Computer science Experience Knowledge SHAring (CEKSHA)

Git - alias to get my last commits

2012-01-30T21:12:00.001+01:00

While working with Git, there is lots of situation where it is interesting to give a quick list of your last commits.
For the current version you're working on, or for any other version.
But usually, you are not alone to commit and so you need to filter result of git log (on author for instance).

This is a very simple GNU/Bash script allowing to list a short form of your last commits (for instance create a file /path/to/gitListMyLastCommits.sh and copy/paste):

#!/bin/bash
#
# Author: Bertrand BENOIT
# Version: 1.1
# Description: lists last commit of current author according to specified (optional) branch and count.

# To use this script, create a git alias.
# git config --global alias.mc '!/path/to/gitListMyLastCommits.sh'

DEFAULT_COMMIT_COUNT=5
export GIT_PAGER=cat

# Manages command line options.
# - case 1: 2 specified arguments
# - case 2: 1 specified argument [-] (with optional negative sign)
# - case 3: no specified arguments -> work on current branch, with 5 last commits of author
if [ $# -gt 2 ] || [ $( echo "$@" |grep -cwE "(-h|--help)" ) -gt 0 ]; then
echo -e "usage: git "
echo -e "The branch can be ommitted"
echo -e "The commit count can be ommitted (default: $DEFAULT_COMMIT_COUNT will be shown)"
echo -e "If only one argument is given, it is regarded as commit count, so you can NOT only specify a branch, without a commit count"
exit 1
fi

# Ensures there is a negative sign for last argument, if needed.
if [ $# -eq 0 ]; then
arguments="-$DEFAULT_COMMIT_COUNT"
else
arguments=$( echo "$@" |sed -e 's/^$[0-9][0-9]*$$/-\1/;s/[ ]$[0-9][0-9]*$$/ -\1/' )
fi

authorInfo=$( git config --get user.email )
[ -z "$authorInfo" ] && echo "WARNING: unable to define author information, ensure you have configured user.email." >&2 && exit 1

echo -e "Looking last commits of author '$authorInfo', with arguments: $arguments"
git log --format='%h - %cd - %s' --date=short --author="$authorInfo" $arguments

echo -ne "\nOrdered commit short SHA1 chain (can be used with cherry-pick): "
git log --format='%h - %cd - %s' --date=short --author="$authorInfo" $arguments |awk '{print $1}' |sed -e 's/$/ /g' |tac |tr -d '\n'
echo ""

At the end, it will give a one-line list of SHA1 corresponding to reversed order of your last commit; it can then be easily copy/paste to cherry-pick.

You can add this script as a new Git alias:
git config --global alias.mc '!/path/to/gitListMyLastCommits.sh'

Then to get your last 5 commits on current version:
git mc

To get your 20 last commits on current version:
git mc 20
or
git mc -20

Finally to get 10 last commits of branch R1_0:
git mc R1_0 20

Migrate from SVN to Git

2011-12-14T19:01:00.001+01:00

There is lots of article about migration from SVN to Git (a powerful distributed version control system), but I didn't find all needed information in only one.
This is the aim of this article.

Instead of blabla, I take the concrete example of the migration of my open-source project Hemera :

- create an author file linking information between commiter name, user name, and user e-mail:

/tmp/authors

projettwk = Bertrand Benoit

- create a Git repository from SVN repository

git svn clone --no-metadata -s -A /tmp/authors https://hemerais.svn.sourceforge.net/svnroot/hemerais hemera

- update Git repository metadata, removing SVN information (more information)

git fetch . refs/remotes/*:refs/heads/*

git branch -rd 0.1 trunk

git branch -d trunk

git config --remove-section svn-remote.svn

rm .git/svn/ -Rf

- create a bare Git repository metadata (more information: here, and here)

git clone --bare /tmp/hemera/.git /tmp/hemera-bare.git

cd /tmp/hemera-bare.git

git config --unset remote.origin.url

- publish the fresh bare Git repository (more information)

git remote add origin ssh://XXX@hemerais.git.sourceforge.net/gitroot/hemerais/hemerais

git config branch.master.remote origin

git config branch.master.merge refs/heads/master

git push --all

- finally, get a local Git repository (with write access)

git clone ssh://XXX@hemerais.git.sourceforge.net/gitroot/hemerais/hemerais

Hemera v0.1.5 released

2011-12-14T18:30:00.002+01:00

Hemera is a Virtual Intelligent System aggregating some more advanced Artificial Intelligence Technologies (speech synthesis, speech recognition, form recognition, motion recognition ...); with applications in daily tasks, domotics and robotics.

Noteworthy changes in version 0.1.5:
* various great infrastructure updates for better flexibility and evolvability
* easy plugins system to add recognized commands
* new configuration auto-diagnosis system (optimal check)
* offline version of Hemera Book
* improved compatibility and robustness with RedHat-based and Debian-based distributions
* fixed all reported bugs in previous Hemera versions

The online documentation has been completely revised for improving legibility and accessibility.

To download release 0.1.5 of Hemera, information about upgrade from 0.1/0.1.1 to 0.1.5, or offline version of the Hemera book:
https://sourceforge.net/projects/hemerais/files/Hemera/0.1.x/

Git, a powerful Distributed Version Control System (DVCS)

2011-12-13T12:16:00.001+01:00

In last ten years, we've seen emergence of Distributed Version Control Systems (DVCS).
IMO, the noteworthy advantages of a Distributed Version Control System against a Centralized one are:

- each working copy can be regarded as a backup
- the most operations can be done offline
- common operations are thus very fast because there is no network overhead
- new kind of very powerful features

You can read more with this illustrated introduction.

Today, the most used DVCS seems to be Git, Mercurial and Bazaar (like SVN which seems to be the mainly used Centralized one).
There are lots arguments for or against a particular system; for instance there, there, or there.

Each one will use the one matching the most his needs.
I've chosen Git because it matches my free and open-source philosophy, it is very incredibly fast and efficient, documentation and samples are complete and of great quality and Git's team is hightly reactive.

IMO, to 'begin' the necessary links are:
- User guide
- Git reference
- the Pro Git Book which is a very complete documentation to deepen the knowledge of the system

Activate logging for Mozilla applications (Firefox, Sunbird, Thunderbird ...).

2011-09-14T21:19:00.000+02:00

There is many applications using NSPR, including Mozilla ones (Firefox, Sunbird, Thunderbird ...).
NSPR provides a set of logging functions which can be activated with
two global variables: NSPR_LOG_MODULES and NSPR_LOG_FILE.
See NSPR Reference for further information.

For Mozilla Thunderbird specific information, see this KB.

It is not easy to find an exhaustive list of [NS]PR module name, this article is the more complete I've found.

For instance, if you want to log POP3 and SMTP warnings, you can define global variables:
export NSPR_LOG_MODULES="POP3:3,SMTP:3"
export NSPR_LOG_FILE="/tmp/nspr.log"

Be aware that some applications use level PR_LOG_ALWAYS for some logs, some whatever the level you define (but 0 for none), you will get those log.
For instance, we can see such instruction in Thunderbird source code.

You can specify any module you want to activate logging of any applications based on NSPR you use.

Change window behavior to prevent focus stealing (under KDE and Gnome)

2011-09-14T20:55:00.002+02:00

Bored of programs stealing focus ?
For instance, you launch an application which takes time to start, you switch to something else to work, and after some time the main window of your just launched program pops up and steals focus (including keyboad ...).

This article gives instructions to change this behavior both under
KDE and Gnome.

Under KDE:
- open 'System Settings' application
- open 'Window Behavior' menu
- select 'Window Behavior' left panel
- select 'Focus' right panel
- update 'Focus stealing prevention level' and policy to your needs

Under Gnome:
- check 'focus new windows' configuration (it is defined to 'smart' by default):
gconftool-2 --get /apps/metacity/general/focus_new_windows
- change 'focus new windows' from 'smart' to 'strict':
gconftool-2 --set /apps/metacity/general/focus_new_windows --type string strict

Under Gnome-Shell:
In addition to previous instructions for Gnome, which don't seem enough for Gnome-Shell (Gnome 3), you can:
- check 'no focus windows' configuration (empty by default):
gconftool-2 --get /apps/metacity/general/no_focus_windows
- change 'no focus windows' to fully disable focus for ANY windows:
gconftool-2 --set /apps/metacity/general/no_focus_windows --type string "(not true)"

You won't be annoyed anymore by windows stealing focus.

Create partition larger than 3TB under GNU/Linux

2011-07-17T16:20:00.000+02:00

You may need to create partition larger than about 3TB under GNU/Linux, with recent Hard Disk Drives having great capacity, with RAID configuration ...
Usually to create/manage partition with command line under GNU/Linux, fdisk is very adapted and powerful, but it is unable to create partition greater than 2/3TB (according to version, and architecture ?), because it needs to manage GPT partition (not supported by fdisk).
For that, you can use GNU Parted which is another powerful tool.

Instead of long explanation, this is a concrete example creating a 4TB partition on a RAID 10 (let's call it /dev/md16):
To begin, activate the GPT:
parted /dev/md126 mklabel gpt

Then, create a primary partition with all spaces (4TB):
parted /dev/md126 mkpart primary ext4 0 100%

Finally, formats it (for instance with ext4 and default block size):
mkfs.ext4 -v /dev/md126p1

Manage Intel RAID under GNU/Linux using mdadm tool

2011-07-17T16:00:00.001+02:00

You may be interesting to read some history about RAID first.
This article is about managing Intel RAID, with mdadm tool, but some information may apply to others kinds.
To begin, you can find lots of information about RAID under GNU/Linux.

IMPORTANT: for Intel's IMSM support you need at least
version 3.2 of mdadm tool (thanks to Iwan for this information).

Usually, there is a container device (let's call it /dev/md127 in this article), and the RAID device (let's call it /dev/md126 in this article).
Instead of long blabla, this article will give some common commands to help manage Intel RAID with mdadm tool.

To report the RAID information from the Option ROM (for instance "Intel(R) Matrix Storage Manager"):
mdadm --detail-platform

To get general details:
mdadm -D /dev/md127

To examine information (including members):
mdadm -E /dev/md127

To assemble the RAID array inside the container, and/or to start it (the -e option declares the style of RAID metadata (superblock) to be used):
mdadm -I -e imsm /dev/md127

To check the current state:
cat /proc/mdstat

Output sample (with a RAID 10 having 4 members):
Personalities : [raid10]
md126 : active raid10 sda[3] sdb[2] sdc[1] sdd[0]
3907023872 blocks super external:/md127/0 64K chunks 2 near-copies [4/4] [UUUU]
[=>...................] resync = 6.6% (258923904/3907024128) finish=305.7min speed=198882K/sec

md127 : inactive sdb[3](S) sda[2](S) sdc[1](S) sdd[0](S)
9028 blocks super external:imsm

To add it to the configuration file (to be activated automagically at boot):
mdadm --examine --scan >> /etc/mdadm.conf

For advanced commands (including repairing), you should read this.

RAID 10 on Intel ICH10R under GNU/Linux (including HDD devices on Marvell 88SE9128)

2011-07-17T15:42:00.000+02:00

After my experience to setup a RAID 0 on Intel ICH10R Vs Marvell 88SE9128; I've recently experienced the setup of
a RAID 10 on Intel ICH10R, using the mdadm tool.

To begin, I've never succeeded to install a bootable Operating System directly on the RAID (tested with Fedora 12, Fedora 13, Fedora 14, Ubuntu 10.04 LTS and Mandriva 2010.1).
So I added an additional HDD device dedicated to the OS, that I plugged on the Marvell 88SE9128 (at the moment, it was more easy with cables).
This time, the OS was bootable, but asap there was more or less I/O stress on drives (OS or RAID ones), the system failed with lots of severe I/O errors, and eventually one of the RAID HDD was declared offline (although it was healthy).

As soon as I plugged all the HDD devices on the Intel ICH10R controller (and after a last OS installation ...), everything became perfect : very efficient and very stable with absolutely no I/O error whatever the stress.
For information, there is no problem with optical devices on the Marvell 88SE9128 controller.

Like in my last article on this subject, I can only conclude that if ever there is a RAID configuration on a system, no HDD devices should be plugged to the Marvell 88SE9128 controller; otherwise the system will be strongly instable and will eventually fail.

Some history about Intel RAID under GNU/Linux (dmraid/mdadm)

2011-07-17T15:22:00.000+02:00

Although RAID is supported under GNU/Linux since many years, it is not trivial either.
This article is about Intel RAID, but some information may apply to others kinds.

There is at least two tools allowing to manage RAID devices:

- dmraid (Device-mapper RAID tool and library)
- mdadm (controls Linux md devices - software RAID arrays)

Intel RAID can be regarded as pseudo-Hardware-based RAID or Software-based RAID, so the 2 tools can be used.
Anyway, mdadm is more commonly used with modern distribution (for instance Fedora switched from dmraid to mdraid since version 12).

