Latest Publications

Howto to install actiTIME on a Debian server

actiTIME Basic is a free easy-to-use timesheet optimized for your basic time tracking needs. It provides you with functionality for:

  • time tracking and reporting
  • user schedule and overtime control
  • project and task management

For more information visit the website: http://www.actitime.com/fw_features.html

This article describes how to install actiTIME on a Debian server or a Debian derived OS such as Ubuntu. actiTIME comes in three flavors and one of them is freeware ;-) That’s the one we are going to install.

So let’s start!

These are the requirements for actiTIME:

  • Java 2 SDK 1.4.2 or higher
  • Apache Tomcat Application Server 4.1 or higher
  • MySQL database, it will need MySQL 4.1.17+, 5.0.x or 5.1.x installed

Install Java
Add “non-free” to the Debian Lenny repositories (/etc/apt/sources.list)

  1. apt-get update && apt-get install sun-java6-jdk
  2. update-java-alternatives -s java-6-sun
  3. echo 'JAVA_HOME="/usr/lib/jvm/java-6-sun"' | tee -a /etc/environment

Install Tomcat
Install it by running:

  1. apt-get install tomcat5.5 tomcat5.5-admin

Change /etc/tomcat5.5/tomcat-users.xml

  1. <?xml version='1.0' encoding='utf-8'?>
  2. <tomcat-users>
  3.   <role rolename="admin"/>
  4.   <role rolename="manager"/>
  5.   <role rolename="tomcat"/>
  6.   <user username="tomcat" password="Some-Obscure-Password" roles="admin,manager,tomcat"/>
  7. </tomcat-users>

Also edit /etc/default/tomcat5.5 and change the value from “yes” to “no” and restart Tomcat.

  1. TOMCAT5_SECURITY=no
  2. /etc/init.d/tomcat5.5 restart

Install mysql-server and create database
If you don’t already have a running mysql server then install it by running:

  1. apt-get install mysql-server

Then setup a database, database user and give it some rights.

  1. mysql -uroot -p
  2. CREATE DATABASE actitime;
  3. CREATE USER 'actitime-usr'@'localhost' IDENTIFIED BY 'Another-Obscure-Password';
  4. GRANT ALL ON actitime.* TO 'actitime-usr'@'localhost';
  5. flush privileges;
  6. \q

Install actiTIME
Download ActiTime and upload it to your server. (http://actitime.com/download.html

  1. cd /var/lib/tomcat5.5/webapps
  2. tar xvzf /tmp/actitime15_unix.tar.gz
  3. cd actitime
  4. bash ./setup_mysql.sh -mysqlhome /usr -username actitime-usr -password Another-Obscure-Password -host localhost -dbname actitime

Settings file (db name and user including pwd) is found in /var/lib/tomcat5.5/webapps/actitime/WEB-INF/classes/com/actimind/actitime/AT.Properties

You should now be able to connect to your actiTIME. Point a browser to http://your.host.name:8180/actitime

Now I have a bit difficulty with this setting. This means that I have to shoot a another hole in my vlan ACL. Let’s make this happen on port 80 shall we!

Install libapache2-mod-jk
Connect Apache2 to Tomcat5.5.

  1. apt-get install libapache2-mod-jk

Create /etc/apache2/workers.properties file with the following content:

  1. #
  2. # This file provides minimal jk configuration properties needed to
  3. # connect to Tomcat.
  4. #
  5. # We define a worked named 'default'
  6. #
  7.  
  8. workers.tomcat_home=/var/lib/tomcat6
  9. workers.java_home=/usr/lib/jvm/java-6-sun
  10. ps=/
  11. worker.list=default
  12.  
  13. worker.default.port=8009
  14. worker.default.host=localhost
  15. worker.default.type=ajp13
  16. worker.default.lbfactor=1

Then add the following to /etc/apache2/apache2.conf

  1. # Where to find workers.properties
  2. JkWorkersFile /etc/apache2/workers.properties
  3.  
  4. # Where to put jk logs
  5. JkLogFile /var/log/apache2/mod_jk.log
  6.  
  7. # Set the jk log level [debug/error/info]
  8. JkLogLevel info
  9.  
  10. # Select the log format
  11. JkLogStampFormat "[%a %b %d %H:%M:%S %Y] "

Finally create a new VirtualHost in /etc/apache2/sites-available and place the following content beteen the VirtualHost tags:

  1. # JkOptions indicate to send SSL KEY SIZE,
  2. JkOptions +ForwardKeySize +ForwardURICompat -ForwardDirectories
  3.  
  4. # JkRequestLogFormat set the request format
  5. JkRequestLogFormat "%w %V %T"
  6.  
  7. # Tomcat serves everything by default
  8. JkMount / default
  9. JkMount /* default
  10.  
  11. # Apache serves the following URLs
  12. JkUnMount /static default
  13. JkUnMount /static/* default

Now enable the VirtualHost and reload Apache2:

  1. a2ensite <site-name> && /etc/init.d/apache reload

So now ActiTime should also be available on http://your.host.name/actitime

Point a browser to the URL and log on with the default credentials. (username: admin / password: manager)
AND CHANGE IT!

