LAN - static IP on fedora 10

It is pretty easy task to set a static ip on LAN in fedora 10 machine. 

First go to network (System -> Administration -> Network) and double click on the device listed in it. Select "Statically set IP address" and enter required fields. In my machine I have used,

Address : 192.168.1.XX

Subnet mask:255.255.255.0

Default gateway address:192.168.1.1

At this point I was unable to connect to internet. By setting fields in DNS tab I was able to connect to internet, but reset them when machine reboots. So I enter those fields in /etc/sysconfig/network-scripts/ifcfg-eth0. It is as follows,

NETMASK=255.255.255.0

GATEWAY=192.168.1.1

DNS1=XX.XX.XX.XX

DNS2=XX.XX.XX.XX

Finally, I was able to set a static ip and connect to internet.

Binary Search Algorithm

Over the weekend I read some chapters of Beautiful Code book. It is really interesting and I am seeing the beauty in programming :). I read the Beautiful Test chapter and impressed with the example it has introduced. It is Binary Search algorithm. It says that this algorithm was published in 1946 but took 12 years for the first binary search without bugs to be published. It is amazing!!!
You can refer the most common bug here.

Further it shows that the algorithm may not returns the first position of the target value. But for the binary search algorithm, only pre-condition is to have a sorted array[1]. A sorted array can have duplicate values. If so algorithm should return the first position, as I guess. But the algorithm published in most sites(in wikipedia too) do not consider this point.
Here is the changed algorithm to handle duplicate values,

public static int binarySearch(int[] array, int target){
int low = 0;
int high = array.length - 1;
while(low <= high){
int mid = calculateMidPoint(low, high);
int midValue = array[mid];

if(midValue < target)
low = mid + 1;
else if(midValue > target)
high = mid -1;
else{
while(mid!=0 && target == array[mid-1]) //loop backward to check whether duplicate values present
mid -= 1;
return mid;
}
}
return -1;
}

Here is my java implementation of this algorithm and test classes. My reference is Beautiful Code book.

[1] http://en.wikipedia.org/wiki/Binary_search

Limit the number of rows in Oracle and MySql

Say we want to select 10 records for a perticular query, we can use the following syntax

// mysql 

select column from table limit 10;

// Oracle 

select column from table where rownum <= 10;

 

If we want to select number of records from the results set in Oracle,

// Oracle 

select * from (select a.*, rownum rnum from (select column from table) a where rownum <=10) where rnum >= 5;

Enable wireless on fedora 10

This article guided me to connect to wireless network from fedora 10
http://forums.fedoraforum.org/showthread.php?t=193555

Install vlc player in fedora 10

I installed vlc player in my fedora partition. It consumes time to search plugins for other players like totem.
Use this link to install vlc player using yum
http://www.videolan.org/vlc/download-fedora.html

A Python Sidebar for Mozilla

When I was doing my Google summer of code in python I had so google search for many times to complete the project. Since, I did it in Python language and I was not familiar, I learn the basics through the Internet. But I couldn't find a place where I could gather all the information. Today I found a handy sidebar for python resources for Mozilla.

http://www.edgewall.org/python-sidebar/

Trac WebAdmin Plugin

I was able to enable trac webadmin plugin to the trac system. Here are the simple steps,

first need to install easy_install utility

wget http://peak.telecommunity.com/dist/ez_setup.py
python ez_setup.py

I got the following error when I try to install it,

unable to open /usr/lib/python2.5/config/Makefile

then I install python-devel using yum

yum install python-devel

then I install webadmin trac plugin,

svn co http://svn.edgewall.com/repos/trac/sandbox/webadmin/
cd webadmin
python setup.py egg_info
python setup.py bdist_egg
cd dist
easy_install *.egg

then edit the trac.ini file which is in ${trac_home}/conf/trac.ini

[components] webadmin.* = enabled

install AccountManager Plugin,

svn co http://trac-hacks.org/svn/accountmanagerplugin/0.10 accountmanagerplugin
cd accountmanagerplugin
python setup.py bdist_egg
cd dist
easy_install *.egg

add following lines to trac.ini

[components] trac.web.auth.LoginModule = disabled acct_mgr.* = enabled  [account-manager] password_format = htpasswd password_file = /path/to/password.file

Finally restart the web server and made the plugins available from web.