Besides, if you ever have a problem after an OS upgrade, with an existing RAID that you manage with dmraid, you should use the kernel option "noiswmd" which will make the system use dmraid for Intel BIOS RAID instead of mdraid.

Get KDE4.6 applications smooth again !

2011-06-09T14:28:00.001+02:00

Since KDE4.6, some of KDE applications can be very ...
very slow sometimes (kate, konqueror, dolphin).

One of the workaround is too specify another Qt graphical system, for instance using raster:
kate --graphicssystem raster
dolphin --graphicssystem raster

Configure easily themes of GTK+ applications, QT4 applications ...

2011-06-08T14:25:00.000+02:00

It's not always immediate to configure Themes of Desktop Environment, GTK+ applications and Qt4 applications separately.

For Desktop environment, there is usually embedded tools.

For GTK+ applications, we can use the gtk-chtheme tool.
Under RedHat-like GNU/Linux, it can be installed with the package of the same name:
yum install gtk-chtheme
Each user wanted to configure his GTK+ applications theme can then use this tool.

For Qt4 applications, we can use the qt-config tool.
Under RedHat-like GNU/Linux, it can be installed like that:

yum install qtconfig-qt4

Each user wanted to configure his Qt4 applications theme can then use this tool.

Lots of themes can be found on OpenDesktop.

Control your GNU/Bash Prompt

2011-06-03T16:48:00.002+02:00

With GNU/Bash, it is easy to control your prompt, defining PS1 (and optionally PS2 to PS4) environment variable.

For better legibility or whatever reason, you may want to
add further information or colors to prompt.
For instance you can see this page, and this one.

But it is very important to escape special characters dedicated to colors.
For instance, this prompt will NOT work:
~~export PS1="\e[0;32m\t\e[0m \u@\h:\e[0;36m\W\e[0m$ "~~

With this prompt, GNU/Bash will NOT work properly while breaking lines (for instance if its length if greater than limit) or while moving to begin/end of them.
To work, it is needed to add escaped bracket between each color escaped characters; this is a working prompt with colors (starting with date and time, and then usual information: user name, host and relative path):
export PS1="\[\e[0;32m\]\t\[\e[0m\] \u@\h:\[\e[0;36m\]\W\[\e[0m\]$ "

To persist your GNU/Bash prompt, you can add this in a GNU/Bash startup file (e.g. $HOME/.bashrc).

RAID 0 on Intel ICH10R Vs Marvell 88SE9128

2011-04-20T21:05:00.000+02:00

I've recently setup a RAID0 of two SATA 3 (6Gb/s) HDD, on a Asus P6X58D-E.
HDD can not reach theorical efficiencies of SATA 3 because of mechanical limits, but in RAID0 configuration,
such limit can be overriden.
So there was 2 choices:
- 2 HDD in SATA 3 mode on Marvell 88SE9128 controller
- 2 HDD in SATA 2 mode on Intel ICH10R controller (2011 specifications update)

For the greatest efficiencies, the reflex is to use the Marvell 88SE9128 controller to benefit from SATA 3 mode.
Despite 2 installation attempts, the system was instable, and performances were drastically variable (easily seen with various Benchmarks).
In addition, it was impossible to get SMART information on any of the HDD.
Because all I've then read on Web was against Marvell, BIOS (1.0.0.1019), and drivers (1.0.0.1036) were up to date, I try Intel ICH10R.

It's as quick and easy as to setup RAID0 on Intel ICH10R controller than on Marvell 88SE9128 controller; but that's where the comparison ends.

The read/write efficiencies are greater with Intel ICH10R controller (despite in SATA 2 mode, instead of SATA 3) and very constant. System is as stable as it can be.
The Intel Rapid Storage is interesting to get quick Health information, and there is no problem to get SMART information for each HDD of the RAID.

This is an interesting article comparing these 2 controllers.

For HDD, I think there is no hesitation to have, Intel ICH10R controller is far more better.
Maybe for SSD, for which it is possible to reach SATA 3 theorical limit, Marvell 88SE9128 controller can be better ?
It is another story ...

Fix kind of "GLib-GIO:ERROR:gdbusconnection.c:2270:initable_init: assertion failed" issue

2011-04-01T11:07:00.000+02:00

When launching some applications (e.g.: emacs, gedit ...) as root user, you may have error message like:
GLib-GIO:ERROR:gdbusconnection.c:2270:initable_init: assertion failed
It means root user doesn't have access to the session bus.
To solve that, simply ensure having a login shell for your root user:
su -

Define google-chrome as default Web browser in Thunderbird

2011-03-31T15:36:00.001+02:00

To define google-chrome as default Web browser in Thunderbird, there is two steps to follow.
Step 1 (may be enough):
- open advanced configuration editor (Edit->Preferences->Advanced->Configuration Editor)
- define following properties with /usr/bin/google-chrome
network.protocol-handler.app.http
network.protocol-handler.app.https

Step 2:
Usually there is embedded configuration in mimeTypes.rdf file of your Thunderbird profile. In this case, step 1 is not enough.

To define default Web browser:

- open advanced configuration editor (Edit->Preferences->Advanced->Configuration Editor)

- define following properties with true

network.protocol-handler.warn-external.http

network.protocol-handler.warn-external.https

- open a http link -> the first time there will be a popup, define /usr/bin/google-chrome as your default Web browser, and activate the check box for memorizing this configuration

- open a https link -> and do the same

You will see that mimeTypes.rdf file will have been edited.

This time, google-chrome will really be the default Web browser in Thunderbird.

Fix google-chrome "error while loading shared libraries: libnss3.so.1d" issue

2011-03-31T15:09:00.000+02:00

After GNU/Linux OS upgrade (at least with Fedora), google-chrome may be broken, complaining about missing libnss3.so.1d library.
To fix this issue:
- ensure you have installed nss v3.xx (corresponding to your architecture)
- create a symbolic link to corresponding library in the google-chrome main directory, usually:
ln -s /usr/lib64/libssl3.so /opt/google/chrome/libssl3.so.1d

There is some other needed symbolic links; if you have same kind of issue for others libraries, you can fix them the same way.

Anyway, at worse, if you don't want to fix it manually, you can reinstall google-chrome and reinstall it:
yum remove google-chrome-stable
yum install google-chrome-stable

Upgrade GNU/Linux Fedora from 10 to 11, 12, 13 and finally 14

2011-03-31T14:50:00.000+02:00

To upgrade GNU/Linux Fedora from 10 to 11, 12, 13 and finally 14, there is no problem.
Although the way to upgrade is
the same for legacy End Of Life (EOF) Fedora versions (until 10 included), and for Fedora 11; it is easier from Fedora 12 with new yum options --releasever and distro-sync.

The principle is globally the same when upgrading:
- from Fedora core N to Fedora core N+1 (see this post)
- from Fedora 8 to 9, then 10 (see this post)
- from Fedora core 6 to Fedora 7 then 8 (see this post).

In addition, the recommendations of Fedora project have greatly evolved and are complete.

Ensure there is no dependencies problem like explained into this post.

Then, you should perform a great configuration files merging campaign to ensure having the up to date functionalities while keeping your own specific configuration (globally the XXX.conf and XXX.conf.rpmnew files).

Google Chrome - some more interesting extensions

2011-03-14T12:29:00.000+01:00

Looking for Google Chrome extensions, I've found some additional ones very interesting (equivalent may exist in other Web browsers but it is not the subject of this post).

To begin, the fact that extensions can be installed very quickly and without restarting the Web browser is something very appreciate.
In addition, each extension usually adds a single little button in toolbar, which is very interesting to have quick access information/functionality, without reducing legibility of the whole Web browser, adding lots of tool bars every where ...

To have some counters in toolbar:
- Mail Checker for Google Mail™
- Google Reader Notifier

To manage (too) many tabs:
- TooManyTabs

Some others:
- Google Dictionary -> quick terms' definition as tooltip
- Status-bar Calculator -> small robust calculator
- The best price -> automatic price comparator

Comparing some default and additional functionalities of Firefox, Chrome and Opera

2011-03-14T12:14:00.000+01:00

Some months ago, I've posted "Firefox add-ons providing more or less same Opera functionalities"; this time, I'll compare Firefox, Google Chrome and Opera functionalities.

The comparison is based on functionalities I find the most useful:
- reduce the GUI (hidden menu, not used buttons ...)

Firefox: Add-on: Personal menu add-on

Google Chrome: default

Opera: Configuration: Show (only) "icons", and "Use small icons"

- create nickname to bookmark and allow quick load
Firefox: Add-on: url_alias add-on
Google Chrome: Alias Links extension
Opera: embedded

- sessions system (set of opened URL and/or Web Browser windows) - save/load; automatic session load when starting Web Browser
Firefox: Add-on: Session Manager add-on
Google Chrome: Session Manager extension
Opera: embedded

- dictionary system while editing
Firefox: embedded
Google Chrome: embedded
Opera: embedded

- speed dial
Firefox: SpeedDial add-on
Google Chrome: SpeedDial extension
Opera: embedded

- "unwanted blocking system" (pop-ups, pubs, ...)
Firefox: noscript add-on (very strict), Adblock Plus add-on
Google Chrome: Adblock & Adblock Plus extensions
Opera: embedded

- bookmarks synchronization
Firefox: Xmarks Bookmark and Password Sync add-on
Google Chrome: embedded
Opera: embedded

- download to various folder according to file extension
Firefox: Automatic Save Folder add-on
No equivalent found.

- left multi-function panel (with the tiniest bar possible)
Firefox: all-in-one_sidebar add-on
Google Chrome: no equivalent found yet
Opera: embedded

- attach Javascript code to a button or a menu

Firefox: all-in-one_sidebar add-on

Google Chrome: no equivalent found yet (may be Stylish extension)

Opera: embedded

Migrate from one MySQL server to another

2011-02-06T10:25:00.001+01:00

Migration from one MySQL server to another can be done in 3 steps; assuming your new server is ready.

Step 1 - your data:
Export all your databases (but mysql and test) and import them to your new server

Step 2 - access definition:
Export user and db tables of mysql database, and import them to your new server

Step 3 - flushing privileges:
Execute the following SQL command :

flush privileges;

Now, you can use your new server (data + access) exactly like the previous one.
Tips: do not forget to update the database host/port information of your third party applications using MySQL server.

Get/probe audio/video codec/container information of a multimedia file

2010-11-26T11:53:00.000+01:00

FFmpeg is a very powerful product allowing to play, create or get information from supported multimedia files.
Nevertheless, a light tool allowing to probe audio/video codec information can be very useful.
FFprobe is the perfect candidate, based on FFmpeg functionalities, it gives quick information about video and audio streams.

If your GNU/Linux version does not provide it, you can compile it from source (see INSTALL file for requirements).
It may be needed to specify ffmpeg include directories for configure to work (it was my case under GNU/Linux Fedora 10):
./configure CFLAGS="-I/usr/include/ffmpeg/"

Migrate Virtual Machines from VMware to VirtualBox

2010-07-27T16:34:00.001+02:00

Even if you can add vmdk file as Hard Disks in the VirtualBox Media Manager, you lost your Virtual Machines.
The way giving the best compatibility to migrate your Virtual Machines from VMware to VirtualBox is to convert them in Open Virtualization Format (.ovf) first.

To do so, VMWare provides an OVF Tool (see User Guide) which is embedded in some recent versions of Workstation and Player.
If you don't have such version, download the OVF Tool version corresponding to your host.

For better compatibility, according to the guest OS there is some instructions to perform first (you may get more information reading this forum or this article) like:
- uninstall vmware-tools
- remove hardware like sound card(s), floppy drive(s), CD/DVD drive(s) (at least unmount iso file), additional IDE controller(s)

Just before converting, you should (otherwise you may get a generic error "Failed to open disk"):
- prepare a destination folder with enough free space
- ensure having rights to read source Virtual Machines, and to write in the destination folder
- switch off Virtual Machine to convert
- ensure there is no lock file/directory in the Virtual Machine directory (with .lck extension); otherwise restart it with VMware and stop it properly (at worst, delete the .lck file/directory)

Then, you can convert your Virtual Machine to Open Virtualization Format:

ovftool /path/to/VMware/VM.vmx /path/to/VM.ovf

Finally, you can import this "new" Virtual Machine in VirtualBox:
- check first you have configured the storage of Hard Disk Drives and Virtual Machines (see preferences) to ensure having enough free space (by default, they are stored under your home directory).
- configure the Virtual Machine: name, operating system, architecture, network mode; and finally add every hardware you need.
- start the Virtual Machine under VirtualBox to ensure it works perfectly.

If it is the case, you can delete the VMWare version and Open Virtualization Format version which are no more needed.