Post to Twitter

Ubuntu Hardy Heron – Conky 1.6.1

Conky 1.6.1 on Hardy Heron

Conky 1.6.1 on Hardy Heron

update: also works on Intrepid Ibex!

Step by step howto install Conky 1.6.1 on Ubuntu Hardy Heron.

Conky is a system monitor for X originally based on the torsmo code. Since its original conception, Conky has changed a fair bit from its predecessor. Conky can display just about anything, either on your root desktop or in its own window. Conky has many built-in objects, as well as the ability to execute programs and scripts, then display the output from stdout.

For more information visit http://conky.sourceforge.net/

Currently in the Hardy repository there is version 1.5.1 available. However, this version does have some issues and there are a lot of fixes between 1.5.1 and 1.6.1.

So I decided to install the latest version from source. Lateron I will describe howto create an deb file from source.

For now you need to install some prerequisite software:

  1. sudo apt-get install libxext-dev libxdamage-dev libglib2.0-dev zlib1g-dev libxft2-dev libiw-dev

Download the latest source from the conky website. Untar it and go into source directory. If you want wlan support for using the variables wireless_* then make sure that you enable it like so.

  1. ./configure –enable-wlan
  2.  
  3. conky 1.6.1 configured successfully:
  4.  
  5.  Installing into:   /usr/local
  6.  System config dir: ${prefix}|>/etc
  7.  C compiler flags:         -I/usr/include/freetype2   -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include   -Wall -W
  8.  Linker flags:       -Wl,-O1
  9.  Libraries:         -lrt  -lm -liw  -lX11   -lXext   -lXdamage -lXfixes   -lXft -lfontconfig   -lglib-2.0  
  10.  
  11.  * X11:
  12.   X11 support:      yes
  13.   XDamage support:  yes
  14.   XDBE support:     yes
  15.   Xft support:      yes
  16.  
  17.  * Music detection:
  18.   Audacious:        no
  19.   BMPx:             no
  20.   MPD:              yes
  21.   XMMS2:            no
  22.  
  23.  * General:
  24.   math:             yes
  25.   hddtemp:          yes
  26.   portmon:          yes
  27.   RSS:              no
  28.   wireless:         yes
  29.   SMAPI:            no
  30.   nvidia:           no
  31. eve-online:       no

If you like you can enable other functionalities but you also need to install the dev files. No worries, the configure script will complain when it can not find whats needed!

Now run make to compile source and test it. If you like you can place a copy of mine .conkyrc in your home directory
Try conky by running…

./src/conky -a top_right &

If everything checks out as you predicted ;-) you can run make install to install conky on your system.

Download configuration file. Just rename it to .conkyrc

conkyrc

Post to Twitter

Ubuntu Hardy Heron – Novell ConsoleOne 1.3.6h

update: also works on Intrepid Ibex AND Jaunty Jackalope!

The need for speed! Everyone who uses ConsoleOne for Windows knows that it’s a pain in the ass referring to speed. So, let’s see if it gets better under a linux workstation. I’m using Ubuntu Heron but debian should work as well.

Just follow these simple steps and you have ConsoleOne up and running in no_time.

Screenshot Novell ConsoleOne 1.3.6h

Screenshot Novell ConsoleOne 1.3.6h

Download the latest version from Novell. [I'm using c1_136h-linux.tar.gz 41771356 bytes]

Save the file in a location and extract it:

  1. tar xvzf c1_136h-linux.tar.gz

As you can see the files are all rpm ones. To use them we need to convert them to deb. Another thing, as you can see there is support for more then one language. Since I’m only using the English version I’ve deleted all the *_de, *_es, *_fr, *_it, *_pt and *_ru files.

