These instructions should work fine on any CentOS [...]]]> Here is some advice I found useful when setting up my RPM build environment, and building a basic RPM.  The main source I used, to save reading my book again or the man pages, was Linc Fessenden’s blog and some of Linc’s blog is repeated here for completeness. Thanks Linc!

These instructions should work fine on any CentOS / RHEL / Fedora system.

First off, we need the rpm-build package to be installed. Check and install if needed.
yum install rpm-build

Login to the system as your user account, then make the following directories:


mkdir -p ~/rpm
mkdir -p ~/rpm/BUILD
mkdir -p ~/rpm/RPMS
mkdir -p ~/rpm/SOURCES
mkdir -p ~/rpm/SPECS
mkdir -p ~/rpm/SRPMS
mkdir -p ~/rpm/tmp


And create an ~/.rpmmacros file with the following in it:


%packager Your Name
%_topdir /home/YOUR USERNAME/rpm
%_tmppath /home/YOUR USERNAME/rpm/tmp


Now you need to create a package source code ditectory in the ~/rpm/SOURCES directory. You should name it package name – major revision number. Eg: ~/rpm/SOURCES/robspackage-1. In that directory place all the files that you wish to package in the RPM. I have put “script.sh” in mine.

Once that is done, make a tarball of that directory in the ~/rpm/SOURCES directory named programname-revision.tar.gz. Eg:

cd ~/rpm/SOURCES
tar -czvf rob-1.tar.gz rob-1/


In the ~/rpm/SPECS directory, create a packagename.spec file for your package.
Eg: rob.spec:

Summary: My first rpm package
Name: rob
Version: 1
Release: 1
Source0: rob-1.tar.gz
License: GPL
Group: CustomGroup
BuildArch: noarch
BuildRoot: %{_tmppath}/%{name}-buildroot
%description
Relevant package description
%prep
%setup -q
%build
%install
install -m 0755 -d $RPM_BUILD_ROOT/opt/rob
install -m 0755 script.sh $RPM_BUILD_ROOT/opt/rob/script.sh
%clean
rm -rf $RPM_BUILD_ROOT
%post
echo ” ”
echo “This will display after rpm installs the package!”
%files
%dir /opt/rob
/opt/rob/script.sh
%changelog
* Wed Feb 24 2010 Rob Jervis
- added something or fixed a bug

* Tue Feb 23 2010 Rob Jervis
- First RPM package of the rob application for EL5

Direct Quote from Linc:
“The install lines tell rpm what to install where and with what permissions. You also have to do any directory creation there as well (the one with the -d in the line).”

“The things after %file are similar in that this tells rpm’s database which files are attached to this package. The %dir signifies a new directory, otherwise the files are listed with their complete paths.”

Now you need to create the package, go to ~/rpm and do “rpmbuild -ba SPECS/rob.spec”.
If your package builds ok, it will end up in the RPMS directory, in this case: ~/rpm/RPMS/noarch/rob-1-1.noarch.rpm.

If it fails to build, first check your spec file carefully. Enjoy!

I thought I would share it, place in “/etc/init.d/dnetc”. It should work fine on current releases of Fedora, RHEL and CentOS. Its designed for the sysadmin to configure dnetc for that [...]]]> This week I was looking into init scripts again, its been a while… Anyhow, I remembered that I had written one for distributed.net’s client app.

I thought I would share it, place in “/etc/init.d/dnetc”. It should work fine on current releases of Fedora, RHEL and CentOS. Its designed for the sysadmin to configure dnetc for that system and then manage its start up and shutdown. (Not a script for end users to be using. )

You will need to change the application path to be the directory you have extracted the into.


#!/bin/bash
#
# dnetc This shell script takes care of starting and stopping distributed.net client
#
# chkconfig: 345 90 12
# description: distributed.net client program, a \
# distributed computing project. The program \
# uses only the computers idle time.
# processname: dnetc

# config: /etc/dnetc/dnetc.ini
# pidfile: /var/run/dnetc.pid

# Get function from functions library
. /etc/init.d/functions

start() {
echo -n “Starting DNET client: ”
/custom/dnetc/dnetc -quiet
touch /var/lock/subsys/dnetc
success $”DNET client startup”
echo
}

stop() {
echo -n “Stopping DNET client: ”
killproc dnetc
rm -f /var/lock/subsys/dnetc
echo
}

# —————————————————————–
case “$1″ in
start)
start
;;
stop)
stop
;;
status)
status dnetc
;;
restart|reload|condrestart)
stop
start
;;
*)
echo $”Usage: $0 {start|stop|restart|reload|status}”
exit 1
esac

exit 0

This script could do with some more work, I did put it together quickly, but this page helped me get it going and its worth looking at if you want to write a System V init script of your own.

(This works on RHEL/CentOS, [...]]]> Do you run a squid proxy? I do, and have been restarting squid to apply config changes for ages. Turns out that you can just run “service squid reload”, users have no down time and your config changes get applied. Great. Wish I had thought to check that before now…..

(This works on RHEL/CentOS, but any install of squid is capable of this via one command or another)