How to install VirtualBox under GNU/Linux

2010-07-27T15:14:00.000+02:00

VirtualBox is now embedded in lots of package managers.
For instance, from Fedora 11, it is available in RPM Fusion repository.
Such a way, to install the open-source edition:

you install VirtualBox-OSE
For older version, it may be available in some no-official yum repositories.

Anyway, whatever your GNU/Linux distribution version, you should find a binary version from the official site (they cover Fedora, Red Hat Entreprise Linux, Mandrake, Debian, Ubuntu, OpenSuse ...).

After the installation, launch VirtualBox with your user (avoid launching it with root), and configure the storage of Hard Disk Drives and Virtual Machines (see preferences) to ensure having enough free space (by default, they are stored under your home directory).

Why choose VirtualBox as Virtualization product ?

2010-07-27T13:25:00.000+02:00

Today, there is various Virtualization Products like VirtualBox, VMware Server/Player/Workstation, Parallels Workstation/Desktop and Windows Virtual PC.
Choice is not always easy, taking consideration of reputation, age, price, press papers, enterprise policies ...

We can find some comparisons, this fresh one is interesting and complete.

VirtualBox has impressively evolved in recent years. In particular with version 2 (released in 2008), and more recently the version 3 (released in june of 2009), it has become an undisputed alternative to other solutions.
Nowadays, we can read things like "With support for up to 32 virtual CPUs per VM, VirtualBox is now the class leader in terms of raw virtualization muscle", and "VirtualBox is now poised to challenge VMware and Microsoft in the datacenter" (source).
VirtualBox Wikipedia article is interesting too.

The list of features is impressive, and the hyper-reactivity (see the changeLog) makes it a very powerful product.
In addition, the integration to various GNU/Linux distribution like Fedora is a very important point making it very easy to install and upgrade the product.
For instance, for these two last points, that is not the case of the free version of VMWare server; its modules' source code has not evolved since months, so it must be patched to be used with newer kernel versions (see this article).

Today, with the maturity of the version 3 (features, powerfulness, hosts/guests support, integration), and in addition because it is an open-source product, and free of charge, VirtualBox has become some kind of obvious choice.

Fix Java Web Start UnsatisfiedLinkError

2010-07-19T17:28:00.000+02:00

In some cases, mixing i386 and x86_64 architectures, Java Web Start can fall into such error:
Exception in thread "main" java.lang.UnsatisfiedLinkError: Can't load library: /usr/local/java/jdk/jdk1.5.0_14/jre/lib/amd64/libdeploy.so

at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1702)
at java.lang.Runtime.load0(Runtime.java:770)
at java.lang.System.load(System.java:1003)
at com.sun.deploy.config.UnixConfig.loadLibDeploy(UnixConfig.java:41)
at com.sun.deploy.config.UnixConfig.(UnixConfig.java:26)
at com.sun.deploy.config.ConfigFactory.newInstance(ConfigFactory.java:11)
at com.sun.deploy.config.Config.getInstance(Config.java:569)
at com.sun.deploy.config.Config.(Config.java:585)
at com.sun.javaws.Main.main(Main.java:82)

To resolve it, you can simply erase your deployment configuration file:
~/.java/deployment/deployment.properties

Improving efficiencies of OCZ SSD Vertex under Windows Seven

2010-07-16T11:33:00.001+02:00

Following my post about SSD, this one is dedicated to OCZ SSD Vertex efficiencies optimization under Windows Seven.
One of the latest improvements of SSD management was the addition of TRIM support. Interesting, but not enough

Comparing to various concurrents, OCZ is very active to improve their products, for instance offering Firmware upgrade regularly (about 3 in less than 1 year).
From the Firmware 1.4, they added support to TRIM.
But they have been much further from the Firmware 1.5, including their Garbage Collection technology (in fact it was included to Firmware 1.4.1 too, but it was exclusive with TRIM support); improved with Firmware 1.6.

In my case, the Firmware 1.5 was an incredible solution, finally allowing my SSD to reach almost the theoretical efficiencies (230MB/s in read, and 130MB/s in write).

However, there was still something to do.
Because of all my tests for improving efficiencies, including those tuning instructions, older versions of my SSD Firmware created meta-data which were now useless.
I don't know every available solution for this, but there is an absolute one which gives 100% success.
It consists to fully remove all meta-data of the SSD (obviously all data are then lost), like it was just out of the box; this operation is called Sanitary Erase.

If there is already something on the SSD (which will usually be the case), you should use a disk/partition cloner to backup/restore; like PING (free), or Acronis (proprietary).

Solid State Drive (SSD), advantages and disadvantages

2010-07-16T10:44:00.001+02:00

Although the notion of Solid State Drive (SSD) exists for a long time, industries have made significant progress last few years.
It has become more accessible, less expensive and offering very interesting efficiencies.

Comparing to conventional Hard Disk Drive (HDD), read/write performances can be increased by ten and sometimes even more.
Because it is a very different technology, there is advantages and disadvantages.
From my point of view, the greatest advantages are the performances, the silent operations, and lower power consumption, and the worst disadvantages are performance loss in function to available free space, lower lifetimes, and finally the prices of GB (SSD with lots of capacity are very very expensives).

The performance degradation is a very delicate point.
There is some articles (like this one for OCZ, and this one of Corsair) giving lots of tuning instructions but it is still a too young "science" and there is still lots of misunderstanding.
It seems, some instructions are not really recommended for operating system (like Windows Seven), although it may be needed to improve SSD lifetimes. It is why it is a delicate subject.

Get Java Web Start Application error when console vanished

2010-07-16T10:00:00.000+02:00

Java Web Start technology allows activating either "trace" or "log" system (or the two of them).

To activate them:

- launch the (same version used when launching JNLP file of) Java Web Start viewer:

/path/to/javaws -viewer

- close the viewer main window

- go to the advanced panel and switch on either "trace" or "log", or the two of them

The next time you will use the same version of Java Web Start, it will produce log files.

It is particularly needed when the console vanishes after showing error stack(s) (usually when the launched application fails startup).

You can get and analyse log/trace, looking under your home directory:

~/.java/deployment/log/

Update Master Boot Record (MBR) replacing the System disk

2010-07-13T14:55:00.001+02:00

Some times ago, I've written an article explaining how to restore MBR with grub. It works only if there is no disk change.

Almost the same procedure can be used if the System disk change.

For instance you have your GNU/Linux on your first disk, and you add another disk (with better efficiencies ?) as System disk (getting the first index), but you want to restore this MBR instead of other (for instance if you install a Windows operating system on your new disk).

The variables of this example are:
- /dev/sda the new disk,
- /dev/sdb the old system disk
- /dev/sdbM the ext2/ext3/ext4 boot partition (corresponds to hd1,M-1 for bootloader)
- grub as bootloader

Instructions:
- boot with a GNU/Linux Rescue or Live CD of your choice
- remove/rename the /boot folder of the launched GNU/Linux Rescue or Live CD instance
- mount your boot partition
mkdir /boot
mount -t ext3 /dev/sdaM /boot
- update the grub configuration file (/boot/grub/grub.conf), replacing all hdN occurrence corresponding to partition of the old disk, by hdP (with P=N+1)
- prefer using LABEL=XXX or UUID=YYY to indicate the root partition on the "kernel" line instead of hdX information
- update the grub device map (very important for old and new disk to be seen)
grub-install --recheck /dev/sdb
- update grub MBR information (the disk/partition information correspond to your /boot partition)
grub
root (hd1,M-1)
setup (hd1,M-1)
quit
- update the MBR (the disk information corresponds to where the MBR must be installed - usually the first disk (your new one))
grub-install /dev/sda

Restore lost partition with fdisk

2010-07-12T12:17:00.005+02:00

Unfortunately, there is various situations where partition can "vanish" (bad manipulation, unwanted application behaviour, issue while using a partition manager ...).
Some years ago, I've started using a very simple rule:

Save the result of fdisk -l into a text file after each disk/partition management

It is very simple, it takes very few time and it can save many hours avoiding awful situations and some times data loss.

If ever you lost a partition:
- start with a GNU/Linux LiveCD (e.g. Fedora)
- retrieve your fdisk output (in particular first and last cylinders, indicated under "start" and "end" columns)
- launch fdisk in interactive mode on the regarded device (e.g. /dev/sda)

fdisk /dev/sda

- restore missing partition using first and last cylinder (typing 'n' for new partition)
- if needed, update the type of partition (typing 't' for changing partition' system ID; then 'L' to get the list of available codes)
- save changes (typing 'w')

If nothing has been written on the corresponding disk space, you will get back lost partition without data loss.

Install VMware-server 2.0.2 under kernel 2.6.31, 2.6.32 or 2.6.33

2010-07-07T17:36:00.002+02:00

The free VMware-server has not evolved since end October of 2009 (v2.0.2), although fortunately Linux kernel keeps evolving.
Such a way, there is a gap between VMware-server modules source code and awaited kernel source code.

It is absolutely needed to patch VMware-server modules source code to install it with recent kernel versions; this post gives instruction to do so with kernel 2.6.31, 2.6.32 or 2.6.33.

To begin, some header files have moved (you can read some information there), which can be fixed so:
cd /usr/src/kernels/$( uname -r )/include/linux
ln -s ../generated/utsrelease.h
ln -s ../generated/autoconf.h
Then to patch the VMware-server modules source code, it is needed to use this patch.
If you have downloaded the gunzip version, you can follow the instructions given with the patch, it is ok.

Otherwise, if you have downloaded the rpm version, those are some instructions to use this patch:
- install the rpm (do not launch the config script)
- backup source tar files (located under /usr/lib/vmware/modules/source), and then uncompress them
- create symbolic link (like if it was the tarball version)
mkdir -p /usr/lib/vmware/modules/source/a/lib/modules && ln -s /usr/lib/vmware/modules/source /usr/lib/vmware/modules/source/a/lib/modules/source
- use the patch
patch -N -p1 --directory="/usr/lib/vmware/modules/source/a" -s < /path/to/source-update.patch
- [re]create the source tar files, with patched source code
cd /usr/lib/vmware/modules/source
for tarFile in vmci vmmon vmnet vsock; do tar cf $tarFile".tar" "$tarFile"-only; done
- patch the config file
patch /usr/bin/vmware-config.pl < /path/to/config-file.patch
- launch the config file
vmware-config.pl

Under Fedora (and perhaps others distribution ?), it is needed to perform additional instructions:
- edit /etc/services and add an entry linking TCP/902 port and vmware-authd
- disable SELinux, editing the /etc/selinux/config file (setenforce 0 can be used at runtime)
- move the VMware-server service file for haldaemon to start before it

mv /etc/rc3.d/S24vmware /etc/rc3.d/S27vmware

Check a File System

2010-05-30T12:13:00.002+02:00

To check a file system you can use the fsck command.
For detailed information checking GNU/Linux ext2/ext3/ext4 file system, you should use e2fsck.

To perform a simple check (if needed):

e2fsck /dev/sdaN

To force a check, with progress information (even if the file system is regarded as clean):

e2fsck -f -C 0 /dev/sdaN

To force a check, with progress information (even if the file system is regarded as clean), and to have the most debug and statistics information:
e2fsck -f -C 0 -t -t -v /dev/sdaN

To perform a check with automatic repairing and exit in case the command discovers a problem which may require system administrator to take additional corrective action:

e2fsck -p /dev/sdaN

To perform a check with automatic repairing, assuming 'yes' to ALL questions (must be used carefully):

e2fsck -y /dev/sdaN

N.B.:
- replace /dev/sdaN by the device path you want to check
- corresponding device must NOT be mounted
- you can combine the debug/statistics and auto-repairing options

Check badblocks on a device

2010-05-30T11:51:00.003+02:00

If error occurs, strange behaviour or if SMART error are reported (Cf. instructions to get report), you should check file System (see next post), including bad blocks.

To have a report as soon as possible, you could focus on checking bad blocks at first.

To perform a quick read-only check:

badblocks -s /dev/sdaN

To perform a non-destructive read-write check:

badblocks -ns /dev/sdaN

To perform a write check (destructive):

badblocks -ws /dev/sdaN

N.B.:
- the -s option is interesting to be notified about check progress
- replace /dev/sdaN by the device path you want to check
- corresponding device must NOT be mounted (do NOT use the -f option !)

