The following is a brief step-by-step on how I've been multi-booting different operatings systems on my machines for the past few years.

It's not necessarily the easy way to do things. It is the way I first learned and the way that I'm most comfortable with using. I'm posting it here, as I believe it may be helpful to some Linux Explorers out there. This tutorial is based on GRUB Legacy. I do not use the new GRUB2 on my machines.

Here we go...

I have multiple drives on my system, so multi-booting becomes a bit more interesting. Currently, I have three hard drives installed. They are /dev/sda, /dev/sdb, and /dev/sdc. We won't concern ourselves at all with /dev/sdc because it is primarily used as a common storage/backup drive. My operating systems are installed on /dev/sda and /dev/sdb.

I use /dev/sda (1st boot device in my BIOS) as the drive that has my primary operating system and my secondary operating system on it. I break my drives down into numerous partitions because I always install GNU/Linux operating systems on at least two partitions; /(root) and /home. I also have a common /swap partition on each drive that I plan on installing GNU/Linux distributions on.

My /dev/sda partitions look like this:

/dev/sda1   *       16065    51215219    25599577+  83  Linux <-- Slackware /(root)
/dev/sda2        51247350   153645659    51199155   83  Linux <-- Slackware /home
/dev/sda3       153661725   488392064   167365170    5  Extended
/dev/sda5       153661788   204860879    25599546   83  Linux <-- Arch /(root)
/dev/sda6       204877008   307275254    51199123+  83  Linux <-- Arch /home
/dev/sda7       307307448   311403959     2048256   82  Linux swap / Solaris
/dev/sda8       311420088   488392064    88485988+  83  Linux <-- Archives

My /dev/sdb partitions look like this:

/dev/sdb1   *       16384    51216383    25600000    7  HPFS/NTFS/exFAT <-- MS Windows XP
/dev/sdb2        51232768   102432767    25600000    7  HPFS/NTFS/exFAT <-- MS Windows Programs
/dev/sdb3       102453181   488270159   192908489+   5  Extended
/dev/sdb5       102453183   106550639     2048728+  82  Linux swap / Solaris
/dev/sdb6       106565823   137289599    15361888+  83  Linux <-- CentOS tester 1 /(root)
/dev/sdb7       137304783   188501039    25598128+  83  Linux <-- CentOS tester 1 /home
/dev/sdb8       188516223   219239999    15361888+  83  Linux <-- Foresight tester 2 /(root)
/dev/sdb9       219255183   270451439    25598128+  83  Linux <-- Foresight tester 2 /home
/dev/sdb10      270466623   301190399    15361888+  83  Linux <-- Debian tester 3 /(root)
/dev/sdb11      301205583   352401839    25598128+  83  Linux <-- Debian tester 3 /home
/dev/sdb12      352417023   383140799    15361888+  83  Linux <-- vacant tester 4 /(root)
/dev/sdb13      383155983   434352239    25598128+  83  Linux <-- vacant tester 5 /home
/dev/sdb14      434367423   488270159    26951368+  83  Linux <-- common storage

You get the above information on your drives using the fdisk -l command as root from the command line.

Alright then, let's say that I now want to install Fedora Core Linux on my tester 4 partitions of /dev/sdb. Here's how I would do that step-by-step:

1) Begin the Fedora installation setting /dev/sdb5 as swap, /dev/sb12 as /(root), and /dev/sdb13 as the /home partition from Fedora's installer.

2) When the Anaconda installer reaches the point where it asks where I want to install the GRUB bootloader, I choose to SKIP installation of GRUB.

3) Once the install is finished, I reboot my main operating system (Slackware) and from the command line I mount the newly installed Fedora /(root) partition:

root_Slackware/home/vtel57:# mount /dev/sdb12 /mnt/tmp

*Note: I have a /mnt/tmp directory in my Slackware. If you don't have a /mnt/tmp, just mount in /mnt. You'll be fine.

4) I list the contents of the Fedora /boot directory:

root_Slackware/home/vtel57:# ls /mnt/tmp/boot

5) I copy down the vmlinuz and initrd information. You'll be looking for the vmlinuz-<kernel version> and the initrd-<kernel version>.img files specifically. You'll need that information for the next step.

