Method of debugging a software system

• US 8,166,348 B1
• Filed: 03/29/2010
• Issued: 04/24/2012
• Est. Priority Date: 03/29/2010
• Status: Expired due to Fees

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Kulkarni, Gururaj Govind, Chopra, Shelesh, Vasanad, Shivashankar
• Primary Examiner(s): MCCARTHY, CHRISTOPHER S
• Application Number:
  US12/748,493
• Time in Patent Office: 757 Days
• Field of Search: None
• US Class Current: 714/38.1
• 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 11/3636 by tracing the execution of the program
• CPC Class Codes: G06F 11/366 using diagnostics G06F11/0703 takes precedence

Abstract

A method, system, and a computer program product for debugging a software system. An aspect of the present invention provides a user with the combination of an error determined in the software system and a set of recommendations for fixing the error, thereby enabling the user to debug the software system more efficiently. In one embodiment, the error is determined by inspecting the execution states of the software system at different time instances prior to receiving a debug indication from the user. The set of recommendations for fixing the error is identified by searching a knowledge base that stores the recommendations (e.g., possible solutions, suggestions, and workarounds) for each of the errors.

Claims

1. A computer implemented method of debugging a software system, the software system comprising software components, the method comprising:

enforcing each of the software components to use a common framework for storing execution states;

storing execution states of the software system in a first format, wherein each of the execution states corresponds to one of the software components at a corresponding time instance,

wherein the common framework is designed to store the execution states in the first format, the first format including a first field identifying the specific one of the software components storing a corresponding execution state,

wherein the common framework stores the execution states in a plurality of log files, each log file corresponding to a software component;

receiving, from a user, an indication to debug the software system, wherein the indication specifies a time instance;

determining the occurrence of a first error in the software system by inspecting at least one of the stored execution states corresponding to time instances prior to the time instance, wherein the first error is determined based on the stored execution states of at least two different software components,

wherein the determining comprises;

generating a set of execution characteristics corresponding to each of the plurality of time instances based on the plurality of execution states;

storing the set of execution characteristics for each of the plurality of time instances in a database; and

querying the set of execution characteristics maintained in the database to determine the occurrence of the first error in the software system,

wherein the generating further comprises;

parsing each of the execution states stored in the first format to identify corresponding values for a set of fields,

wherein the parsing comprises using a common parser to parse all of the plurality of log files to determine the set of execution characteristics of the software system; and

creating a first subset of execution characteristics based on the values for the set of fields identified for an execution state, wherein the first subset of execution characteristics is contained in the set of execution characteristics for a corresponding time instance;

maintaining a knowledge base specifying a plurality of sets of recommendations for a corresponding set of errors, wherein the set of errors includes the first error;

identifying a first set of recommendations for fixing the first error,

wherein the identifying further comprises;

searching, after the determining the first error, the knowledge base for recommendations for the first error; and

finding the first set of recommendations in the plurality of sets of recommendations corresponding to the first error; and

providing, to the user, the first error and the first set of recommendations to enable the user to debug the software system,

wherein the determining, the identifying and the providing are performed in response to the receiving the indication to debug the software system.

2. The method of claim 1, wherein the first subset of execution characteristics includes a execution state identifier, an execution state category, a time instance at which the corresponding execution state was stored, an identifier of a software component corresponding to the first field, a state message corresponding to information in the execution state and a file identifier corresponding to the identifier of the log file in which the execution state was stored.

3. The method of claim 1, wherein the generating further comprises:

computing a second subset of execution characteristics based on the values for the set of fields identified for some of the execution states, wherein the second subset of execution characteristics is contained in the set of execution characteristics for a corresponding time instance.

4. The method of claim 3, wherein the second set of execution characteristics includes a count of the number of occurrences of an execution state identifier in the execution states.

5. The method of claim 1, wherein the indication further specifies a set of criteria for debugging the software system, wherein the parsing is performed according to the set of criteria,

wherein the set of criteria includes a first criteria indicating the specific software components contained in the software system to be debugged, a second criteria indicating the specific execution state category to be inspected, a third criteria specifying a time interval in which execution states are to be inspected and a fourth criteria specifying a text to be searched in the state messages of the execution states.

6. The method of claim 5, wherein the storing stores the set of execution characteristics for each of the time instances in the database according to a second format,

wherein the determining further comprises:

maintaining a set of rules for determining the occurrences of errors;

forming a first query based on a first rule contained in the set of rules, wherein the first rule specifies a first set of conditions to be satisfied by a set of execution characteristics for determining the occurrence of the first error; and

executing the first query on the database to determine the occurrence of the first error in the software system if there exists a first set of execution characteristics corresponding to a first time instance in the database which satisfies the first set of conditions.

