How to do core dump analysis

UNIX for Advanced & Expert Users
1

Hi All,
I am new to unix environment.
Please tell me how to do coredump analysis. Please explain clearly with example. What are the details are available in the core.

Thanks in advance

2

You're asking a lot.

Basically, you use a debugger. Since you did not specify an OS I'll assume you have
gdb. You must have compiled the file

gcc -g -o myfile myfile.c

in order for symbols to be available. If you are analyzing a core dump of somebody else's code you are in trouble.

The core dump file is called core

gdb myfile core

This will show you a backtrace (stack dump) of the call tree that lead to the crash.
You will have to find using the stack dump where in the code (not in a C library) the crash occurred. In other words the last line of the program's code that actually led to the crash.

try man gdb
and
this tutorial site for gdb:

3

dbx debugger is used to analysis the coredumps ............. i will explain with anexample

cc -g first.c
dbx ./a.out

Reading a.out
Reading ld.so.1
Reading libc.so.1
Reading libdl.so.1
Reading libc_psr.so.1
(dbx 1) sif main
(2) stop infunction main
(dbx 2) run
Running: a.out
(process id 6156)
stopped in main at line 7 in file "first.c"
7 printf("%d\n",time(&t1));
(dbx 3) n
1117798709
stopped in main at line 8 in file "first.c"
8 for ( i=0 ; i<10; i++ ){
(dbx 4) n
stopped in main at line 9 in file "first.c"
9 printf("%d\n",rand());
(dbx 5) n
16838
stopped in main at line 8 in file "first.c"
8 for ( i=0 ; i<10; i++ ){
(dbx 6) exitbx 6)

4

Hi,
I think I have similar question, please bear me to put here.
I recently created a testing program containing only one thread to generate core file to analyze, but I couldn't use gdb to print stack trace, It says "cannot access memory ...".
Why? did I miss any command?
I did the test both on Linux and Solaris, same result.

tc11:/home/canzh/test> gcc -g devide.c

tc11:/home/canzh/test> ./a.out
Segmentation fault (core dumped)

tc11:/home/canzh/test> gdb a.out core
GNU gdb 5.0rh-5 Red Hat Linux 7.1
Copyright 2001 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "i386-redhat-linux"...
Core was generated by `./a.out'.
Program terminated with signal 11, Segmentation fault.
Cannot access memory at address 0x40016bec
#0 0x400a68d6 in ?? ()
(gdb) bt
#0 0x400a68d6 in ?? ()
Cannot access memory at address 0xbffff8a8
(gdb) where
#0 0x400a68d6 in ?? ()
Cannot access memory at address 0xbffff8a8
(gdb) quit
tc11:/home/canzh/test> uname -a
Linux tc11 2.4.2-2 #1 Sun Apr 8 20:41:30 EDT 2001 i686 unknown

5

try

gdb a.out
gdb> r

you mentioned threads, is your signal handler part of the thread you create?

6

Thanks for yor reply.
Yes, that started the program. This is my testing program, but if I have a real program, I cannot reproduce the problem easily by simply re-run it because there could be thousands of inputs and I don't know where and how it can crash. The purpose of running gdb is to find the problem (postmortem analyze).