6) I edit my /boot/grub/menu.lst file to add the new Fedora Core entry. Note: my MBR-controlling GRUB is in my secondary OS, Arch Linux. I can still mount and modify it from within Slackware, though.

root_Slackware/home/vtel57:# mount /dev/sda5 /mnt/tmp

root_Slackware/home/vtel57:# vim /mnt/tmp/boot/grub/menu.lst

I'll add this entry to the file:

# Fedora Core

title          fedora (t4)

root         (hd1,11)

kernel     /boot/vmlinuz-<kernel version> root=/dev/sdb12 ro

initrd       /boot/vmlinuz-<kernel version>.img

7) I reboot and choose "fedora (t4)" from the GRUB menu to boot my new Fedora installation.

That's it, folks. Straightforward, but not necessary the easiest way to do this. That's the beauty of GNU/Linux. You can do it your way.

Have fun!

~Eric

P.S. I'm still working on the next shell scripting lesson. It's going to be about variables and parameters, a necessary lesson for further BASH learning. Unfortunately, it's also kinda' complicated, so it's taking me a bit to figure out how to formulate a lesson and still stay loyal to my primary goal here... keeping it as simple as possible. Stay tuned...

*Originally published on my Nocturnal Slacker v1.0 blog at Wordpress.com
  

If you use Amazon Web Services EC2 for virtualising your infrastructure, you may find that once you have a few instances started, that it may be hard (especially if you aren't using a configuration management tool) to trace back 'which machine belongs to what security group?' or 'which AMI did I use to bring up this instance?.'

Well theres a few quick tricks you can use to get some info out of your running instances. To do this you need to access the instances meta-data, and this is available simply by using 'Curl' from your running Linux instance, as follows:

'curl http://169.254.169.254/latest/meta-data/<metadata-object>'

here are some of the metadata objects:

ami-launch-index

ami-manifest-path

block-device-mapping/

hostname

instance-action

instance-id

instance-type

kernel-id

local-hostname

local-ipv4

placement/

profile

public-hostname

public-ipv4

public-keys/

reservation-id

security-groups

ami-manifest-path will tell you from where on AWS the AMI (the Amazon Machine Image for the linux virtual machine) was launched. If you rolled/composed your own AMI this helps you match up which Bucket on S3 relates to your AMI.

for example:curl http://169.254.169.254/latest/meta-data/ami-manifest-path

public-ipv4 is the world-facing IP address of the running virtual machine.

If your not sure from looking at the AWS Dashboard (or dont have access to it) you can run:

curl http://169.254.169.254/latest/meta-data/instance-id

this will help you make sure if you want to terminate an instance that you kill off the correct one.

Its easy to experiment with, and the fact that you use curl to access the metadata information, it means that its also something thats pretty easy to compose a shell script for commonly needed tasks.

Hope someone finds this helpful.

We are pleased to announce our openSUSE Weekly News #172

In this Issue:

• Evergreen needs you
• Evergreen & Tumbleweed status update
• First openSUSE Conference 2011 Program Commitee Meeting
• Installing latest Intel graphics driver to openSUSE 11.4
• and many more ...

You can download it there:

• Weekly News in PDF Format
• Weekly News in HTML (Bento Format)
• Weekly News in HTML (Susebooks Format)

We hope you enjoy the reading :-)

If you want to help us collecting interesting articles for the openSUSE Weekly News, so you can all your stuff into our new ietherpad: http://os-news.ietherpad.com/2.

Found Bugs? Please place it in our Bugtracker: http://developer.berlios.de/bugs/?group_id=12095

Features, Ideas and Improvements can placed in our Featuretracker: http://developer.berlios.de/feature/?group_id=12095

Older content can be found there.

I posted this on my own external blog, but thought I would share the pics of the article in Linux Format that mentions the Linux Foundation!

This years Top Ten Linux Guru’s as judged by the Linux Foundation where picked earlier last month. Imagine my surprise to find my name (Matt Palmer) amongst them when I opened this months (UK June Edition) of Linux Format!!

I thought I would put up a couple of pictures of the article for prosperity. Well done to all the other guru’s as well, who have done a brilliant job of helping new users to Linux, as well as assisting others with questions,whitepapers and best practices, and inspiring innovation and development within the organisation.

Here were the Top Ten contributors to the Foundation:

1)Matthew Fillpot

2)Aaron Aceves

3)Andrea Benini

4)Shane

5)Istimsak Abdulbasir

6)Marco Fioretti

7)Matt Palmer (Me!)

