Discovery troubleshooting based on machine learning

Patent metadata

Field	Value
Publication number	US-12222834-B2
Application number	US-202318323072-A
Country	US
Kind code	B2
Filing date	May 24, 2023
Priority date	May 24, 2023
Publication date	Feb 11, 2025
Grant date	Feb 11, 2025

Abstract

A method includes obtaining a discovery pattern that indicates a plurality of operations associated with a corresponding computing resource type of a plurality of computing resource types. The method also includes identifying a variable parameter value associated with execution of the discovery pattern with respect to a computing resource of the corresponding computing resource type, and determining an error value by using a machine learning model to process the variable parameter value. The error value indicates a likelihood that execution of the discovery pattern, when associated with the variable parameter value, with respect to the particular computing resource results in a corresponding error type. The method further includes receiving, based on the error value, an input comprising one or more of (i) an instruction to execute the discovery pattern or (ii) a modification applicable to the variable parameter value.

First claim

What is claimed is:

1. A method comprising:
obtaining a discovery pattern that indicates a plurality of operations configured to acquire information about a corresponding computing resource type of a plurality of computing resource types;
identifying a variable parameter value that is associated with and affects execution of the discovery pattern with respect to a computing resource of the corresponding computing resource type;
determining, prior to execution of the discovery pattern with respect to the computing resource, an error value by using a machine learning model to process the variable parameter value, wherein the error value indicates a likelihood that execution of the discovery pattern, when associated with the variable parameter value, with respect to the computing resource results in a corresponding error type, and wherein occurrence of an error of the corresponding error type during the execution of the discovery pattern prevents the discovery pattern from obtaining at least some of the information about the computing resource; and
receiving, based on the error value, an input comprising one or more of (i) an instruction to execute the discovery pattern or (ii) a modification applicable to the variable parameter value.

2. The method of claim 1 , wherein determining the error value comprises:
determining a plurality of error values by processing the variable parameter value by the machine learning model, wherein each respective error value of the plurality of error values is associated with a corresponding error type of a plurality of different error types and indicates a likelihood that execution of the discovery pattern, when associated with the variable parameter value, with respect to the computing resource results in the corresponding error type.

3. The method of claim 2 , wherein the variable parameter value comprises a plurality of variable parameter values of a plurality of different variable parameters of the discovery pattern, wherein each respective error type of the plurality of different error types is associated with a corresponding variable parameter of the plurality of different variable parameters, wherein one or more variable parameter values of the corresponding variable parameter are associated with occurrence of the respective error type, and wherein the method further comprises:
determining, based on the plurality of error values, at least one error value that exceeds a threshold error value; and
displaying an indication to modify, prior to execution of the discovery pattern, a variable parameter value of the variable parameter that is associated with the error type of the at least one error value.

4. The method of claim 2 , wherein the machine learning model comprises a plurality of machine learning models, wherein each respective machine learning model of the plurality of machine learning models is (i) associated with a corresponding error type of a plurality of different error types and (ii) configured to generate the respective error value for the corresponding error type.

5. The method of claim 2 , wherein the plurality of different error types comprises two or more of: (i) a bad request error indicating that a request made by the discovery pattern is invalid, (ii) a graceful termination error indicating that the computing resource does not exist, (iii) a throttling error indicating that a number of requests to a computing system containing the computing resource exceeds a threshold number of requests, (iv) a socket timeout error indicating that a flow of data from the computing system has been interrupted, or (v) a permissions error indicating that the discovery pattern is not permitted to access information about the computing resource.

6. The method of claim 1 , wherein the machine learning model has been trained based on log data representing outcomes of prior executions of the discovery pattern, wherein the log data comprises, for each respective prior execution of the prior executions, (i) a corresponding variable parameter training value and (ii) a ground-truth outcome of the respective prior execution associated with the corresponding variable parameter training value.

7. The method of claim 6 , wherein the log data comprises (i) internal log data generated by a managed network with which the computing resource is associated and (ii) external log data generated by one or more other managed networks, and wherein the machine learning model has been pre-trained using the external log data and fine-tuned using the internal log data.

8. The method of claim 1 , wherein receiving the input comprises:
causing the error value to be displayed; and
receiving the input after causing the error value to be displayed.

9. The method of claim 1 , further comprising:
determining that the error value exceeds a threshold error value;
based on determining that the error value exceeds the threshold error value, displaying a prompt indicating to modify the variable parameter value prior to executing the discovery pattern; and
after displaying the prompt, obtaining the modification applicable to the variable parameter value.

10. The method of claim 9 , further comprising:
based on determining that the error value exceeds the threshold error value, blocking execution of the discovery pattern until the variable parameter value is modified.

11. The method of claim 1 , further comprising:
determining that the error value does not exceed a threshold error value; and
based on receiving the instruction to execute the discovery pattern and determining that the error value does not exceed the threshold error value, executing the discovery pattern.

12. The method of claim 1 , wherein the variable parameter value represents an input to the discovery pattern configured to allow the discovery pattern to execute with respect to the computing resource.

13. The method of claim 1 , wherein the variable parameter value represents an attribute of one or more prior executions of the discovery pattern with respect to one or more computing resources of the corresponding computing resource type.

14. The method of claim 1 , wherein the variable parameter value represents an attribute of a context in which the discovery pattern is planned to execute with respect to the computing resource.

15. A non-transitory computer-readable medium, having stored thereon program instructions that, upon execution by a computing system, cause the computing system to perform operations comprising:
obtaining a discovery pattern that indicates a plurality of operations configured to acquire information about a corresponding computing resource type of a plurality of computing resource types;
identifying a variable parameter value that is associated with and affects execution of the discovery pattern with respect to a computing resource of the corresponding computing resource type;
determining, prior to execution of the discovery pattern with respect to the computing resource, an error value by using a machine learning model to process the variable parameter value, wherein the error value indicates a likelihood that execution of the discovery pattern, when associated with the variable parameter value, with respect to the computing resource results in a corresponding error type, and wherein occurrence of an error of the corresponding error type during the execution of the discovery pattern prevents the discovery pattern from obtaining at least some of the information about the computing resource; and
receiving, based on the error value, an input comprising one or more of (i) an instruction to execute the discovery pattern or (i

Assignees

• Servicenow Inc

Inventors

• Herzog Shay
• Garty Asaf

Classifications

• H04L41/069 Primary

  using logs of notifications; Post-processing of notifications · CPC title

• G06F11/3006

  where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems (multiprogramming arrangements G06F9/46; allocation of resources G06F9/50) · CPC title

• H04L41/16

  using machine learning or artificial intelligence · CPC title

• G06F11/3051 Primary

  Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs (verification or detection of system hardware configuration G06F11/2247) · CPC title