Setting up Centos 5.5

spn

Put following in your /etc/rc.local file:

echo "CentOS release 5 (Final)" > /etc/issue
echo "Kernel \r on an \m" >> /etc/issue
echo "">> /etc/issue
echo "$(ifconfig eth0 | grep inet | cut -d : -f 2 | cut -d \ -f 1)" >> /etc/issue
echo "" >> /etc/issue

http://crazytoon.com/2008/04/22/linux-how-to-change-login-prompt-in-linux/
http://linuxhelp.blogspot.com/2005/10/change-system-login-banner-in-linux.html

spn

vi /etc/sysconfig/network

HOSTNAME=Cent.localdomain

#change to Cent

http://wiki.centos.org/FAQ/CentOS5

spn

etc/httpd/conf/httpd.conf

<Directory "/var/www/cgi-bin">
Options ExecCGI
AddHandler cgi-script .py
# still allow all script, not just .py :(
</Directory>

I think this should only allow .py script, but it allow any executable file.

Options ExecCGI

AddHandler cgi-script .py

</Directory>

As the above only allow .py script in /home/*/public_html/cgi-bin.

or I missed something :(

spn

edit /etc/httpd/conf/httpd.conf
add these section

<Directory "/home/*/public_html/cgi-bin">
Options ExecCGI
AddHandler cgi-script .py
</Directory>

Then restart the server
/usr/sbin/apachectl restart

If there is any error, check httpd error log

tail /var/log/httpd/error_log

[Mon Nov 08 10:21:05 2010] [error] [client 192.168.106.31] suexec policy violation: see suexec log for more details

[Mon Nov 08 10:21:05 2010] [error] [client 192.168.106.31] Premature end of script headers: ...py

error from suexec, check suexec log

tail /var/log/httpd/suexec.log

[2010-11-08 10:22:51]: uid: (500/s) gid: (500/500) cmd: ..py

[2010-11-08 10:22:51]: directory is writable by others: ..

The directory should not be writable by others (security reason)

chmod 711 /home/user-abc/public_html/cgi-bin

If there is another error, check both logs again.

spn

edit /etc/httpd/conf/httpd.conf

remove this line
UserDir disable

and uncomment this line
UserDir public_html

(or can just change the first line)

Then restart the server
/usr/sbin/apachectl restart

If SELinux is enabled (I did), you must also need to run these

/usr/sbin/setsebool -P httpd_enable_homedirs true
chcon -R -t httpd_sys_content_t /home/user-abc/public_html

(do not forget to change the path on the second line)

from Apache Userdir with SELinux on Fedora 13/14, CentOS 5.5, Red Hat (RHEL) 5.5/6

spn

Using telnet command
(assuming the ip address of your machine is 192.168.106.120 or you can use localhost)

telnet 192.168.106.120 80

Then type "abc" and Enter

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">
<html><head>
<title>501 Method Not Implemented</title>
</head><body>
<h1>Method Not Implemented</h1>
<p>a to / not supported.<br />
</p>
<hr>
<address>Apache/2.2.3 (CentOS) Server at localhost.localdomain Port 80</address>
</body></html>
Connection closed by foreign host.

Try using telnet again but this time type "GET" then Enter

Different output?

Now open a browser and put 192.168.106.120 as the address

Should see a lot of info (including below)

"...
If you are the website administrator:
You may now add content to the directory /var/www/html/..."

Now go into /var/www/html and create index.html containing this line
Hello from CentOS

Then refresh the page on the browser
(also telnet and see)

spn

Check the status of httpd

/usr/sbin/apachectl status
ELinks: Connection refused

It is not started, then start it :)

/usr/sbin/apachectl start

These can use to restart (after change its config)
/usr/sbin/apachectl restart
or
/sbin/service httpd restart

From
centos_doc
centos doc2

spn

After installing git (see this)
clone (copy) from kitenet.net

git clone git://git.kitenet.net/etckeeper

Then go into etckeeper and edit etckeeper.conf

# The VCS to use.
VCS="hg"
#CS="git"
#VCS="bzr"
#VCS="darcs"

# Options passed to git commit when run by etckeeper.
GIT_COMMIT_OPTIONS=""

# Options passed to hg commit when run by etckeeper.
HG_COMMIT_OPTIONS="-u etckeeper@a.com"

# Options passed to bzr commit when run by etckeeper.
BZR_COMMIT_OPTIONS=""

# Options passed to darcs record when run by etckeeper.
DARCS_COMMIT_OPTIONS="-a"

# Uncomment to avoid etckeeper committing existing changes
# to /etc automatically once per day.
#AVOID_DAILY_AUTOCOMMITS=1

# Uncomment to avoid etckeeper committing existing changes to
# /etc before installation. It will cancel the installation,
# so you can commit the changes by hand.
#AVOID_COMMIT_BEFORE_INSTALL=1

# The high-level package manager that's being used.
# (apt, pacman-g2, yum etc)
HIGHLEVEL_PACKAGE_MANAGER=yum

# The low-level package manager that's being used.
# (dpkg, rpm, pacman-g2, etc)
LOWLEVEL_PACKAGE_MANAGER=rpm

Then we can install and create an initial version.