Now we need to convert them to deb. Make sure that you have `alien` installed. (apt-get install alien)

  1. sudo alien -v -d *.rpm

And install the deb files

  1. sudo dpkg -i *.deb

Copy Icon file to icon location

  1. sudo cp ConsoleOneIcon.png /usr/share/pixmaps/

ConsoleOne is a Java based application so we need to install java preferable from repository.

  1. sudo apt get install openjdk-6-jre

Now lets make sure that this java version is the default one! If not, select this version. (works for me)

  1. update-alternatives config java

At this point we are ready to start ConsoleOne.

  1. sudo /usr/ConsoleOne/bin/ConsoleOne

If it all works well ConsoleOne is started. Now go to “File” and select “Authenticate”.
Give your credentials and the IP-ADDRESS of the server to authenticate with.

ConsoleOne Login screen

ConsoleOne Login screen

And there you go…. your authenticated to the tree!

Post to Twitter

Ubuntu Hardy Heron – Novell Netware

update: also works on Intrepid Ibex!

Setup Novell client for a Ubuntu Heron workstation to connect to a Novell Netware Cluster.

First we need to install some additional software which you can find in the default repository :

  1. sudo apt-get install ncpfs

If you like you can find a complete file list here for what is installed:

  1. less /var/lib/dpkg/info/ncpfs.list

Command and tools:

  1. nw tools:
  2. nwauth      nwborm      nwbpset     nwfsctrl    nwmsg       nwpurge     nwtrustee
  3. nwbocreate  nwbpadd     nwbpvalues  nwfsinfo    nwpasswd    nwrevoke    nwtrustee2
  4. nwbols      nwbpcreate  nwdir       nwfstime    nwpjmv      nwrights    nwuserlist
  5. nwboprops   nwbprm      nwdpvalues  nwgrant     nwpqjob     nwsfind     nwvolinfo  
  6. ncplogin   ncplogout  ncpmap     ncpmount   ncpumount

All of the ncp* command you need root privileges. So when you map (sudo ncpmap) you can not access the files as a regular user! However, if change SUID bit for these files, you can. Take notice that this introduces a security risk!

  1. sudo chmod u+s /usr/bin/ncp*

So, what is the procedure to log-on and map drives to a Novell Netware Server? All of this is very trivial.

!! Change all variables between the [brackets] to your needs!

Step 1: Login to server

  1. sudo ncplogin -S [ServerName]  -A [Host]  -U .[UserName].add.your.context

You could add the ” -P ” flag and provide your password but you don’t want people to shoulder surf and snag your password! We have a cluster called atom, so my definition for both flags -S and -A is atom ;-)
!! Notice that we start which a ” . ” by flag -U

Step 2: Test if are connected:

  1. nwuserlist -S [ServerName
  2. ]

Step 3: Map the drives

  1. sudo ncpmap -S [ClusterResource] -A [ClusterResource] -V [VolumeName] -R [directory_to_map] ~/Novell/data-drive

As you can see I’ve logged-on to the Netware cluster and now I’m mapping to a cluster resource.

Step 4: Logout

  1. ncplogout -a

If you like, here is a little script I’m using to connect to the server:
Syntax ~/nwlogin [no username | username]
If no username is given then your linux username will be default.

  1. #!/bin/bash
  2.  
  3. # Setup directory structure
  4. if [ ! -d ~/Novell ]; then
  5.  mkdir ~/Novell
  6.  chmod 700 ~/Novell
  7. fi
  8.  
  9. if [ ! -d ~/Novell/data-drive ]; then
  10.  mkdir ~/Novell/data-drive
  11.  chmod 700 ~/Novell/data-drive
  12. fi
  13.  
  14. if [ ! -d ~/Novell/appl-drive ]; then
  15.  mkdir ~/Novell/appl-drive
  16.  chmod 700 ~/Novell/appl-drive
  17. fi
  18.  
  19. # Set UserName if none given fetch your current linux loginname
  20. if [ -e $1 ]; then
  21.  echo $1
  22.  NDSUSER=$USER
  23. else
  24.  NDSUSER=$1
  25. fi
  26.  
  27. # Set alternative context
  28. if [ ! -e $2]; then
  29.  CONTEXT=$2
  30. else
  31.  CONTEXT=your.context.here
  32. fi
  33.  
  34. # Close active connections
  35. ncplogout -a > /dev/null
  36. echo "Enter NDS password for $NDSUSER.$CONTEXT"
  37. read -s NDSPASS
  38.  
  39. ncplogin -S [ServerName] -A [Host] -U .$NDSUSER.$CONTEXT -P $NDSPASS -o tcp ||
  40. echo "Couldn't authenticate to NDS server, aborting…"
  41.  
  42. ncpmap -S [ServerResource] -A [ServerResource] -V [VolumeName] -R DriveToMap ~/Novell/data-drive || echo "Couldn't map to your atomdata, aborting…"
  43. ncpmap -S [ServerResource] -A [ServerResource] -V [VolumeName] -R DriveToMap ~/Novell/appl-drive || echo "Couldn't map to your atomappl, aborting…"
  44.  
  45. unset NDSPASS

Cheers,
Raymond.