Migrate you URL nicknames from Opera to Firefox

2010-04-23T18:43:00.002+02:00

Follow those instructions:
- under Opera, export your bookmarks (e.g. to /tmp/bookmarks.adr file)
- into a console, parse the file like that :
cat /tmp/bookmarks.adr |tr -d '\n' |sed -e 's/#/\n/g;' |grep "SHORT NAME" |sed -e 's/.*URL=$[^ \t]*$[ \t].*SHORT[ ]NAME=$[^ \t]*$[ \t].*$/\2\t\1/g;' > /tmp/urlAlias.txt
- under Firefox, go to Preferences of Url Alias add-on
- copy/paste contents of /tmp/urlAlias.txt in "Configuration" panel

Firefox add-ons providing more or less same Opera functionalities

2010-04-23T15:50:00.001+02:00

Those are Firefox add-ons which fit more or less some Opera functionalities I found the most useful:

- reduce the GUI (hidden menu, not used buttons ...)
Configuration: Show (only) "icons", and "Use small icons"
Add-on: Personal menu

- create nickname to bookmark and allow quick load
Add-on: url_alias

- sessions system (set of opened URL and/or Web Browser windows) - save/load; automatic session load when starting Web Browser
Add-on: Session Manager

- dictionary system while editing
Embedded

- speed dial
Add-on: SpeedDial

- "unwanted blocking system" (pop-ups, pubs, ...)
Add-ons: noscript (very strict), Adblock Plus

- bookmarks synchronization
Add-on: Xmarks Bookmark and Password Sync

- download to various folder according to file extension
Add-on: Automatic Save Folder

- left multi-function panel (with the tiniest bar possible)
Add-on: all-in-one_sidebar

There is some functionalities for which I've not found equivalent yet
- synchronization of speed dial, url_alias, blocked elements ... (and configuration ?) to share with other computers (anyway, it'd never fully worked with Opera under GNU/Linux -> usually duplicate entries creation by the synchronization process itself)
- per-site configuration (charset encoding, cookies, language ...)
- attach Javascript code to a button or a menu

To conclude:
- something very interesting as compromise to automatic synchronization of almost everything : Full profile backup/restore system
Add-on: FEBE
- improved download notification system
Add-on: Download status bar

Goodbye Opera, Welcome back my Beloved Firefox Web Browser

2010-04-20T17:33:00.002+02:00

I'd used Opera from May of 2008 to March of 2010, against my strong philosophy of open-source software use.

The 9.x versions come with lots of improvements of GNU/Linux version, and then with a greater plug-ins support.
At the time, Opera had got the greater efficiencies, greater standard respect and better result in some tests like ACID.
After months of use, I'd faced some issue (usually little ones but boring), for each one I'd reported a richly detailed bug.

After (almost) 2 years of use, I can observe that:
- the GNU/Linux version of Opera still needs some improvements to better fit the OS components,
- the greatest and the more boring is - sometimes - an important resources use for no obvious reason, dropping down the performance of the whole computer,
- there is absolutely no visibility of reported bugs,
- I've never received any return of my - at least 10 - reported bugs (never corrected, never referenced into Release Notes, never contacted for further information request ...)
- they are late of 2 versions for GNU/Linux (although the "10.50 available shortly" mention is specified for month(s))
- since creation of Unite, obviously - according to resources - there is less effort done on the "browser part" evolution
- there is still some incompatibilities with "important"/useful WebSites like various Google services (gmail, blogger, documents ...), Bloglines ...

Because of the impressive richness of the Web Browser, it can't be replaced by everything.

During last 2 years, Firefox too has greatly evolved and lots of plug-ins too.
I'll post various information to switch from Opera to Firefox with some compromises (instead of loss).

Flush a queue of e-mails

2010-04-20T16:11:00.000+02:00

For various reasons it's possible that some e-mails are queued (for instance because of issue into an element of a full and safe emails server - e.g. too old packages, broken anti-virus, broken anti-spam ...).

After having fixed the issue, it's interesting to flush all queued e-mails, which can easily be done with the following command:
postqueue -f

Stop updating packages as often

2009-10-25T19:12:00.003+01:00

When managing computer administration, there is a big trap to avoid: update the packages as soon as they are available.

Yes, there is often bug fix or security improvement and it's very important; but it can also be some new functionalities and potentially some unfortunate regression.
When such regression strikes one software/tool, it can generally be fixed (at worst downgrading to the previous version); but when it strikes the operating system or the graphical system, it becomes very boring, frustrating and sometimes very difficult to fix.

So instead of constantly updating packages, it seems better to be informed, see if we are concerned by fixed bug, security issues or if we need the new functionalities.
This way, there is more chance the regarded computer will work better and longer with less service interrupt.

Stop polling for incoming e-mails

2009-10-25T18:50:00.003+01:00

Unfortunately, it seems there are lots of us (computer users) polling for incoming e-mails very, very often; like if a very important thing might be received and to be answered almost immediately.

It can be a very important efficiency loss, because the user cannot focus on his current task.
Although that seems obvious, I've recently realized that it was also my case only after having read an interesting article about that.

Since some weeks, I've seriously changed my way to manage incoming e-mails:
- no more polling
- I've disabled all kind of notification of my e-mailer (no more popup which can be disturbing)
- I've disabled the automatic "read mark" to remove the pressure to answer as soon as possible when I finally check my e-mails.

It's little things but at the end it gives interesting results.
Yes, I answer e-mails later, but there is no dead, no fire, no tsunami, no storm ... and among all others, it gives me more opportunity to focus on my work, and so to be more efficient.

Help ! rpm or yum freezes

2009-09-18T10:07:00.004+02:00

Sometimes, rpm, yum or even other commands are freezing without message to understand why (e.g. for yum, just after the message "Running Transaction Test").

Almost each time I've seen that, it was caused by broken mount (usually network ones, when the server at the other side has failed, or has restarted).

To check it:
mount

If you see some network mount points you do not need anymore, or you're not sure about, umount them.
If they are broken or if the system wrongly thinks it is busy you can use the -l option (for lazy) to ensure the umount - in fact the system won't fully perform the umount but it will clean the list of mounted point, such a way the launched commands won't freeze anymore.

For instance, to umount (lazy) a broken mount point /mnt/myMountPoint:

mount -l /mnt/myMountPoint

Compile LZO, TUN and OpenVPN under Solaris 64bits (revised)

2009-07-28T09:41:00.009+02:00

(This post is a revision of this one for Solaris 64bits, and in accordance with new source code versions).

It is possible to get LZO, OpenVPN and TUN under Solaris, compiling the source code.
Because of some compatibility issues, it is not as easy as it can be thought.

You can follow those steps to get openvpn executable under Solaris:
- ensure GNU gcc and make are installed
pkg install SUNWgcc
- download the LZO source code and uncompress it under [LZO_DIRECTORY],
- download the TUN source code and uncompress it under [TUN_DIRECTORY],
- download the openVPN source code and uncompress it under [OPENVPN_DIRECTORY],
- update the path for all needed tools (gcc, gmake, install, ar ...) to be available
export PATH=/usr/sfw/bin:/usr/ucb/:/usr/ccs/bin:/usr/sbin:/usr/bin:$PATH

- compile LZO
cd [LZO_DIRECTORY]
./configure --libdir=/usr/local/lib
make
make install

- configure TUN
cd [TUN_DIRECTORY]
./configure
- edit solaris/Makefile to add options as followed
CFLAGS = -m64 -mcmodel=kernel -mno-red-zone -ffreestanding $(DEFS) -O2 -Wall -D_KERNEL -I.
- compile TUN (ignore the error "Warning: 64-bit version of driver found at /usr/kernel/drv/tun")
make
make install
- move the created tun file (to be regarded as 64bits)
mv /usr/kernel/drv/tun /usr/kernel/drv/amd64/
- load it
devfsadm -i tun

- compile OpenVPN
cd [OPENVPN_DIRECTORY]
./configure --with-lzo-lib=/usr/local/lib --with-lzo-headers=/usr/local/include --prefix=/usr
make
make install

It is now possible to use LZO, OpenVPN and TUN under Solaris.

Master your installed tools

2009-07-19T14:45:00.004+02:00

Sometimes, we use tools from so long time that we don't know/care/remember where it comes from.
It is particularly true for system tools (e.g. mount, fsck, tune2fs, ls...).

To get such information, installed RPM packages can be requested (like explained in this article).

For instance, to get information about the package providing the mount tool (under GNU/Bash):
rpm -qfi $( which mount)

Then, to see what other tools are provided by this package, check the list of provided files:
rpm -qfil $( which mount)

Usually, using those instructions help mastering installed tools, and to discover some you don't use yet.

Create a specific user for MySQL Database backup

2009-07-19T14:37:00.004+02:00

For security reasons, it is interesting to create a specific user for MySQL Database backup with the less privileges possible.
In fact, only LOCK TABLES and SELECT privileges are requested to use mysqldump.

So it is enough to use the following SQL query:
GRANT LOCK TABLES, SELECT ON XXX.* TO 'YYY'@'localhost' IDENTIFIED BY 'ZZZ';
GRANT LOCK TABLES, SELECT ON XXX.* TO 'YYY'@'localhost.localdomain' IDENTIFIED BY 'ZZZ';
FLUSH PRIVILEGES;

With variables:
XXX: the database name
YYY: the specific backup user login
ZZZ: the specific backup user password

The following instruction allows full backup with this limited privileges user:
mysqldump -u "YYY" --password="ZZZ" --all-databases > /path/to/my/dump/file

Manage SMART disks

2009-07-14T15:34:00.003+02:00

Modern hard disks (from ATA-3 and SCSI-3) provide built-in system called SMART (for Self-Monitoring, Analysis and Reporting Technology).

smartmontools provides two utility programs (smartctl and smartd) to control and monitor storage systems using SMART.

It can be installed in RedHat-like systems using yum (yum install smartmontools).

To get "general" Info like device model number, serial number, firmware version... (e.g. /dev/sda):
smartctl -i /dev/sda

To get Health Status of a device - for instance for predicting its own failure within the next 24 hours (e.g. /dev/sda):
smartctl -H /dev/sda

To get SMART Attributes of a device (e.g. /dev/sda):
smartctl -A /dev/sda

To get SMART Capabilities of a device (e.g. /dev/sda):
smartctl -c /dev/sda

To get error information of a device (e.g. /dev/sda):
smartctl -l error /dev/sda

To get all those information and even more about a device (e.g. /dev/sda):
smartctl -a /dev/sda

The smartctl command allows launching some tests.
For instance to launch a short test:
smartctl -t short /dev/sda

To get progress and result information:
smartctl -l selftest /dev/sda

See smartctl MAN page for further information.

Remount a read-only mounted partition

2009-07-14T15:25:00.002+02:00

In some cases (kind of maintenance mode for instance), partitions can be read-only mounted, and the situation stays the same after operations (like fsck) and reboot.

For instance, it can happen if the system is wrongly configured, forcing the fsch of unknown (or removed) devices.

To leave such end-less situation, it is possible to remount a read-only mounted partition, which can not be umounted (like /), to update configuration and then get a working boot:
mount -n -o remount,rw /mountPoint

The -n option allows to request the mount without attempting to write on the disk (which is obviously not possible on read-only mounted partition).
The -o option allows updating mount options (same syntax than when mounting the partition the "first" time)

Install 64 bits Java plug-in for Opera 64 bits under GNU/Linux

2009-04-06T11:09:00.005+02:00

Since SUN jdk 6 update 10, a new generation plug-in is provided.
This is an article with giving lots of information.
To install it under Opera 64 bit, add the [jdk_installation_dir]/jre/lib/amd64/ under the plug-ins path definition (Tools-> preferences-> advanced-> contents-> plug-ins options-> change path-> add). The corresponding library is libnpjp2.so.

You can check your installation with this site applet.

Install VMWare-server v1.x under kernel 2.6.27 or 2.6.28

2009-03-13T11:04:00.003+01:00

To install VMWare-server v1.x under kernel 2.6.27 or 2.6.28, use the following patch (the vmware-any-any-update115 does not work from this kernel version).

For x86_64 architecture, there may be a lock problem while launching VMware-server, ending with following final error message:
vmware: xcb_lock.c :77 : _XGetXCBBuffer: L'assertion « ((int) ((xcb_req) - (dpy->request)) >= 0) » a échoué.

To solve this issue, install the gtk-nodoka-engine.i386 package (and needed packages).

Install Windows after GNU/Linux on the same hard disk

2009-03-02T20:38:00.006+01:00

There can be various issues when attempting to install Windows on which is installed a GNU/Linux distribution.
One of the more subtle is hardware detection problem which can occur after the first second while Windows installation CD is booting (See this interesting post which gives some explanations in specific cases).

It can be the case if there is a dedicated boot partition for GNU/Linux.
The solution is to temporarily delete the partition, perform the Windows installation, restore the partition, and finally restore the MBR.