8)Per Lindholm

9)GoinEasy9

10)Adam

Take a look at the Linux Foundations Website linux.com for some great articles and topics for debate, and the latest news on the greatest operating system on the planet!

Blimey!

I thought this was amazing. This nifty phone runs on Android, and has the ability to integrate into a Tablet or as a slimline PC with a real keyboard.

Its amazing take a look. It includes bluetooth headset that you can use if you have the phone integrated into the tablet when someone calls you.

http://www.compufon.com/features

This could prove to be an interesting competitor to other Smartphone vendors when this reaches production stage, as it is basically the iPhone/iPad and Laptop all in one. Because its Android, it also opens the door to allow enthusiasts to develop a wider range of applications,etc. 

Summary

I recently did a migration from Bugzilla 3.0 running on one server to Bugzilla 4.0 running on a new server.  Since I was already writing these down as I went, I thought that I would share these instructions with the greater Linux community.  You may find that some of these instructions may need to be run in a slightly different order, but this is a pretty complete account of what it took to get this going on the new server.  I know that I had to periodically go back the the ./checksetup.pl script over and over throughout the process to check myself and figure out what needed to occur next.  I tried to arrange all of these in an order that will work, but if you have to install the bugzilla folder first and continually re-run ./checkseutp.pl, it's pretty much what I had to do.

These instructions are for RedHat 5.5.

Backup Your Current Bugzilla 3.0 Database

These directions all assume that you have a Bugzilla 3.0 database that you've backed up with teh following commands: 

sudo mkdir /data/backups/mysql-bugzilla-3.0

sudo chown www-data.www-data /data/backups/mysql-bugzilla-3.0

mysqldump -u bugs --password=XXXXXXXXX bugs | gzip -9v > `date '+/data/backups/mysql-bugzilla-3.0/bugs_%Y%m%d.sql.gz'`

Setup Apache

Install the Web Server, set the run levels, and start the service:

[root@mybugzilla]# yum groupinstall "Web Server"

[root@mybugzilla]# yum chkconfig --levels 2345 httpd on

[root@mybugzilla]# service httpd restart

Try the main site to ensure that apache is setup correctly by browsing to your site name or to http://localhost

You should see the default apache site showing that the webserver is installed properly.

Modify the httpd.conf file

Open the /etc/httpd/conf/httpd.conf file in your favorite editor.  Add these lines to the end of your httpd.conf file:

PerlSwitches -w -T

PerlConfigRequire /var/www/html/bugz/mod_perl.pl

Also make sure that you have the line "KeepAlive Off" per the bugzilla install instructions.

Restart Apache

[root@mybugzilla]# service httpd restart

Setup MySQL

Install MySQL, set the service to start on boot, and set the mysql root password:

[root@mybugzilla]# <b>yum groupinstall "MySQL Database"</b>

[root@mybugzilla]# <b>chkconfig --levels 2345 mysqld</b>

[root@mybugzilla]# <b>service mysqld restart</b>

[root@mybugzilla]# <b>/usr/bin/mysqladmin -u root password 'XXXX'</b>

Now login to mysql:

[root@mybugzilla]#  <b>mysql -pXXXX</b>

Once logged into mysql as root, create the bugs user and the database:

mysql> create user 'bugs'@'localhost' IDENTIFIED BY 'XXXpasswordXXX';

mysql> create database bugs;

mysql> show databases;

Restore data

These instructions assume that you followed the above instructions for exporting your bugzilla database from Bugzilla 3.0.

[root@mybugzilla]# cd /path/to/restore/file
[root@mybugzilla]# gunzip bugs_20110414.sql.gz
[root@mybugzilla]# mysql -u root -pXXXX bugs < bugs_20110414.sql

Grant Permissions

Login to mysql as above and run

mysql> grant all on bugs.* to 'bugs'@'localhost';

Fix MySQL Defaults in /etc/my.cnf

Edit the /etc/my.cnf file and add these lines under the [mysqld] section:

[mysqld]

# Allow packets up to 4MB

max_allowed_packet=4M

# Allow small words in full-text indexes

ft_min_word_len=2

This ensures that uploaded attachments to bugs is now 4MB instead of the default 1MB.

