busy, then an analyzer skips analysis of the pollable object and selects a next item from the queue.

41. The system of claim 32, wherein pollable objects in a Fault Area are operable to be processed before pollable objects in a Far-From-Fault Area.

42. A computer readable medium storing instructions that, when executed on a programmed processor, reduce fault analysis time in a network, comprising:

instructions for receiving one or more poll responses from one or more of the plurality of pollable objects;

instructions for assigning priorities to each received poll response of the one or more received poll responses in accordance with one or more of a received poll response value, a pollable object status and a type of pollable object; and

instructions for processing the assigned priorities and the corresponding received poll responses to determine any fault locations, wherein for the plurality of pollable objects a poll result of a pollable object with a higher priority is processed before a second poll result of a second pollable object with a lower priority.

43. A system for locating a fault in a network, comprising:

means for receiving poll responses from one or more of the plurality of pollable objects;

means for assigning, in conjunction with one or more analyzers, priorities to each received poll response of the received poll responses in accordance with one or more of the received poll response value, a pollable object status and a type of pollable object; and

means for using the assigned priorities and the corresponding received poll responses to determine a fault location, wherein for the plurality of pollable objects a poll response for a pollable object with a higher priority is processed before a second poll response for a second pollable object with a lower priority.