Method for debugging computer program

Abstract

The invention proposes a debugging method. A computer running a program that generates a master core dump file and one or more slave core dump files in response to a triggering event such as an internal failure. Another computer analyzes the core dump file to debug the program. The master core dump file includes an index list, each entry of the index list corresponds to a slave core dump file of the one or more slave core dump files. The master core dump file may first be transferred to the other computer. Based on the information in the index list, the other computer selectively requests one or more of the generated slave core files for debugging.

Description

technical field [0001] The present invention relates generally to computer technology, and more particularly to a method of debugging a computer program. Background technique [0002] Debugging is an important process in developing computer programs. A so-called "bug" is an error, defect, failure or malfunction in a computer program (hereinafter referred to as a "program") or system. Vulnerabilities could cause errors or unexpected results, or cause programs or systems to behave unexpectedly. Vulnerabilities in programs can be detected and corrected through the debugging process. One way to perform debugging is to detect the area in the computer memory where information corresponding to the program is stored while the program is running in the computer. The information stored in the memory area can be retrieved as a file or a so-called "dump". Such a file is called a core dump file, and may also be called a crash dump file, checkpoint file, or snapshot file. A core dump...

AI Technical Summary

Problems solved by technology

[0004] However, a program's core dump file can be large, for example, over 100 megabytes (MB) or even over 1 gigabyte (GB)

Transferring a core dump file from a remote computer to the host computer consumes a lot of bandwidth and takes a long time to transfer

Images