Those are some instructions to help you solving this issue.
The variables of this example are:
- /dev/sda the main hard disk,
- /dev/sdaN the boot partition (ext3) before deletion (corresponds to hd0,N-1 for bootloader)
- CF and CL the first and last cylinders of the /dev/sdaN partition (given by fdisk -l)
- /dev/sdaM the boot partition after restore (corresponds to hd0,M-1 for bootloader)
- grub as bootloader

Instructions:
- backup your /boot/grub/grub.conf
- backup your boot partition into a file
dd if=/dev/sdaN of=/tmp/boot.bin
- temporarily delete the partition
fdisk /dev/sda
d (to delete partition)
select the partition number (N in this instance)
- perform the Windows installation
- boot with a GNU/Linux Rescue or Live CD of your choice
- restore the partition
fdisk /dev/sda
n (to add new partition)
specify CF and CL as first and last cylinders

- use fdisk -l to identify the new partition number, in case it has changed (M in our case)
- format it
mkfs.ext3 -L boot /dev/sdaM
- restore your boot partition
dd if=/tmp/boot.bin of=/dev/sdaM
- remove/rename the /boot folder of the launched GNU/Linux Rescue or Live CD instance
- mount your boot partition
mkdir /boot
mount -t ext3 /dev/sdaM /boot
- edit the /boot/grub/grub.conf file to replace hd0,N-1 by hd0,M-1
- restore the MBR

Restore Master Boot Record (MBR) with grub

2009-03-02T19:02:00.009+01:00

In case your MBR is broken, those are some instructions to help you restore it.

The variables of this example are:
- /dev/sda the master hard disk,
- /dev/sdaM the ext3 boot partition (corresponds to hd0,M-1 for bootloader)
- grub as bootloader

Instructions:
- boot with a GNU/Linux Rescue or Live CD of your choice
- remove/rename the /boot folder of the launched GNU/Linux Rescue or Live CD instance
- mount your boot partition
mkdir /boot
mount -t ext3 /dev/sdaM /boot
- restore the MBR
grub
root (hd0,M-1)
setup (hd0,M-1)
quit
grub-install /dev/sda

Make some of your log files rotate and compress

2009-01-24T18:27:00.003+01:00

logrotate is a very useful and powerful tool.

To make some of your log files to be rotated and compressed, create a new configuration file into the logrotate "include" directory (usually /etc/logrotate.d), and add the rotate/compress rule:

/var/log/XXX.log {
rotate 3
weekly
compress
size 1M
create 0640 myUser myGroup
}

In this example, the log will go through three weekly rotations before being removed. There can be rotations if the size exceeds 1MB.
Rotated files will be compressed.
New empty file (after rotation) will be created with "0640" mode, and myUser:myGroup ownership.

Replace courier-imap by dovecot for emails server under GNU/Linux

2009-01-18T12:37:00.008+01:00

The courier-imap package, and linked packages, have been replaced by dovecot from Fedora 8 (or earlier).
Anyway, it's still possible to maintain mailboxes part of your full and secured GNU/Linux emails server, compiling the source code of courier-imap.

If you still want to benefit from up-to-date package, using yum or if you want a more secured system, you should migrate to dovecot.

To begin, there is a tool allowing to migrate "metadata" from courier-imap to dovecot.

This is how to configure dovecot editing the /etc/dovecot.conf file, according to the configuration of the full and secured GNU/Linux emails server:
- define clearly what protocols you want to avoid wasting resources and potential security hole (See this post for equivalent for courier)
protocols = pop3 pop3s

- define the certificate and the key to use (can be the same of the smtp server)
ssl_cert_file = /etc/postfix/smtpd.cert
ssl_key_file = /etc/postfix/smtpd.key

- define the mail location
mail_location = maildir:/home/vmail/%d/%n

- define the mail UID, GID and the privileged group:
mail_uid = 5000
mail_gid = 5000
mail_privileged_group = vmail

- comment the imap or pop3 begin/end line according to your needs

- specify pop3 UIDL format for it to be compatible with the existing courier metadat
pop3_uidl_format = %u-%v

- specify the mechanisms into the auth default part
mechanisms = plain login

- comment all lines corresponding to auth you do not want (like for pam for instance)

- specify path of ONE specific file (you will create) to request the SQL database to get password and user information in the same request (optimization) (respect the order which is important, "userdb prefetch" must be before "userdb sq" part)
passdb sql {
args = /etc/dovecot-mysql.conf
}
userdb prefetch {
}
userdb sql {
args = /etc/dovecot-mysql.conf
}

- create a file with SQL query and connect information (/etc/dovecot-mysql.conf), replacing mail, admin and XXX with your corresponding databse name, privileged user login and password (default_pass_scheme is very important and must be defined according to the function used when adding password to database; in this case it correspond to the ENCRYPT function):
## /etc/dovecot-mysql.conf contents ##
driver = mysql
default_pass_scheme = CRYPT
connect = host=localhost dbname=mail user=admin password=XXX

# Extended request allowing to get password and all user information at same time.
password_query = SELECT password, '/home/vmail/%d/%n' as userdb_home, 'maildir:/home/vmail/%d/%n' as userdb_mail, 5000 AS userdb_uid, 5000 AS userdb_gid, concat('dirsize:storage=',quota) AS userdb_quota FROM users WHERE email = '%u'

# Used only for deliver (see LDA).
user_query = SELECT '/home/vmail/%d/%n' as home, 'maildir:/home/vmail/%d/%n' as mail, 5000 AS uid, 5000 AS gid, concat('dirsize:storage=',quota) AS quota FROM users WHERE email ='%u'
## /etc/dovecot-mysql.conf contents ##

- (facultative) activate debug information in case something goes wrong
mail_debug = yes
auth_debug = yes
auth_debug_passwords = yes

- restart the dovecot service
service dovecot restart

- attempts to connect to your imap/pop server with your favorite email client, checking the log file (default is /var/log/maillog)

- (facultative) disabled debug information if no more needed

Upgrade GNU/Linux Fedora from 8 to 9, then 10

2009-01-18T12:18:00.002+01:00

To upgrade GNU/Linux Fedora from 8 to 9, then 10, there is no problem.
It is easier than previous upgrade, and there is less kind of issue.

The principle is globally the same when upgrading from Fedora core N to Fedora core N+1 (see this post), from Fedora core 6 to Fedora 7 then 8 (see this post), or from Fedora 7 to Fedora 8 (see this post).
In addition, the recommendations of Fedora project have greatly evolved and seem now complete.

Upgrade from Fedora 8 to Fedora 9:
- do not forget to clean all the yum metadata with yum clean all,
- upgrade the Fedora release:
rpm -Uvh ftp://download.fedora.redhat.com/pub/fedora/linux/updates/9/i386.newkey/fedora-release-*.noarch.rpm
- upgrade your repository if needed (N.B.: livna has now merged into RPM fusion)
- remove and reinstall thunderbird to avoid specific issue.

Upgrade from Fedora 9 to Fedora 10:
- do not forget to clean all the yum metadata with yum clean all,
- upgrade the Fedora release:
rpm -Uvh ftp://download.fedora.redhat.com/pub/fedora/linux/releases/10/Fedora/i386/os/Packages/fedora-release-*.noarch.rpm
- Be careful if you have RAID, there is an important issue which may prevent the OS from booting properly

Ensure there is no dependencies problem like explained into this post.

Then, you should perform a great configuration files merging campaign to ensure having the up to date functionalities while keeping your own specific configuration (globally the XXX.conf and XXX.conf.rpmnew files).

Have a quick look on installed packages providing services

2009-01-18T11:37:00.003+01:00

Sometimes, it is interesting to have a quick look on installed packages providing services to check if there is possibility to clean unused packages, or to disable (temporary or not) services which are not used and which are so wasting resources.

To get information about installed packages providing services (excluding inet and xinet) which are NEVER started (and so may be removed):
for service in $( chkconfig --list |grep "1:" |grep -v "on" |awk '{print $1}' ); do rpm -qi "$( rpm -qf /etc/init.d/$service )"; done |less

Important:
- some packages provide several services so it can appear several times with this simple command
- before removing such a package, it is important to ensure it is not providing something you need (libraries, tools, various files ...)
- you may use something else than "on" according to your OS language

To get information about installed packages providing services (excluding inet and xinet) which are started (and so using resources):
for service in $( chkconfig --list |grep "5:on" |awk '{print $1}' ); do rpm -qi "$( rpm -qf /etc/init.d/$service )"; done |less

Important:
- some packages provide several services so it can appear several times with this simple command
- you can perform this check with every runlevel replacing "5" by anything else
- you may use something else than "on" according to your OS language

Launch Jack Audio Connection Kit daemon with realtime scheduling

2008-05-31T13:37:00.003+02:00

Today, it is very simple to launch the Jack Audio Connection Kit daemon with realtime scheduling, with a no-root user.
When Jack Audio Connection Kit is installed, it adds specific "limit" to jackuser group, into Pluggable Authentication Modules (pam).
Cf. the /etc/security/limits.conf file.

So the only need is to add the user which must need to launch the Jack Audio Connection Kit daemon with realtime scheduling into the jackuser group:
/usr/sbin/usermod -G jackuser "username"

Then, use the -R option when launching Jack Audio Connection Kit daemon:
jackd -R ...

N.B.: those instructions have been performed (at least) under Fedora 8

Use MIDI software "soundfonts mapping"

2008-05-31T13:26:00.007+02:00

This is a quick "how to", to use softwares installed like explained on this post:
- launch Jack Audio Connection Kit daemon with options according to your configuration (let's said you are using ALSA)
jackd -R -d alsa
The -R option is used for "real-time", I explain it into my next post.
- launch a fluidsynth GUI frontend
qsynth
- setup the wished soundfonts (there we're using the path of the fluid R3 soundfonts), setup->Soundfonts->Open and selects the path /usr/share/soundfonts/FluidR3_GM.SF2 (under Fedora 8),
- finally launch Rosegarden and play MIDI files or Rosegarden Studio of your choice
rosegarden
- you may need to update configuration of the MIDI devices, Studio->Manage MIDI devices.

N.B.: those instructions have been performed (at least) under Fedora 8

Setup MIDI software "soundfonts mapping" (instead of hardware)

2008-05-31T13:02:00.010+02:00

In this post, I've (quickly) explained how to set up environment to play MIDI file, loading soundfonts into wavetable of the sound card.
Nevertheless, the soundfonts which can be loaded into the sound card is obviously limited to available memory and so compromise must be done between quality and size.

Instead of this, it is possible to use software "soundfonts mapping". Today, there is lots of very powerful and mature softwares although all must be set up with caution to avoid latency issue.
There is various interesting explanations on the subject, for instance Planet CCRMA at home, and Linux MAO.

This is the software which an be used (prefixed by corresponding package name under Fedora):
- jack (Jack Audio Connection Kit): a low-latency audio server,
- qjackctl: a simple application to control the JACK sound server,
- fluidsynth: a real-time software synthesizer based on the SoundFont 2 specifications,
- qsynth: a fluidsynth GUI front-end application,
- fluid-soundfont-R3: fluid R3 soundfonts of good quality,
- rosegarden4: a professional audio and MIDI sequencer, score editor, and general purpose music composition and editing environment.

N.B.: those instructions have been performed (at least) under Fedora 8

Prepare Fedora environment to become an audio plateform

2008-05-27T20:14:00.001+02:00

After some search of interesting tools to play midi musics (with external devices) under GNU/Linux, I finally find an unbelievable project: Planet CCRMA at home.
They provide all needed packages for updating an environment "to transform it into an audio workstation with a low-latency kernel".

There is rich installation information on the official site.

My first aim is to discover the astonishing features of Rosegarden.
To get it fully installed, I have had to install some additional packages XML::Twig and LilyPond:
yum install perl-XML-Twig lilypond

Currently the real-time kernel does not boot on my computer, I'll surely post information about this later ...

N.B.: those instructions have been performed (at least) under Fedora 8

Play MIDI under GNU/Linux with ALSA

2008-05-27T18:55:00.005+02:00

There is lots of articles on the subject but it is not so trivial to get needed information to use ALSA tools to play MIDI files.

This is some quick information:
- the first is obviously to have a working ALSA installation,
- if you have several sound cards, ensure the default is well the wanted one (for instance using system-config-soundcard under Fedora),
- then it is important to ensure not having mute channel(s), you can use alsamixer for this,
- then ensure to have a way to load soundfonts into the wavetable of the sound card; for instance of sound card managed AWE32/Emu10k1 sound driver, you can install the awesfx package
yum install awesfx
- get a soundfonts to load, for instance this one,
- at any moment, to check if there is a loaded soundfonts:
cat /proc/asound/card?/wavetableD?
- you can clear currently loaded soundfounts but you want to keep it:
/bin/asfxload -i
- then load it:
/bin/asfxload "soundfonts file path"
- find the "addresses" into the "/proc/asound/card?/wavetableD?" file (for instance: 17:0 17:1 17:2 17:3),
- finally test installation using aplaymidi to play a MIDI file
aplaymidi -p "addresses"* "MIDI file path"
* addresses must be coma separated, 17:0,17:1,17:2,17:3 for instance.