7. A non-transitory machine readable medium storing one or more sequences of instructions for causing a system to debug a software system, the software system comprising software components, wherein execution of the one or more sequences of instructions by one or more processors contained in the system causes the system to perform the actions of:

storing execution states of the software system, wherein each of the execution states corresponds to one of the plurality of software components at a corresponding time instance;

maintaining a knowledge base specifying a plurality of recommendations for fixing a set of errors;

receiving, from a user, an indication to debug the software system, wherein the indication specifies a time instance;

determining the occurrence of a first error in the software system by inspecting a subset of the execution states corresponding to time instances prior to the time instance, wherein the first error is determined based on the stored execution states of at least two different software components;

identifying, in the knowledge base, a first set of recommendations for fixing the first error, wherein the first set of recommendations is contained in the plurality of recommendations; and

providing, to the user, the first error and the first set of recommendations to enable the user to debug the software system,

wherein the determining, the identifying and the providing are performed in response to the receiving the indication to debug the software system

wherein the knowledge base comprises a error message repository, a set of product release notes, a support knowledge base and a defect tracking system,

wherein the recommendations include possible solutions, suggestions, and workarounds for fixing the first error.

8. The non-transitory machine readable medium of claim 7, wherein each of the set of errors is uniquely identified in the software system by a corresponding system error identifier,

wherein the maintaining maintains each of the plurality of recommendations associated with the system error identifier of a corresponding error, wherein the first set of recommendations is maintained associated with a first system error identifier of the first error,

wherein the identifying identifies the first set of recommendations in the plurality of sets of recommendations based on matching the first system error identifier of the first error.

9. The non-transitory machine readable medium of claim 7, wherein the software system stores each of the execution states according to a first format, wherein the determining further comprises:

parsing each of the execution states stored in the first format to identify corresponding values for a set of fields;

generating a set of execution characteristics corresponding to each of the time instances based on the values for the set of fields identified for an execution state;

storing the set of execution characteristics for each of the time instances in a database in a second format, wherein the second format is different from the first format; and

querying the set of execution characteristics maintained in the second format in the database to determine the occurrence of the first error in the software system.

10. The non-transitory machine readable medium of claim 9, wherein the querying further comprises:

maintaining a set of rules for determining the occurrences of the set of errors;

forming a first query based on a first rule contained in the set of rules, wherein the first rule specifies a first set of conditions to be satisfied by a set of execution characteristics for determining the occurrence of the first error; and

executing the first query on the database to determine the occurrence of the first error in the software system if there exists a first set of execution characteristics corresponding to a first time instance in the database which satisfies the first set of conditions.

11. A computing system comprising:

a plurality of systems to execute software components forming a software system, wherein the software system is designed to store the execution states of the software components corresponding to different time instances,

wherein the software system stores in the one or more files, each of the execution states according to a first format;

a database to maintain data in a second format, wherein the second format is different from the first format;

a user system to enable a user to send an indication to debug the software system; and

at least one processor executing instructions retrieved from a memory to operate as a debug tool, the debug tool to receive the indication at a time instance and in response operable to;

determine the occurrence of a first error in the software system by inspecting a subset of the execution states corresponding to a plurality of time instances prior to the time instance, wherein the debug tool determines the first error based on the stored execution states of at least two different software components;

identify a first set of recommendations for fixing the first error; and

provide, to the user, the first error and the first set of recommendations to enable the user to debug the software system,

wherein the debug tool to determine the first error is operable to;

parse each of the execution states stored in the first format to identify corresponding values for a set of fields;

generate a set of execution characteristics corresponding to each of the time instances based on the values for the set of fields identified for an execution state;

store the set of execution characteristics for each of the time instances in the database in the second format; and

query the set of execution characteristics maintained in the second format in the database to determine the occurrence of the first error in the software system; and

a knowledge base maintaining a plurality of recommendations for fixing a set of errors including the first error, wherein each of the set of errors is uniquely identified in the software system by a corresponding system error identifier,

wherein the knowledge base maintains each of the plurality of recommendations associated with the system error identifier of a corresponding error, wherein the first set of recommendations is maintained associated with a first system error identifier of the first error,

wherein the debug tool identifies the first set of recommendations in the plurality of sets of recommendations based on matching the first system error identifier of the first error,

wherein the knowledge base is stored in a non-volatile storage separate from one or more files in which the execution states are stored.