Post to Twitter

Oracle Calendar on Ubuntu Hardy Heron (8.0.4)

[update apr 2009]:
also works on Debian Lenny. Eric Gerlach mentions the following : If you’re running the amd64 port of Debian or Ubuntu, you have to install ia32-libs and ia32-libs-gtk for the installer to work.


[update dec 2008]: also works on Intrepid Ibex!

Recently I reinstalled my computer at work and missed my Oracle Calendar. Here is a small howto to setup the calendar client.

Download the latest client from the Oracle website. Currently client version is cal_linux_1012.tar.gz (39,732,486 bytes)

Extract the client…

  1. tar xvzf cal_linux_1012.tar.gz

After extraction you find a directory OracleCalendar_inst/ go there and do the following..

  1. ./text_install.sh
  2. Preparing to install
  3. Extracting the JRE from the installer archive…
  4. Unpacking the JRE…
  5. Extracting the installation resources from the installer archive…
  6. Configuring the installer for this system's environment…
  7. nawk: error while loading shared libraries: libm.so.6: cannot open shared object file: No such file or directory
  8. dirname: error while loading shared libraries: libc.so.6: cannot open shared object file: No such file or directory
  9. /bin/ls: error while loading shared libraries: librt.so.1: cannot open shared object file: No such file or directory
  10. basename: error while loading shared libraries: libc.so.6: cannot open shared object file: No such file or directory
  11. dirname: error while loading shared libraries: libc.so.6: cannot open shared object file: No such file or directory
  12. basename: error while loading shared libraries: libc.so.6: cannot open shared object file: No such file or directory
  13. Launching installer…
  14. grep: error while loading shared libraries: libc.so.6: cannot open shared object file: No such file or directory
  15. /tmp/install.dir.25009/Linux/resource/jre/bin/java: error while loading shared libraries: libpthread.so.0: cannot open shared object file: No such file or directory

As you can see the installer produces some errors. That’s because it assumes your kernel version is the one it is set in the install script. So what we need to do is to disable the assumption. First let’s see what the script does…

  1. cat cal_linux | grep LD_ASSUME_KERNEL
  2.  
  3. linux_LD_ASSUME_KERNEL_hack=0;
  4.                                                                 linux_LD_ASSUME_KERNEL_hack=1
  5. # LD_ASSUME_KERNEL for Native POSIX Threading Library on some Linux distros
  6.                 export LD_ASSUME_KERNEL=2.2.5
  7.                 # unset the LD_ASSUME_KERNEL in cause we don't need it
  8.                 unset LD_ASSUME_KERNEL
  9.                 # check our rules for setting LD_ASSUME_KERNEL
  10.                         linux_LD_ASSUME_KERNEL_hack=1
  11. if [ $linux_LD_ASSUME_KERNEL_hack -eq 1 ]; then
  12.         LD_ASSUME_KERNEL=2.2.5
  13.         export LD_ASSUME_KERNEL

As you can see the kernel version that is assumed is version 2.2.5. Doing a `uname -r’ it shows me on a Ubuntu Heron that I’m currently running 2.6.24-19-generic kernel version.
So what we need to do is fool the script so it can not export the variable LD_ASSUME_KERNEL

Run the following to backup / modify / and cleanup the mess you made ;-)

  1. cp cal_linux cal_linux.save
  2. cat cal_linux.save | sed "s/export LD_ASSUME_KERNEL/#xport LD_ASSUME_KERNEL/" > cal_linux
  3. rm cal_linux.save

Just run `cat cal_linux | grep LD_ASSUME_KERNEL’ again to see what changes you made.