Define the different installed packages between GNU/Linux installations

2008-05-13T20:30:00.003+02:00

Let's call one installation local, and the other remote.
On the first one, use rpm to get the list of installed packages:
rpm -qa --qf "%{NAME}.%{ARCH}\n" |sort > /tmp/localInstallationInstalledPackages

Idem, on the second one:
rpm -qa --qf "%{NAME}.%{ARCH}\n" |sort > /tmp/remoteInstallationInstalledPackages

Then, get the differences between each list using diff.

To see the package installed on the local installation, but not on the remote one:
diff /tmp/localInstallationInstalledPackages /tmp/remoteInstallationInstalledPackages |grep "^<"

To see the package installed on the remote installation, but not on the local one:
diff /tmp/localInstallationInstalledPackages /tmp/remoteInstallationInstalledPackages |grep "^>"

In the two cases, to get a list of regarded packages on one line, use the following instructions.
For instance, for the local installation:
diff /tmp/localInstallationInstalledPackages /tmp/remoteInstallationInstalledPackages |grep "^<" |awk '{print $2}' |sed -e 's/$/ /' |tr -d '\n'

Then, you can use this list to [un]install corresponding packages on local or remote installation.

In addition to this post, those instructions allow to get exactly the same installed packages between GNU/Linux installations.

Install the same packages between two GNU/Linux installation, using yum

2008-05-13T18:16:00.011+02:00

There is some cases where it is interesting to ensure having the same installed packages between two GNU/Linux installation.
Those installations must be of same architecture and be as near as possible (ideally the same version), otherwise it is obvious it won't work. In addition, it is needed to have the same rpm packages sources (for instance the same repositories if using yum).
For instance, it gives the ability to create a virtual machine with almost the same configuration, and then to perform lots of "tests" (upgradibility ...) without risk (but the one to get a "dead virtual machine") before performing the same instructions on the "original" installation.
Personally, I've used this ability to test upgradibility between Fedora 8 and Fedora 9.
To get the list of installed packages of an installation (including the architecture, which is particularly important there):
rpm -qa --qf "%{NAME}.%{ARCH} " > /tmp/allPackagesList

Then copy the file to the other installation.

Finally, use yum (or something else), to install all those packages (for instance, under GNU/Bash):
yum install $( cat /tmp/allPackagesList )

Such a way, all the packages installed on the first installation, will be installed on the other one.
Nevertheless, there may be other packages already installed on the second one, so the installed packages would not be exactly the same.
I'll write another post to explain how to remove such packages.

Personal Knowledge Base

2008-05-13T18:10:00.002+02:00

I've recently used Trac to create a little personal Knowledge Base.
The only aim of this Knowledge Base is to centralize access to various quick information corresponding to specific computer science issues/situations.

Initially, I've set up this system for personal purposes.
Nevertheless, if you find current tickets (Knowledge Base elements) interesting and you wish to add some yourself, send me an email to admin[dot] knowledgeBase[at]bsquare[dot]no-ip[dot]org.

There is currently very few tickets (lack of time), but as soon as possible, I'll add more.

Get sound working with plugins for Firefox or Opera 9 under Fedora 8 (i386 and x86_64)

2008-05-12T20:35:00.003+02:00

In addition to plug-ins installation (see this post), it is needed to ensure having an installed soundwrapper to get sound working under a Web Browser.
Whatever the architecture of the OS, it seems absolutely needed to have a i386 installed soundwrapper.
For instance, install the arts (KDE) soundwrapper:
yum install arts.i386

Install multimedia plugins for Opera 9 under Fedora 8 (i386 and x86_64)

2008-05-08T23:17:00.007+02:00

There should be no problem to get multimedia plug-ins installed for Opera 9 under Fedora with a "full" i386 or x86_64 architecture installation.
In some cases, this post may help (Opera being "Mozilla compatible" as Firefox) with installation having i386 and x86_64 components.
Nevertheless, other situations may still need additional instructions to be performed.

It was my case with my installation: Opera 9.27 (i386), under a Fedora 8 x86_64 installation.
In particular, the gecko-mediaplayer package (replacing the mplayerplug-in one) was available only in x86_64 architecture. The nspluginwrapper method does not seem to work with Opera, so a i386 version of gecko-mediaplayer package is needed.
This is quick way to get it installed:
- get a i386 version, for instance browsing the livna repository online, or using yumdownloader under a i386 installation,
- install it using yum (it will install all the needed dependencies)
yum localinstall "gecko-mediaplayer-XXX.rpm"
- the i386 plug-ins might be installed under /usr/lib/mozilla/plugins

To enable plug-ins into Opera, follow the usual way:
- Menu Tools->Preferences->Advanced->Content->Plug-in options->Change path
- add the /usr/lib/mozilla/plugins path (or equivalent)
- valid, and then click "Find new..." button.

Close and launch again Opera.
You may use the -debugplugin option to get debug information about plug-ins.

Finally, under Opera, take a look to the "opera:plugins" information page to check if plug-ins are available.

Install Opera 9 under Fedora 8 (i386 and x86_64)

2008-05-07T21:46:00.008+02:00

Those instructions might be interesting for other configuration too (for instance Fedora 7 and prior).
The GNU/Linux Opera 9 package does not seem providing all needed dependencies information. Such a way, even if the rpm is well installed, there might be trouble to launch Opera.

To begin, to install Opera:
- download the rpm of the wished version, Opera 9.27 being currently the latest stable one;
- install it with yum to get the maximum (but not all) needed packages installed too:
yum --nogpgcheck localinstall "opera-XXX.rpm"
- ensure to have qt package installed (i386 for v9.27 and prior; x86_64 might be enough for newer version)
- ensure to have a up to date Java installation to benefit from the maximum functionalities (again, i386 for v9.27 and prior; x86_64 might be enough for newer version)
- it might be needed to update the LD_LIBRARY_PATH to specify where to find some libraries: libqt-mt.so (qt), libjvm.so and libawt.so (Java):
For instance, if you have installed the i386 JDK 6u6 under /usr/share/java/jdk1.6.0_06.i386/
export LD_LIBRARY_PATH=/usr/lib64:/usr/lib/qt-3.3/:/usr/share/java/jdk1.6.0_06.i386/jre/lib/i386/client/:/usr/share/java/jdk1.6.0_06.i386/jre/lib/i386/

Compile VMwareTools on Linux kernel 2.6.22

2008-05-07T15:12:00.005+02:00

To compile the vmhgfs VMwareTools on Linux kernel 2.6.22, it is needed to patch the source code, like explained there, from a tar.gz installation.
The same fix works from rpm installation, although the VMwareTools source code directory may be different (/usr/lib/vmware-tools/modules/source for me):
- move to /usr/lib/vmware-tools/modules/source/
- untar vmhgfs.tar and backup it
- then follow the same instructions:
"Now edit vmhgfs/driver.c and change lines 44 and 45 from
#define INODE_SET_II_P(inode, info) do { (inode)->u.generic_ip = (info); } while (0)
#define INODE_GET_II_P(inode) ((HgfsInodeInfo *)(inode)->u.generic_ip)
to
#define INODE_SET_II_P(inode, info) do { (inode)->i_private = (info); } while (0)
#define INODE_GET_II_P(inode) ((HgfsInodeInfo *)(inode)->i_private)

Also, remove line 763: inode->i_blksize = HGFS_BLOCKSIZE;
"
- create a new vmhgfs.tar with patched source code
- the compilation can now be fully performed -> launch /usr/bin/vmware-config-tools.pl again

Upgrade manually an installed perl module

2008-04-30T18:23:00.003+02:00

When an application requires a minimum perl module version which is not provided by installed RPM, it can be upgraded manually.
For instance, it is the case of amavis (see create a full and secured emails server), when upgrading some perl rpm packages (perl-CPAN for instance).

To upgrade CPAN perl module manually, opens a perl shell:
perl -MCPAN -e shell

Then request the "install" of the newest available perl module version. For instance:
install File::Temp

Define the current version of a perl module

2008-04-30T18:14:00.004+02:00

To get the current version of an installed perl module, the following command can be used:
perl -le 'use MODULE_NAME; print MODULE_NAME->VERSION'
With MODULE_NAME the name of the wished perl module.

For instance with File::Temp it gives:
perl -le 'use File::Temp; print File::Temp->VERSION'

Get the list of gcc defined macro after preprocessing

2008-04-25T19:11:00.004+02:00

It various situation, it is very useful to get the list of all defined maco after gcc preprocessing.
Particularly when porting C/C++ source code to another OS/Architecture to know what macro to use to identify it.

A simple way is to use gcc with the following options:
-E to end after preprocessing step,
-dM to print all defined macros.

For instance to known the "default" macros, use those options with a simple source code file "test.c" defining a main function:
gcc -E -dM test.c

Fix empty /var/log/boot.log file issue

2008-02-23T00:19:00.004+01:00

The local7 facility is used to write into the /var/log/boot.log file.
But since the Fedora core 4, the system has been disabled waiting for a new generation system which does not seem to be available yet (see this bug report and this one)

A quick way to fix the empty /var/log/boot.log file, is to edit the /etc/rc.d/init.d/functions file.
Into the success, failure, passed and warning functions, you can add a line like the following, just before the return instruction:
/usr/bin/logger -p local7.info "$1"

It works but each time the initscripts package is updated, it is needed to edit the file ...

Make Red Hat Graphical Boot detailed

2008-02-23T00:12:00.003+01:00

There is unfortunately no configuration file allowing to switch on/off the details of Red Hat Graphical Boot.
Anyway, for it to be detailed, the /etc/rc.d/init.d/functions can be edited:
- simply add --details=yes into the update_boot_stage function on the line using the /usr/bin/rhgb-client.

It works but each time the initscripts package is updated, it is needed to edit the file ...

Updating courier-imap configuration

2008-02-22T23:40:00.003+01:00

By default, the courier-imap creates pop3 (110), pop3s (995), imap (143) and imaps (993) servers.
Generally, they are not all needed, and for obvious security reasons, it is better to not start useless servers.

Under GNU/Linux Fedora, the corresponding configuration files, which are under /usr/lib/courier-imap/etc/, are respectively pop3d, pop3d-ssl, imapd and imapd-ssl.
Into each, there is a variable XXXSTART (with XXX corresponding to POP3D, POP3DSSL, IMAPD or IMAPDSSL) which can be YES or NO.

To not start such a server, it is enough to replace corresponding YES by NO.

Finally, restart the courier-imap service.

Upgrade GNU/Linux Fedora from core 6 to 7, then 8

2008-02-22T20:31:00.006+01:00

To upgrade GNU/Linux Fedora from core 6 to 7, then 8, there is no [really] problem today.
It was not the case when I've tried it as soon as Fedora 7 was available, but I've successfully performed such an upgrade without difficulties some days ago.

The principle is globally the same when upgrading from Fedora core N to Fedora core N+1 (see this post), or from Fedora 7 to Fedora 8 (see this post).
In addition, the recommendations of Fedora project have greatly evolved and seem now complete.

This is some additional instructions in case you have specific issues (like I had):
Update your /etc/fstab file:
It is very important to use label (LABEL=myLabel) to identity your device instead of path (/dev/xxx). The identification has changed from Fedora core 6 to Fedora 7, generally from /dev/hdX to /dev/sdY BUT there is absolutely no certainties that X and Y will equal between versions.

Using labels ensure your mount points are always the same.
Particularly it ensures your root partition (/) will be the good, and your computer will well boot after upgrade.
To update the label of a device, you can use the tune2fs command.

Upgrade from Fedora core 6 to Fedora 7:
- do not forget to clean all the yum metadata with yum clean all,
- upgrade the Fedora release:
rpm -Uhv ftp://download.fedora.redhat.com/pub/fedora/linux/releases/7/Fedora/i386/os/Fedora/fedora-release*.noarch.rpm
- upgrade your repository, for instance for livna:
rpm -Uvh http://rpm.livna.org/livna-release-7.rpm
- remove and reinstall authconfig to avoid specific issue.

Upgrade from Fedora 7 to Fedora 8:
- do not forget to clean all the yum metadata with yum clean all,
- upgrade the Fedora release:
rpm -Uvh http://mirror.anl.gov/pub/fedora/linux/releases/8/Everything/i386/os/Packages/fedora-release-*.noarch.rpm
- upgrade your repository, for instance for livna:
rpm -Uvh http://rpm.livna.org/livna-release-8.rpm

Ensure there is no dependencies problem like explained into this post.

Then, you should perform a great configuration files merging campaign to ensure having the up to date functionalities while keeping your own specific configuration (globally the XXX.conf and XXX.conf.rpmnew files).

Clean useless packages under 64bits GNU/Linux

2008-02-22T16:50:00.004+01:00

It seems there is some dependency "issues" when installing a 64bits GNU/Linux. It is, for instance, the case under Fedora.
Although the architecture of the installed OS is 64bits, there is some i386 or i686 installed packages.
Some are needed, but lots might be removed.