Restart MySQL

Restart the service to ensure any changes above are loaded:

service mysqld restart

Setup Bugzilla

Prep Bugzilla

Create the directories and download Bugzilla:

[root@mybugzilla]# cd /var/www

[root@mybugzilla]# mkdir bugz

[root@mybugzilla]# cd bugz

[root@mybugzilla]# wget http://ftp.mozilla.org/pub/mozilla.org/webtools/bugzilla-4.0.tar.gz

[root@mybugzilla]# tar xzvf bugzilla-4.0.tar.gz

[root@mybugzilla]# cd /var/www/html

[root@mybugzilla]# ln -s ../bugz/bugzilla-4.0 bugz

[root@mybugzilla]# cd /var/www/bugz/bugzilla-4.0

Install Bugzilla Dependencies

./checksetup.pl --check-modules

You will be told you should run several commands to install all the perl modules.  For example,
I was told to run these:

[root@mybugzilla]# /usr/bin/perl install-module.pl Digest::SHA

[root@mybugzilla]# /usr/bin/perl install-module.pl Date::Format

[root@mybugzilla]# /usr/bin/perl install-module.pl DateTime

[root@mybugzilla]# /usr/bin/perl install-module.pl DateTime::TimeZone

[root@mybugzilla]# /usr/bin/perl install-module.pl Template

[root@mybugzilla]# /usr/bin/perl install-module.pl Email::Send

[root@mybugzilla]# /usr/bin/perl install-module.pl Email::MIME

[root@mybugzilla]# /usr/bin/perl install-module.pl List::MoreUtils

or just simply:

  [root@mybugzilla]# /usr/bin/perl install-module.pl --all

  DateTime won't build so I'm trying this:

  [root@mybugzilla]# yum install mysql-devel gd gd-devel perl-DBD-MySQL mod_perl-devel

    (per instructions at http://www.bugzilla.org/docs/4.0/en/html/installation.html)
  The perl-DBD-MySQL package is 3.07 which is too old (4.0 is needed) so I had to be sure to
run the cpan install:

[root@mybugzilla]# /usr/bin/perl install-module.pl DBD::mysql

  I noticed that DateTime complains that I only have Archive::Tar 1.3901 which it found in the
perl-Archive-Tar package.  I upgraded with the following:

[root@mybugzilla]# /usr/bin/perl install-module.pl Archive::Tar

  Which got me Archive::Tar 1.76, which stopped the complaints in DateTime.  I installed datetime
running these:

[root@mybugzilla]# /usr/bin/perl install-module.pl Module::Build

[root@mybugzilla]# /usr/bin/perl install-module.pl DateTime

Install Additional Modules

I installed these extra modules:

[root@mybugzilla]# /usr/bin/perl install-module.pl Net::LDAP

[root@mybugzilla]# /usr/bin/perl install-module.pl GD

[root@mybugzilla]# /usr/bin/perl install-module.pl Chart::Lines

[root@mybugzilla]# /usr/bin/perl install-module.pl Email::Reply

[root@mybugzilla]# /usr/bin/perl install-module.pl Apache2::SizeLimit

[root@mybugzilla]# /usr/bin/perl install-module.pl GD::Graph

[root@mybugzilla]# /usr/bin/perl install-module.pl PatchReader

[root@mybugzilla]# /usr/bin/perl install-module.pl Email::MIME::Attachment::Stripper

    I know that out of the above, if you don't have Apache2::SizeLimit then there will be a problem when restarting the httpd server, which (per the above instructions) includes a reference to the mod_perl.pl script in the bugzilla folder, which in turn uses Apache2::SizeLimit.  If you have trouble (as I did), make sure you have the =mod_perl-devel= package installed in order to install Apache2::SizeLimit.

Edit Configuration

Now you'll need to edit the localconfig file to setup the variables.  Make sure you have these variables set.  You'll probably have to manually set
those listed in bold:

$create_htaccess = 1;

$webservergroup = 'apache';

$use_suexec = 0;

$db_driver = 'mysql';

$db_host = 'localhost';

$db_name = 'bugs';

$db_user = 'bugs';

<b>$db_pass = 'XXXpasswordXXX';</b>

$db_port = 0;

$db_sock = '';

$db_check = 1;

$index_html = 0;

$cvsbin = '/usr/bin/cvs';

$interdiffbin = '/usr/bin/interdiff';

$diffpath = '/usr/bin';

$site_wide_secret = 'XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX';

Re-Run CheckSetup

Now re-run checksetup now that you've fixed the localconfig file.  This should test the settings in the localconfig to ensure the correct db name, user name and password:

./checksetup.pl --check-modules

The script may remind you to perform some actions documented above.  If you've missed anything, go ahead and fix it now before continuing.

