insmod: ERROR: could not insert module / Invalid module format

Build kernel with this config off and feel free to load new kernel module without recompiling kernel

# CONFIG_MODULE_FORCE_UNLOAD is not set
# CONFIG_MODVERSIONS is not set
# CONFIG_MODULE_SRCVERSION_ALL is not set
# CONFIG_MODULE_SIG is not set
# CONFIG_MODULE_COMPRESS is not set

Download this Kernel_build.config

Mount QCOW2 image

QCOW_IMG=$1
NBD_DEV=/dev/nbd0

modprobe nbd
qemu-nbd -c $NBD_DEV $QCOW_IMG

#Find start byte (offset) of a partition
parted $NBD_DEV -s unit B print

#Associate that offset with a loop dev
losetup -o 525336576 -f /dev/nbd0
#Mount loop device to a directory

gdb: source indicator goes out of order

Problems:
If program is compiled with option "-O2", gdb might not recognize order of source lines correctly and the current statement  would point to incorrect line.

Solution:
Change "-O2" to "-O0"

http://www.linuxquestions.org/questions/programming-9/gdb-doesn%27t-provide-the-correct-line-in-the-source-code-858155/

Dual boot: windows 7 & centos

menuentry "Windows 7" {
	set root=(hd0,1)
	chainloader +1
}

grub2-mkconfig -o /boot/grub2/grub.cfg

C pointer variable

What is the difference between int* n and int *n

If int *n, that is a fallacy that there are 2 type of variables. In fact, variable is a place holder of data only, the variant is the type of data it holds.

int is data structure of 4 bytes while int* is the one of 8 bytes (for 64 bit system).

#include <stdio.h>
#include <stdlib.h>
#include <sys/socket.h>
#include <sys/types.h>
#include <netdb.h>
#include <netinet/in.h>
#include <errno.h>
#include <string.h>

int main(int argc, char* argv[]){
 int fd = socket(AF_INET, SOCK_DGRAM, 0);
 unsigned int zero = 300;
 
 // building address structure 
 struct in_addr* ipaddr = malloc(sizeof(struct in_addr));
 if(ipaddr == NULL){
  perror(strerror(errno));
 }

 zero = *((unsigned int*)ipaddr);
 printf("size of struct in_addr*: %d\n", sizeof(struct in_addr*));
 printf("size of int*: %d\n",sizeof(int*));
 printf("%x\n",zero);
 printf("ipaddr = %x",*((int*)(ipaddr)));
 
 return 0;
}

Populate a whole-root directory structure

Using yum utility, we can populate a directory structure that holds bunch of software utilities. This structure could be later used as changed root environment (jail). The following command ran on Centos 6:

$ yum install --releasever=6  --installroot=/absolute/path/to/dir <package_name>


Linux software RAID-1

Recover RAID-1 from the disk that survived disaster

$ mdadm --create /dev/md0 --level=1 --raid-disks=2 missing /dev/sdc 

where /dev/sdc is the surviving disk

Color on bash shell prompt

vim ~/.bashrc



COLOR1='\[\033[38;5;34m\]'

COLOR2='\[\033[38;5;161m\]'

BOLD='\[\033[1m\]' # Work only if "allow bold text" setting is enabled

STOP='\[\033[0m\]'

PS1="[${BOLD}${COLOR1}\u${STOP}@${BOLD}${COLOR2}\h \W${STOP}]\$ "

For more

Postgresql: getting started

# service postgresql initdb
to create /var/lib/pgsql/data which contains pg_hba.conf

createdb

set password for postgres user

psql -h -d -u -W

crond edit

Format: minute hour day month weekday /path/to/script

Cron table takes time value as time point not interval.