To begin, to get a sorted list of such potential packages, you can use:
rpm -qa --qf "%{NAME}.%{ARCH}\n" |grep -v x86_64 |grep -v noarch |grep -v "(none)" |sort

Then, you can remove all packages you are sure you do not need, but be careful with the dependencies.
To make it quicker, you can request the remove of package(s) you know it is needed by lots of others (like some libraries).

After all of this, ensure there is no dependencies problem. For instance, with yum utilities:
- ensure the database is up to date with rpm --rebuilddb,
- check if there is problem with package-cleanup --problems.

Finally, to complete the cleaning, you can use the package-cleanup --leaves and ensure you need all those "independent" package(s), else remove them.

Personally, I have removed about 150 packages like this.
For information, there is some i386/i686 packages which are needed by x86_64 ones.
It was the case for me with compat-libstdc++-33.i386, libgcc.i386 and glibc.i686.

Get 32bits firefox plugins available under 64bits system

2008-02-18T13:18:00.005+01:00

There is various article about this, but I have not found all needed information into only one.
This is a quick method to get 32bits firefox plugins available under 64bits system:
- install wished 64bits plugins,
- install 32bits plugins if there is no 64bits version (like for macromedia flash),
- install nspluginwrapper 64 and 32 bits (important to have the two versions),
- ensures 32bits plugins are available under /usr/lib/mozilla/plugins (or create symbolic links),
- ensures 64bits plugins are available under /usr/lib64/mozilla/plugins (or create symbolic links),
- according to your OS, it may then be enough to restart firefox (like under Fedora 8),
- else, use the mozilla-plugin-config -i command.

It is possible to not create symbolic link, and to specify path of corresponding .so file with the mozilla-plugin-config -i.

Each wrapped plugins should be under
/usr/lib64/mozilla/plugins-wrapped (or maybe equivalent). If it is not the case, you can use the -v option while using mozilla-plugin-config.

Finally, under Firefox, you should see all available plugins under "about:plugins".

Reinstall packages from another source/Vendor without broking dependencies

2008-02-11T13:13:00.000+01:00

There is various situations leading to the wish to reinstall packages from another source/Vendor without broking dependencies.
It is not so easy when continuity of services is an important need.
To avoid difficulties, try to make it with iterative manner, "package set" by "package set".

This is the steps which can be followed for each iteration:
- define with the greatest precision the packages which must be updated (this post can help for this),
- define the dependencies which might be broken while removing regarded packages (this post can help for this),
- ensure there is another source for each package which must be reinstalled, for instance with yum:
yum search packageName1 ... packageNameN
- with great caution, remove the packages regardless to dependencies:
rpm -e packageName1 ... packageNameN --nodeps
- finally, install the new version from wanted source, for instance with yum:
yum --disablerepo=unwantedSource install packageName1 ... packageNameN
- ensure there is no problem, particularly about dependencies:
package-cleanup --problems

As final check, restart the services which may have been affected to ensure there is no warning/error.

Split a file into several ones with dd

2008-02-11T13:08:00.000+01:00

To split quickly a file in several ones (for greatest management, better legibility ...), the dd command can be used.

To create the /path/to/my/destination/file with the first N MBytes of the /path/to/my/source/file:
dd if=/path/to/my/source/file of=/path/to/my/destination/file bs=1048576 count=N

Manage services under Solaris

2008-02-11T13:06:00.000+01:00

A simple refresh of what can be done with svcs and svcadm tools.

Get list of all services:
svcs

Add a service:
To begin, a manifest file is needed to define the service (some information there).
Then, it is needed to ensure it is valid:
xmllint --valid /path/to/my/service/manifest/file

Finally, the service can be add:
svccfg import /path/to/my/service/manifest/file

Activate a service (which has been priorly added):
svcadm enable * serviceName
The -s option can be used to get focus only after the service is fully activated (or failed).

Disable a service:
svcadm disable * serviceName
The -s option can be used to get focus only after the service has been disabled.

Put a service under maintenance:
svcadm mark * serviceName

Remove the "maintenance" status of a service:
svcadm clear * serviceName

Update the configuration of a service:
svcadm refresh * serviceName

Get the current status of a service:
svcs -x serviceName stop

Manage services under GNU/Linux distributions providing service and chkconfig

2008-02-11T12:57:00.001+01:00

A simple refresh of what can be done with service and chkconfig tools.

Get list of all services and their state for each runlevel (including inet and/or xinet if available):
chkconfig --list

Get state for each runlevel of one service (including inet and/or xinet if available):
chkconfig --list serviceName

Remove a service:
chkconfig --del serviceName

Add a service (with state for each runlevel defined automatically according to the service definition file):
chkconfig --add serviceName

Add a service with specific state for some runlevels:
chkconfig --level * serviceName on|off**
* one or several runlevel(s) specified without separator
** only "on" as sense there (to add/activate the service)

Define the state for some runlevels of an installed service:
chkconfig --level * serviceName on|off*
* "on" to "activate" the service, "off" otherwise

Get the current status of all services:
service --status-all

Get the current status of a service:
service serviceName status

Start a service:
service serviceName start

Stop a service:
service serviceName stop

Restart a service:
service serviceName restart

Request DEB packages ("Debian-like" GNU/Linux distribution)

2008-02-11T12:44:00.001+01:00

A simple refresh of what can be done with dpkg tool.

All those commands query installed packages.

Get the list of packages:
dpkg -l

Get the list of packages matching a specific pattern:
dpkg -l pattern
For instance for libraries: dpkg -l lib*

General information about a package:
dpkg -s myPackageName[-myCompletePackageVersion]

Get the list of files of a package:
dpkg -L myPackageName

Get the package providing a specific file:
dpkg -S myFilePathOrPattern*
*: absolute path or pattern

To get the general information of the package providing a specific file (under GNU/Bash) with only one command line:
dpkg -s $( dpkg -S myFilePath |awk '{print $1}' |sed -e 's/://g;' )

Request RPM packages ("Redhat-like" GNU/Linux distribution)

2008-02-11T12:22:00.000+01:00

Request RPM packages ("Redhat-like" GNU/Linux distribution)
A simple refresh of what can be done with rpm tool.

To begin, the -q option allow to select the RPM packages to request.
The -a option allow to select all (potentially with a pattern to match).
Name, pattern and even version can be specified for greatest precision in selection.

All those commands query installed packages.

Get the list of packages:
rpm -qa

Get the list of packages matching a specific pattern:
rpm -qa pattern
For instance for X libraries: rpm -qa lib*X*

General information about a package:
rpm -qi myPackageName[-myCompletePackageVersion]
For instance, for a specific kernel: rpm -qi kernel-2.6.20-1.2962.fc6

Get the list of files of a package:
rpm -ql myPackageName[-myCompletePackageVersion]

Get ChangeLog of a package:
rpm -q --changelog myPackageName[-myCompletePackageVersion]

Get the list of packages needed by another package:
rpm -qR myPackageName[-myCompletePackageVersion]

Get the list of what provides a package:
rpm -q --provides myPackageName[-myCompletePackageVersion]

Get the list of packages providing a specific functionality:
rpm -q --whatprovides functionalityName
For instance, for the perl functionality "Authen::SASL": rpm -q --whatprovides "perl(Authen::SASL)"

Get the package providing a specific file:
rpm -qf myFilePath*
*: the path can be relative to current directory or absolute

To get the general information of the package providing a specific file (under GNU/Bash) with only one command line:
rpm -qi $( rpm -q --file myFilePath )

Define what is the installed GNU/Linux distribution

2008-02-11T12:05:00.000+01:00

The "release" part of the Linux Standard Base (LSB) allow answering this question.

Tips:
- to install it under GNU/Linux Fedora, use yum install redhat-lsb
- to install it under GNU/Linux Debian, use apt-get install lsb-release

Then, simply use the command lsb_release -i.

This is output instances:
- under GNU/Linux Fedora, it answers "Distributor ID: Fedora"
- under GNU/Linux Debian, it answers "Distributor ID: Debian"

Define installed packages of specific Vendor

2008-01-27T18:07:00.000+01:00

When lots of things are installed on a computer, it is sometimes difficult to define what are the installed packages of a specific vendor.
This is a little command allowing to get such Information :
rpm -qa --qf "%-30{NAME}\t%{VENDOR}\n" |grep -i VENDOR_NAME

VENDOR_NAME must be replaced by what you want ("Red Hat" for instance).

The opposite can be done. To define what are the installed packages which are not provided by a specific vendor :
rpm -qa --qf "%-30{NAME}\t%{VENDOR}\n" |grep -v VENDOR_NAME

Fix setup of pulseaudio under GNU/Linux Fedora 8

2007-11-18T15:38:00.000+01:00

It is important to ensure that pulseaudio is well installed, and used as wished.
After having installed the kde-settings-pulseaudio package (and perhaps something else ?), I have had issue with sound support.
There is lots of interesting information on the "The Perfect Setup" page of official pulseaudio site.

According to my needs, I have followed the instructions about "ALSA Applications" and "KDE" (mplayer and xmms works well without the given instructions).

Unfortunately, I have had still problems, and found that several other users have too.
The solution can be found into this topic:
- simply add the MODE="0666" mention after some instructions of the /etc/udev/rules.d/40-alsa.rules file,
- ensure there is no more /tmp/pulse-XXX (with XXX username) directory with not rights enough.

But it is not enough neither, additional information can be found on this topic, it solves my problem one for good.

Hang problems under GNU/Linux Fedora 8

2007-11-17T23:12:00.000+01:00

It seems there is several situations leading to kernel hang under GNU/Linux Fedora 8.
The bug 283161 report gives lots of information about solution.
For instance, this is kernel options which can be used (not all together):
- nohz=off highres=off
- clocksource=acpi_pm
- nolapic_timer

Unfortunately, I have tests all "sets" but without success.
Lots of tests and several discussions lead mind to potential graphical driver issue.
As soon as I have replaced use of radeon by vesa, I had no more problem (but I have had no time enough to keep on testing).
To benefit from my graphical card efficiencies, I have installed the last Fedora 7 radeon drivers version and have still no more issue (after having switched back use of those radeon drivers).

This is only a temporarily fix until an official solution is given.
I will report as soon as I would know about it.

Upgrade GNU/Linux Fedora from 7 to 8

2007-11-17T22:37:00.000+01:00

Upgrade from Fedora 7 to Fedora 8 can be done with the same method I have explained into another post.
In addition, there is recommendations of Fedora project.
Eventually, this is additional instructions allowing to find potential issues with installed packages (need yum-utils package):
package-cleanup --problems
to find problems
package-cleanup --orphans
to list installed packages which are not available from currently configured repositories. It allows to find orphan installed packages which may be removed now.

Personally, I have had hang problems after upgrade (independent from it).
My next post will be linked to it, and known solutions.

Make Microsoft Windows stops swapping

2007-10-09T18:58:00.000+02:00

I've recently found something which should be very useful in various situation.
A register key to update to disable the swap of drivers and system.
It might be interesting after having minimized an application and let it during some times, for instance.
In other words, a performance improvement VS some more memory use.

I almost do not use Microsoft Windows anymore, so let me know if it does something for you ...

Remove network interface under Solaris

2007-09-23T01:25:00.001+02:00

Under Solaris, it is very more easy to remove than to add. The network interface can be found thanks to the couple (name, inet address):
N.B.: All checks have been removed for better legibility, but it is very important to check returned code after each request.

// We consider the existence of the variables (unsigned char *) "networkInterfaceName", "inetAddress" which could be respectively "eth0" and "192.168.80.1" for instance.

int socketDescriptor;
struct sockaddr_in *addr;
struct lifreq lifr;

sd = socket(AF_INET, SOCK_DGRAM, 0);

memset(&lifr, 0, sizeof(lifr));
addr = (struct sockaddr_in *) &(lifr.lifr_addr);
strncpy(lifr.lifr_name, networkInterfaceName, sizeof(lifr.lifr_name));
addr->sin_family = AF_INET;
addr->sin_addr.s_addr = inet_addr(inetAddress);
ioctl(socketDescriptor, SIOCLIFREMOVEIF, (caddr_t) &lifr)
close(socketDescriptor);

Remove network interface under GNU/Linux

2007-09-23T01:18:00.000+02:00

It is enough to switch down the corresponding alias network interface:
N.B.: All checks have been removed for better legibility, but it is very important to check returned code after each request.