Bugzilla Upgrade

Upgrading can cause serious harm to the database.  Make sure you've backed up or you have a file migrated from another server that you can use to restore the database 'bugs' table in case the following fail.  There are probably more ways to migrate, but these are the actions that I took.

Ensure the character encoding is correct

Your database must be in UTF-8.  If the previous database used another encoding, you should fix it with the recode.pl script.

[root@mybugzilla]# /usr/bin/perl install-module.pl Encode::Detect

[root@mybugzilla]# cd /var/www/bugz/bugzilla-4.0/

[root@mybugzilla]# contrib/recode.pl --dry-run --guess

Review what changes then do the command for real:

[root@csobugzilla]# contrib/recode.pl --guess

Continue with the checksetup.pl

Try the checksetup.pl script again.  You can now safely continue when prompted about the UTF-8 stuff.

References

http://www.bugzilla.org/docs/4.0/en/html/configuration.html
http://www.velikan.net/import-sql-dump-file-to-mysql-database/
http://dev.mysql.com/doc/refman/5.1/en

Introduction

This is something I used to need a lot, that I dug out of my documentation to share.  These are steps to using dsh on ubuntu to administer multiple workstations.  This does not include instructions for setting up public/private key authentication in ssh, but these instructions work the best when this has been setup.  If you don't setup ssh keys, you'll end up having to type the root password for each machine you ssh into.

Setup

Installation

On Ubuntu, run this to install:

sudo aptitude install dsh

Configuration

Start by creating a new group of workstations.  I have several CentOS workstations, so I create a new file that contains all of these:

/etc/dsh/group/centos-workstations

# These are standard workstations conforming to the typical centos configuration

root@host1

root@host2

root@host3

Running Commands

You can run pretty much anything, but here's an example:

dsh -g centos-workstations yum clean all
dsh -g centos-workstations yum -y upgrade

Some Shortcuts

I simplify matters even more by creating an alias

alias cent='dsh -g centos-workstations'

I add this to my .zshrc file.  This buys the the ability to run things like:

cent yum -y upgrade

Conclusion

I realize these instructions are a bit basic, but I hope it points out new possibilities to those who have never used dsh before.  I have found myself using this pretty extensively with about 50 workstations and publickey authentication and it works wonderfully.  I found it especially useful a year or more ago when I was trying to modify 50+ workstations to conform to a more standard configuration.  I tried several other solutions over a year ago (parallel ssh or something like that) and I didn't find anything quite as nice as dsh.

I'm one of the 150 people invited to the first "Bloggers in Vatican" meeting next month. I've put all details about the goals of the meeting and why I was invited here, and now I have a question:

Which questions and proposals do you think I may suggest for a meeting like that if I had the possibility, both in general and as a FOSS/Open Standards advocate?

Thanks! Any feedback is welcome.

WordPress is quick and easy to install and update, but the quicker you can make these operations the better, right? If you have shell access to the server where your WordPress copy is installed, it is possible to make some operations much quicker and safer with this shell script.

Just by looking at the amount of voluntary work users engage in to help each other with their computer problems I wonder if we could not use this massive source of human intelligence for educational purposes.

The idea is that we first aim high and help each other on entry level university mathematics. Young adults would then see a another real benefit to join the Linux community.

In my previous blog entry I wrote about the potential benefit from bundling student books with some popular distributions. I believe this where you need to start. Having a couple of good open textbooks for users to engage in. Perhaps its enough to have them in the repository. The import thing is to get enough users reading the same books.

I wrote the below blog entry explaining how the cooperation could work.

http://blogg.idg.se/perrabyte/2011/04/08/idea-how-to-make-swedish-students-much-better-in-mathematics/

Here I focus on explaining the textbook theory in multi fashionable way. What I forgot to write is that the explanations need to be approved before publishing and you can link the specific theoretical part to whatever is relevant be it videos, exercises or forum discussions for questions not answered.

We need advanced innovative software in all fields. By helping with the educational bit I think we get a lot in return.