03-12-2002
Most core dumps that I make are caused by one of two things:
1. Not terminating strings, or processing strings (char arrays) outside the array bounds.
2. Passing a variable into a function that requires a pointer to that variable. i.e. use &var instead of var in the function call.
Give your code the once over to see if it is one of the above two.
Core dumps can be difficult to deal with because the program is stopped in its tracks and the point of failure might not be your bug - that is probably elsewhere.
If you don't have access to a debugger then, identifying the line of code the program fails on, will go a long way to solving the issue. Use a log file or printf statements to locate the point at which the program core dumps. It you get a log message, then you know your program is still alive. It shouldn't take too long to locate the problem.
10 More Discussions You Might Find Interesting
1. Solaris
I am trying to set up coredumps on a server, I have enabled the coreadm, and have also set a path to whre the coredumps will go.
I now need to set the ulimit , as this is set to 0. I made the change:
ulimit -c 8192
but when I logged out and back in, it was set back to 0 again. I need to know... (1 Reply)
Discussion started by: mzyvn6
1 Replies
2. Programming
haiu all
what makes strncpy to coredump (0 Replies)
Discussion started by: vijaysabari
0 Replies
3. Solaris
SunOS unknown 5.9 Generic_118558-10 sun4u sparc SUNW,Sun-Fire-880
I'm trying to set up mirroring and am running into a problem:
As you can see, I set up the metadb, but am getting a coredump when running metainit. I saw this through google, but don't know the validity of it. Any ideas?
#... (5 Replies)
Discussion started by: dangral
5 Replies
4. UNIX for Advanced & Expert Users
Every time I reboot our solaris 9 (SunOS 5.9 Generic_118558-19 sun4u sparc SUNW,Sun-Fire-480R) box,
I get the messages below:
# dmesg | grep dump
Mar 24 12:39:55 hostname savecore: initial dump header corrupt
Mar 24 12:39:55 hostname genunix: dump on /dev/dsk/c1t0d0s1 size 700 MB
Mar 24... (2 Replies)
Discussion started by: xnightcrawl
2 Replies
5. UNIX for Dummies Questions & Answers
This works for all my normal executions. But on one machine, a coredump is expected on one command. We don't care and know it will always happen, but need to confirm the version info from the start of the output. This is output from the exe:
$ ./Up
-*-*- XXXXXXX UPDATE XXXXX Version... (7 Replies)
Discussion started by: brdholman
7 Replies
6. Red Hat
Hi All,
I just wondering how can we read core files (core.****). I have an appication, when it is trying to crash I got the coredump. since it is in binary format, I dont know how to read that. Thanks (1 Reply)
Discussion started by: s_linux
1 Replies
7. AIX
Hello!
I do have an IBM pSeries 7043 server running AIX 5.2.0.0. Well, I do have trouble to run xsnaadmin tool remotely, I mean, the xsnaadmin tool can't run on exported X display.
What I'm doing is
aixserver>export DISPLAY=xxx.xxx.xxx.xxx:0.0
aixserver>xsnaadmin
Segmentation... (0 Replies)
Discussion started by: jssanche
0 Replies
8. Shell Programming and Scripting
Hi Gurus,
I tried bc 1000 % 10 on tcsh and ksh and its throwing a core dump on a sun solaris machine.
uname -a
SunOS azote 5.9 Generic_118558-39 sun4u sparc SUNW,Ultra-4
Please let me know if you find anything.
Thanks,
Kinny (8 Replies)
Discussion started by: kinny
8 Replies
9. HP-UX
Hi,
I'm looking for the help on how to implement coredump using the tools that comes with HP-UX 11.23 and HP-UX 11.31.
To also avoid the the dump files fill-up the root disk when dump occurs.
I'm new to this and would need step by step please.
Thanks in advance! (3 Replies)
Discussion started by: lamoul
3 Replies
10. Programming
hi
I have a method which returns char*. In this method am using switch case. I am getting segmentation error in case 49 and my code is
case 49:
if(intFlag == 0210)
{
iiIDCode = atoi(getsubstring(sReq,262,2));
l =... (7 Replies)
Discussion started by: AAKhan
7 Replies
CORE(5) BSD File Formats Manual CORE(5)
NAME
core -- memory image file format
SYNOPSIS
#include <sys/param.h>
DESCRIPTION
A small number of signals which cause abnormal termination of a process also cause a record of the process's in-core state to be written to
disk for later examination by one of the available debuggers. (See sigaction(2).) This memory image is written to a file named by default
core.pid, where pid is the process ID of the process, in the /cores directory, provided the terminated process had write permission in the
directory, and the directory existed.
The maximum size of a core file is limited by setrlimit(2). Files which would be larger than the limit are not created.
The core file consists of the Mach-O(5) header as described in the <mach-o/loader.h> file. The remainder of the core file consists of vari-
ous sections described in the Mach-O(5) header.
NOTE
Core dumps are disabled by default under Darwin/Mac OS X. To re-enable core dumps, a privileged user must do one of the following
* Edit /etc/launchd.conf or $HOME/.launchd.conf and add a line specifying the limit limit core unlimited
* A privileged user can also enable cores with launchctl limit core unlimited
* A privileged user can also enable core files by using ulimit(1) or limit(1) depending upon the shell.
SEE ALSO
gdb(1), setrlimit(2), sigaction(2), Mach-O(5), launchd.conf(5), launchd.plist(5), sysctl(8)
HISTORY
A core file format appeared in Version 6 AT&T UNIX.
BSD
June 26, 2008 BSD