make installl
cd /etc
/usr/sbin/etckeeper init
/usr/sbin/etckeeper commit -m "Initial"

spn

To add rpm from RPMforge

rpm --import http://apt.sw.be/RPM-GPG-KEY.dag.txt

wget http://packages.sw.be/rpmforge-release/rpmforge-release-0.5.1-1.el5.rf.x86_64.rpm

rpm -K rpmforge-release-0.5.1-1.el5.rf.x86_64.rpm

rpm -i rpmforge-release-0.5.1-1.el5.rf.x86_64.rpm

# if not using x64
# rpm -i rpmforge-release-0.5.1-1.el5.rf.i386.rpm

Now we can use
yum install mercurial
yum install p7zip
yum install git
yum install django
It still does not have etckeeper :(

more software installed
yum install lynx

from
http://wiki.centos.org/AdditionalResources/Repositories/RPMForge

spn

Start again :)

After re-installing yesterday (this time I set grub on the correct drive),

however it hanged at starting udev :(

I found this CentOS 5.0 hangs at boot

vi /etc/sysconfig/modules/udev-stw.modules

--> remove all modules : )

#!/bin/sh

MODULES="nvram floppy parport lp snd-powermac"

[ -f /etc/sysconfig/udev-stw ] && . /etc/sysconfig/udev-stw

for i in $MODULES ; do

modprobe $i >/dev/null 2>&1

done

This time, I picked only "Server - GUI" package.

I do not think this causes the problem.

Anyway I just saved the partition image using Partimage on SystemRescueCd.

Later I might try install using the same set of package (as I did earlier).

spn

I am starting to play with Centos 5.5 again this month.

Starting from 1 NOV 2010,

I downloaded CentOS-5.5-x86_64-netinstall.iso

then created a bootable USB drive, using Universal USB Installer

It cannot see iso files on FAT32 partition : (

so I have to chose HTTP install, from this server.

I selected these packages: Desktop+Server+Virtualisation

It took about 30 minutes.

I did not read *carefully* when it asked to install grub (boot loader).

So it was installed to the USB drive.

The result is that I can boot to Centos if the USB drive is plugged in.

: (

Try to fix by following http://www.sorgonet.com/linux/grubrestore/

root (hd0,1)

setup (hd0)

quit

But it does not work.

Finally I have to edit /boot/grub/menu.lst

I am not sure but I think just change

hd1 to hd0 in the file.

# NOTICE: You have a /boot partition. This means that

# all kernel and initrd paths are relative to /boot/, eg.

# root (hd1,8)

# kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00

# initrd /initrd-version.img

#boot=/dev/sdb

default=2

timeout=5

splashimage=(hd0,8)/grub/splash.xpm.gz

hiddenmenu

title CentOS (2.6.18-194.17.4.el5xen)

root (hd0,8)

kernel /xen.gz-2.6.18-194.17.4.el5

module /vmlinuz-2.6.18-194.17.4.el5xen ro root=/dev/VolGroup00/LogVol00

rhgb quiet

module /initrd-2.6.18-194.17.4.el5xen.img

title CentOS (2.6.18-194.el5xen)

root (hd0,8)

kernel /xen.gz-2.6.18-194.el5

module /vmlinuz-2.6.18-194.el5xen ro root=/dev/VolGroup00/LogVol00 rhgb

quiet

module /initrd-2.6.18-194.el5xen.img

title Other

rootnoverify (hd0,0)

chainloader +1

I do not know why, but I have to activate the network card on the first couple of times that it booted.

I can get http server to work, but not Python CGI.

Then I tried installing mercurial.

cannot use: yum install mercurial

The first time was older version - 1.3.1-3.el5

http://download.fedora.redhat.com/pub/epel/5/i386/epel-release-5-4.noarch.rpm

then a newer one - 1.6.4

http://packages.sw.be/rpmforge-release/rpmforge-release-0.5.1-1.el5.rf.i386.rpm

not sure if this site is safe :)

After messing it a little bit, I found that there is a configuration management software

http://docs.puppetlabs.com

can use: yum install puppet

But I do not think I have a time to learn and use it.

So I found etckeeper which can use mercurial for its vcs.

- Version Control for Linux Configuration (/etc) with etckeeper

- How to setup etckeeper with Mercurial in Ubuntu?

edit /etc/etckeeper/etckeeper.conf

HG_COMMIT_OPTIONS=" -u etckeeper@a.com"

cd /etc

sudo etckeeper init

sudo etckeeper commit "Initial import"

NOTE: The information here may not be accurate.

That is why I just re-installed yesterday.

Please continue on the next post

Setting up Centos 5.5

Tuesday, November 30, 2010

Show IP address before login

Change hostname

Sunday, November 7, 2010

Bug in Apache? Allow all script in /var/www/cgi-bin

Enable python CGI script in user directory (public_html/cgi-bin)

Enable user directory (public_html)

Saturday, November 6, 2010

Testing httpd using telnet

start httpd

Install etckeeper

Add RPMforge

Re-install

Starting ...