/sbin/service httpd reload

Make a trac backup

I was able to make a backup of trac database and import all the data to other machine. So all the tickets, milestones, etc. were imported to another machine. Here is the procedure.

Make a backup of the current database.

mysqldump -u charith -p --databases trac > dbdump.sql

Then I copy it to the target machine and import it and create a new database called "trac" and allow permission to user charith,

grant all privileges on trac.* to charith@localhost identified by 'password';

I tried to use mysqlimport command,

mysqlimport trac --replace --user=root dbdump.sql

mysqlimport: Error: Table 'trac.dbdump' doesn't exist, when using table: dbdump

But I was able to import it successfully as root by,

mysql trac < dbdump.sql

But, still trac give the following error: "The 'repository_dir' has changed, a 'trac-admin resync' operation is needed".
So I synchronize the repository by,

trac-admin /path/to/env/ resync

Finally, I was able to import the data to the new environment.

Install trac - project management tool

I spent few days to install trac software in my machine. I learnt lot of things while installing this software. Here is the procedure I followed,

1) Start Web server

First it need to confirm web server is running. We can start is as a service by,

System -> Administration -> Services

enabling the httpd and starting it.

Then can test it is running by browsing it using, http://localhost

To allow acces to the outsiders, need to enable webserver as a trusted service in firewall,

System -> Firewall

And need to enable port forwarding in the router which is depend on the router type.

Then you can get the public assigned to the router by browsing the www.whatismyip.com

You should be able to access http://public_ip

2) Running a subversion server

install subversion using yum as root

yum install subversion

Create svn repository

svnadmin create /srv/svn/

make httpd can access all files in repository

chown -R apache.apache /srv/svn chmod g+s /srv/svn/db

then create password file

vim /srv/trac/conf/trac.htpasswd

then add user password to it

htpasswd -m /srv/trac/conf/trac.htpasswd charith

configure apache by editing /etc/httpd/conf.d/subversion.conf

add folowwing lines to it

    DAV svn    SVNPath /srv/svn    AuthType Basic    AuthName "svn"    AuthUserFile /srv/trac/conf/trac.htpasswd    Require valid-user  

restart httpd service. This service need to be running to be reload. If not it will out put as "failed"

/sbin/service httpd reload

then upload the code to the repository

svn import -m "intial upload" /home/charith/SvnTestfile:///srv/svn/SvnTest

Can test the subversion surver running properly by checking out the code to some other folder

svn co http://localhost/srv/svn/SvnTest

Make repository access by apache

chcon -R -h -t httpd_sys_content_t /srv/svn/

3) Installing trac

install trac using yum

yum install trac

create a directory to store trac project

mkdir -p /srv/trac

create new trac environment using trav-admin

trac-admin /srv/trac/test initenv

make folders own by apache

chown -R apache.apache /srv/trac

edit the file /etc/httpd/conf.d/trac.conf to point to new project environment

# Replace all occurrences of /srv/trac with your trac root below
# and uncomment the respective SetEnv and PythonOption directives.
    SetEnv TRAC_ENV /srv/trac/charith
   AuthType Basic
   AuthName "trac"
   AuthUserFile /srv/trac/conf/trac.htpasswd
   Require valid-user
    SetHandler mod_python
    PythonHandler trac.web.modpython_frontend
    PythonOption TracEnv /srv/trac/charith

restart httpd service. This service need to be running to be reload. If not it will out put as "failed"

/sbin/service httpd reload

Now you should be able to access the trac in local mahine using

http://localhost/cgi-bin/trac.cgi

If you have enable SELinux there are few things to do, I used following commands to enable web server to access trac and subversion repositories.

chcon -R -t httpd_sys_content_t /srv/trac chcon -R -t httpd_sys_content_t /srv/svn

Easiest thing is to view the SELinux warning message from the top pannel. It contains the command to configure the SELinux.