// We consider the existence of the variable (unsigned char *) "aliasInterfaceName" ("eth0:myAlias" for instance).
int sd;
struct ifreq ifr;

sd = socket(AF_PACKET, SOCK_RAW, htons(ETH_P_ALL));

memset(&ifr, 0, sizeof(struct ifreq));
strncpy(ifr.lifr_name, aliasInterfaceName, sizeof(ifr.ifr_name));
ifr.ifr_flags = ~IFF_UP;
ioctl(sd, SIOCSIFFLAGS, &ifr);
close(sd);

Add network interface under Solaris

2007-09-23T01:13:00.000+02:00

Under Solairs, it is not trivial to add a network interface, there is no alias, and it seems to not be possible to specify the additional network interface name (so it must be got and kept).
N.B.: All checks have been removed for better legibility, but it is very important to check returned code after each request.

// We consider the existence of the variables (unsigned char *) "networkInterfaceName", "inetAddress", "broadcastAddress" and "netmaskAddress" which could be respectively "eth0", "192.168.80.1", "192.168.80.255" and "255.255.255.0" for instance.

int sd;
struct sockaddr_in *addr;
struct lifreq lifr, lifrBroadcast, lifrNetmask, lifrFlags;
const char *aliasName;

// Resets structure.
memset(&lifr, 0, sizeof(lifr));
strncpy(lifr.lifr_name, networkInterfaceName, sizeof(lifr.lifr_name));

sd = socket(AF_INET, SOCK_DGRAM, 0);
ioctl(sd, SIOCLIFADDIF, (caddr_t) &lifr);

// Very important to get the alias name to update network interface parameters.
aliasName = lifr.lifr_name;

// Manages broadcast.
memset(&lifrBroadcast, 0, sizeof(lifrBroadcast));
addr = (struct sockaddr_in *) &(lifrBroadcast.lifr_broadaddr);
strncpy(lifrBroadcast.lifr_name, aliasName, sizeof(lifrBroadcast.lifr_name));
addr->sin_family = AF_INET;
addr->sin_addr.s_addr = inet_addr(broadcastAddress);
ioctl(sd, SIOCSLIFBRDADDR, (caddr_t) &lifrBroadcast);

// Manages netmask.
memset(&lifrNetmask, 0, sizeof(lifrNetmask));
addr = (struct sockaddr_in *) &(lifrNetmask.lifr_addr);
strncpy(lifrNetmask.lifr_name, aliasName, sizeof(lifrNetmask.lifr_name));
addr->sin_family = AF_INET;
addr->sin_addr.s_addr = inet_addr(netmaskAddress);
ioctl(sd, SIOCSLIFNETMASK, (caddr_t) &lifrNetmask);

// Defines the address of the new interface.
addr = (struct sockaddr_in *) &(lifr.lifr_addr);
addr->sin_family = AF_INET;
addr->sin_addr.s_addr = inet_addr(inetAddress);
ioctl(sd, SIOCSLIFADDR, (caddr_t) &lifr);

// Sets the new interface UP.
memset(&lifrFlags, 0, sizeof(lifrFlags));
strncpy(lifrFlags.lifr_name, aliasName, sizeof(lifrFlags.lifr_name));
ioctl(sd, SIOCGLIFFLAGS, (caddr_t) &lifrFlags)
lifrFlags.lifr_flags |= IFF_UP;
ioctl(sd, SIOCSLIFFLAGS, (caddr_t) &lifrFlags);
close(sd);

Add network interface under GNU/Linux

2007-09-23T01:07:00.000+02:00

It is not so hard to perform it under GNU/Linux with alias.
The aim, is to add an alias to an existing network interface with the wished inet, broadcast and netmask addresses:
N.B.: All checks have been removed for better legibility, but it is very important to check returned code after each request.

// We consider the existence of the variables (unsigned char *) "aliasInterfaceName", "inetAddress", "broadcastAddress" and "netmaskAddress" which could be respectively "eth0:myAlias"; "192.168.80.1", "192.168.80.255" and "255.255.255.0" for instance.

int sd;
struct sockaddr_in *addr;
struct ifreq ifr, ifrBroadcast, ifrNetmask;

sd = socket(AF_PACKET, SOCK_RAW, htons(ETH_P_ALL));

// Add alias.
memset(&ifr, 0, sizeof(ifr));
strncpy(ifr.lifr_name, aliasInterfaceName, sizeof(ifr.ifr_name));
addr = (struct sockaddr_in *) &(ifr.ifr_addr);
addr->sin_family=AF_INET;
addr->sin_addr.s_addr=inet_addr(inetAddress);
ioctl(sd, SIOCSIFADDR, &ifr);

// Manages broadcast.
memset(&ifrBroadcast, 0, sizeof(ifrBroadcast));
strncpy(ifrBroadcast.lifr_name, aliasInterfaceName, sizeof(ifrBroadcast.ifr_name));
addr = (struct sockaddr_in *) &(ifrBroadcast.ifr_addr);
addr->sin_family=AF_INET;
addr->sin_addr.s_addr=inet_addr(broadcastAddress);
ioctl(sd, SIOCSIFBRDADDR, &ifrBroadcast)

// Manages netmask.
memset(&ifrNetmask, 0, sizeof(ifrNetmask));
strncpy(ifrNetmask.lifr_name, aliasInterfaceName, sizeof(ifrNetmask.ifr_name));
addr = (struct sockaddr_in *) &(ifrNetmask.ifr_addr);
addr->sin_family=AF_INET;
addr->sin_addr.s_addr=inet_addr(netmaskAddress);
ioctl(sd, SIOCSIFNETMASK, &ifrNetmask);

close(sd);

Get MAC address under GNU/Linux or Solaris

2007-09-23T00:33:00.000+02:00

This is a way to get MAC address of a network interface in C language.
Under GNU/Linux it can be done thanks to a RAW socket:
N.B.: All checks have been removed for better legibility, but it is very important to check returned code after each request.

// Consider a variable "networkInterfaceName" of type char * is defined ("eth0" for instance).
int sd;
struct ifreq ifr;
const unsigned char *hardwareAddress;

sd = socket(AF_PACKET, SOCK_RAW, htons(ETH_P_ALL));
memset(&ifr, 0, sizeof(ifr));
setNameInIfreq(&ifr, networkInterfaceName);
hardwareAddress = (unsigned char *) &ifr.ifr_hwaddr.sa_data;
close(sd);

Under Solaris, it begins to be less easy, we need to use DLPI code:
// Consider a variable "devicePath" of type char * is defined ("/dev/eth0" for instance).
int sd;
dl_phys_addr_req_t dlpareq;
dl_phys_addr_ack_t *dlpaack;
struct strbuf msg;
char buf[128];
int flags = 0;
const unsigned char *hardwareAddress;

sd = open(devicePath, O_RDWR));
dlpareq.dl_primitive = DL_PHYS_ADDR_REQ;
dlpareq.dl_addr_type = DL_CURR_PHYS_ADDR;
msg.buf = (char *)&dlpareq;
msg.len = DL_PHYS_ADDR_REQ_SIZE;
putmsg(sd, &msg, NULL, 0);
dlpaack = (dl_phys_addr_ack_t *)buf;
msg.buf = (char *)buf;
msg.len = 0;
msg.maxlen = sizeof (buf);
getmsg(sd, &msg, NULL, &flags);
memcpy(hardwareAddress, &buf[dlpaack->dl_addr_offset], dlpaack->dl_addr_length);
close(sd);

Computer science Experience Knowledge SHAring

2007-09-22T00:29:00.000+02:00

For more understanding, the blog is now named "Computer science Experience Knowledge SHAring", which could be shortly called CEKSHA.

Upgrade GNU/Linux Fedora

2007-09-15T00:26:00.000+02:00

Upgrade from Fedora core N to Fedora core N+1 can be done thanks to yum, although it is not really recommended.
I have successfully performed such upgrade from core 3 (perhaps core 2, I'm not sure), to core 4, then core 5 and finally core 6. Upgrade from Fedora core 6 to Fedora 7 fails with this way because of great distribution architecture update (core and extras merged).

This is steps which can be followed to perform such upgrade:
- open a terminal in run level 3,
- temporary set default run level to 3 (edit /etc/inittab file) to avoid potential problem with card detection or something else during next boot,
- ensure to have the last version of yum thanks to yum upgrade yum,
- clean the yum meta-data to avoid some rare conflict yum clean all,
- upgrade the release thanks to fedora-release (and fedora-release-notes from Fedora core 6) package(s)
For instance, from core 5 to core 6 : rpm -Uvh fedora-release fedora-release-notes
- finally upgrade the full system, ensuring to keep information about all progress into a file, thanks to yum upgrade 2>&1 |tee /tmp/systemUpgrade.log
- fix potential little problem like missing symbolic link (I had one about i686-redhat-linux-gnu one time),
- think to upgrade depositories if needed (livna, dries, dag ...),
- update the /etc/grub.conf or /etc/lilo.conf file according to your environment if you want to ensure the default use of a kernel,
- reboot.

How to choose a *nix shell

2007-09-08T20:56:00.000+02:00

Today, there is very lots of shells like sh, bash, csh, tcsh, ksh, zsh, ash ...
For various reasons, it makes sens to choose one, a time for all, although it is not so easy.
Personally, I have first chosen tcsh for its functionalities and its legibility very near the C language. On the first hand, it was not a bad choice because thanks to it, I had done all I had needed, on the other hand I had faced two issues.
The first was the fact that function does not exists under tcsh, and that alias become dirty very quickly.
The second, the more important, is this is not installed by default on all *nix operating systems.

The wish to create something which can be used on lots of operating system, particularly on GNU/Linux, Solaris, and Cygwin which was my first shell multi-OS need, my second and definitive choice has been for GNU/Bash. It resolves the two points and it perfectly fits my needs.

To lead me on this choice, I have read lots of articles and Web Site, and I have concluded that GNU/Bash is now embedded as installation base on almost all operating systems. In the case it would not, it can be used as sh compatible.

It is why, I think it is the best choice in most cases. If you have not made your choice, you might make this one too, it will be a guarantee of perenity and should answer to all your needs.

Compile OpenVPN and TUN under Solaris

2007-09-04T20:59:00.000+02:00

I've recently worked on Solaris and need to compile the source code of openVPN and TUN.
At first, it seemed easy but because of some compatibility issues, it was not.
You can follow those steps to get openvpn executable under Solaris:
- download the openVPN source code and uncompress it under [OPENVPN_DIRECTORY],
- download the TUN source code and uncompress it under [TUNE_DIRECTORY],
- copy the [TUNE_DIRECTORY]/solaris/if_tun.h file to [OPENVPN_DIRECTORY],
- edit the [OPENVPN_DIRECTORY]/tun.c file and add the #include "if_tun.h" line after the #include "tun.h" one,
- under [OPENVPN_DIRECTORY], execute ./configure --disable-lzo,
- ensure there is a make executable into your path, for instance create a symbolic link executing ln -s /usr/sfw/bin/gmake /usr/bin/make,
- then execute make under [OPENVPN_DIRECTORY],
- finally the openvpn executable is ready for Solaris.

Then, you can follow those steps to get the tun functionality under Solaris:
- create the subdirectory [TUNE_DIRECTORY] /solaris/sys,
- download the dditypes.h file and put it under [TUNE_DIRECTORY] /solaris/sys (it fixes an incompatibility issue),
- ensure there is a ld executable into your path, for instance create a symbolic link executing ln -s /usr/sfw/bin/gld /usr/bin/ld,
- execute make install under [TUNE_DIRECTORY],
- finally the tun functionality is ready for Solaris.

It is now possible to use openVPN and TUN under Solaris.

Create a full and safe emails server under GNU/Linux

2007-09-01T20:41:00.000+02:00

Not so easy ? right.
Anyway, almost one year ago, I've successfully followed a very interesting How To Forge Tutorial which allows me to have such a fully functional emails server. Because it is for my own purpose, there is only few users using it, and I so don't know how must it is scalable.

This emails server is very interesting thanks to its completeness and security. The system is compounded of :
- postfix for the MTA layer,
- courier-imap for mailboxes,
- clamAV, openssl, openldap, cyrus-sasl and courier-authlib-mysql for authentication and anti-virus,
- mysql for users configuration and administration,
- spamassassin, dcc, razor, pyzor and amavis-new for spamfilters.

It works perfectly under Fedora core 5, and Fedora core 6 with my additional instructions.
During my documentation research in 2006, I have read articles enough to think that this system should work on several others GNU/Linux distributions, although packages name/configuration may differ.

Introduction

2007-08-31T23:00:00.000+02:00

Hi everybody,

The aim of this blog is simply to share knowledge earned into design/development/technical delicate and/or difficult situations in which loneliness takes all its meaning.
Like lots of other people, in my personal and professional activities, I face such situations like very specific shell scripts needs, object oriented programming, native source code under GNU/Linux, Sun Solaris ...
Documentation, source code and simply theories are sometimes so difficult to find that it would be a waste to not share acquired knowledge.
I'll try to share such information as regularly as time will allow ...

All the comments are welcomes. Do not hesitate to interact with me.