Use * to ignore value.
# Run script at 4:03
3 4 * * * script

Use / to  specify interval
# run script every 14 minute
*/14 * * * * script

VIM: some useful settings

set laststatus=2
set ruler
set cursorline
set hlsearch
set nowrap
set expandtab
set tabstop=4
set mouse=a

filetype plugin on

highlight CursorLine cterm=NONE ctermbg=24

Bash: special redirection character

Character	Action
>	Redirect standard output
>&	Redirect standard output and standard error
<	Redirect standard input
>!	Redirect standard output; overwrite file if it exists
>&!	Redirect standard output and standard error; overwrite file if it exists
|	Redirect standard output to another command (pipe)
>>	Append standard output
>>&	Append standard output and standard error

VIM text editor

Put permanent settings in  ~/.vimrc (create one if not existed)

# enable syntax highlight
syntax on


# bottom status
set laststatus=2






set ruler

SSHFS mount does not allow other users to access

Append the following option:
sshfs ... -o allow_other

Login via serial console in linux system

When a user want to log in linux system via physical serial port (eg ttyS0), a login program (agetty or mingetty) should have already been waiting for user input (username and password) on that port.
To make the login program ready on serial port, append the following line in /etc/inittab:
1:23:respawn:/sbin/mingetty ttyS0

The next step is to add parameter console=ttyS0 at kernel boot command in GRUB, the kernel will expose interaction on that serial port so that user can login.

C++ Variable of reference

#include <stdio.h>

struct point{
    int x;
    int y;
};

typedef struct point Point;
int main(){
    
    Point p1;
    Point &px = p1;
    
    p1.x = 10;
    p1.y = 20;
    
    Point p2 = p1;
    
    px.x = 11111;
    px.y = 1122;
    
    printf("Point 1: %d %d", p1.x, p1.y);
    printf("Point 2: %d %d", p2.x, p2.y);
    printf("There are two distinct object: Point 1 and 2");
}

psql: FATAL: Ident authentication failed for user "username" Error and Solution

"By default Postgresql uses IDENT-based authentication. All you have to do is allow username and password based authentication for your network or webserver. IDENT will never allow you to login via -U and -W options. Append following to allow login via localhost only:" (from http://www.cyberciti.biz/faq/psql-fatal-ident-authentication-failed-for-user/)

To access database via tcp connect with username/password, comment this line:
#host    all         all         127.0.0.1/32          ident

Not getting automatically web.xml file while creating servlet in eclipse

Answer

Make Centos listen to ACPI signal

yum install acpid
chkconfig acpid on


When you press power-off button on chassis, an ACPI signal is generated. If OS have acpid daemon listen for that kind of signal, shutdown command will be issued instead of sudden power-off ( in case of power outage).

Centos: make on-boot startup script

This post guide how to make script that start up at boot time on CentOS.

Write a script named, e.g launchd, that follows the pattern below:

Sample script 

#----------------------------------------------------------
#!/bin/bash

# name of service
# chkconfig: 345 90 90
# description: describe what your script would do

start(){
   # create a lock file
   touch  /var/lock/subsystem/launchd

   # tasks to do on boot
}


stop(){
    # remove the lock file
    rm -f /var/lock/subsystem/launchd
   # tasks to do to stop scripts
}

case "$1" in
        start)
                start
        ;;

        
        stop)
                stop
        ;;

*)
esac
# ------------------------------------------------------------------------------------

Elaboration

This script must implement at least start() for the system to call.
The line that starts with # chkconfig  is required to let chkconfig utility know which levels to run this script, start and stop priority.

Put this script in /etc/init.d directory.
Run: chkconfig --add launchd so that chkconfig can recognize your script.
Run: chkconfig launchd on to enable the script at some run-levels. 

Lockfile

Lockfile lets the system know whether there has been a running instance of this script. When the system starts up your script, it checks if lock file exists or not so that it won't have to run the script again and similar for stop(). This is why the function stop() won't be called if you haven't created a lockfile earlier.
If you look at other script in /etc/init.d, you will find all scripts have lock file created in start() and removed in stop().