Rerun the installer to give it another try…

  1. ./text_install.sh

There is just one more thing we need to do. If you finished the installation you can find `ocal’ in the default directory ~/OracleCalendar/bin. Running this produces another error…

  1. /OracleCalendar/bin$ ./ocal
  2. Starting Oracle Calendar for Linux
  3. /home/raymond/OracleCalendar/bin/Ocal: error while loading shared libraries: libstdc++.so.5: cannot open shared object file: No such file or directory

This is because Oracle Calendar depends on the older version of libstd that is installed on Heron. You can fix this by doing…

  1. sudo apt-get install libstdc++5
  2. The following NEW packages will be installed:
  3.   gcc-3.3-base libstdc++5

That’s it, now you can start the calendar

  1. ~/OracleCalendar/bin/ocal

Cheers,
Raymond

Post to Twitter

Mediawiki Farm – Administer multiple wiki environments

2009/07/28 Update: I’m working on a robust version that handles the diversity of mediawiki’s databases. Therefore the script should work with newer versions of mediawiki.
2008/06/19 Update: Added entry, howto setup VirtualHosts / CNAME
2008/04/28 : version 0.4 is available.

Overview

As we all know, wiki’s are hot. And at the University we’ve got request constantly for facilitating them. BUT, there’s also a big issue and that is the fact that there isn’t any easy administration tool for installing and archiving wiki’s. So, with that in mind I started to work on a solution to administer multiple wiki’s, in this case mediawiki’s.

So, what does it need to do

Well basically, I just want to run a command line script that creates a wiki-directory linked to the source directory for easy updating and create a wiki database and user.

If this is what you want….. then read on ;-)

Server setup and software requirements

The server in question runs Linux Debian Etch. Web service is provided by Apache2, php5 and a mysql server v5 runs at localhost. The scripts I’ve created for this purpose will run on Debian, Ubuntu, Linux Mint, and lots of other Linux flavors. The script uses Debian package makepasswd. Just apt-get it and you’ll be fine. All other commands are pretty default. The mediawiki software that I’m running is version 1.12 from backports.org but it should work fine with 1.7 and I’ve tested successfully with 1.10 (Linux Mint).

If you like here is how you install mediawiki from the debian backports.org.

Install mediawiki from backports.org
  1. $echo deb http://www.backports.org/debian etch-backports main contrib non-free >> /etc/apt/sources.list
  2. If you are using etch and you want apt to verify the downloaded backports you can import backports.org archive’s key into apt:
  3. $apt-get install debian-backports-keyring
  4. $apt-get update
  5. And finally install mediawiki
  6. apt-get -t etch-backports install mediawiki

Global architecture

When you fetch mediawiki from the repo it will be installed in /urs/share/mediawiki. This will be the “source” directory for all your wiki installations. Within the source directory there are some symbolic links to “config”, “extensions” and “images”. Hereby an overview of shared and created directories:

# Main source directory also includes, extensions, skins, languages, etc, etc.

/var/www/new_wiki -> /usr/share/mediawiki

# A separate directory for uploading files. We like to keep this separate for other wiki’s.

/var/www/new_wiki/images

NB. config is not used since we are doing a one click button install.

The script

This is a very basic bash script, easy to understand and to configure. The directory structure is as follows:

/root/mediawiki-setup/ # Base script directory

/root/mediawiki-setup/archive # Directory used when archiving a wiki. Complete wiki html directory, mysql data dump and a mysql cleanup query.

/root/mediawiki-setup/bin # Place where the script “wiki-admin” lives.

/root/mediawiki-setup/conf # Here are the default files like mediawiki database sql script, and LocalSettings.php

/root/mediawiki-setup/reports # While creating a new wiki, specific information about it will be stored in the reports directory. Per wiki it saves LocalSettings.php and the SQL create statements.

You do need to configure at least the following for the wiki-admin script:

WIKI_SOURCE=/usr/share/mediawiki # Change it if it differs from yours.

SQL_PWD=YourVeryStrongPassword # Change pwd and if you like add the server if your not on localhost.

What does it work?

Wiki-admin script wil check and create a directory at /var/www/new_wiki. Then it will create the necessary symbolic links to the source directory. Next it will create the image directory and set it up with appropriate rights. When this is done a sql script will run to create the database, db-user and rights for the db-user. Last step is setting up LocalSettings.php and copy it to the wiki directory.

For archiving just run wiki-admin archive . The script will basically backup your wiki to the archive directory (tgz) and then dump the database save it and finally drop the database and cleanup the database user and his rights.

The “WikiAdmin” account will be installed in every wiki you install. Therefore, change the WikiAdmin password. Of course you can install a test wiki, change the WikiAdmin password, drop the schema, save it in conf directory and use that for your basic wiki installation.

Here’s a code snippet, you can download it below!

#!/bin/bash
  1.  
  2. # Wiki-admin: maintenance script for creating, archiving and listing mediawiki's (http://www.mediawiki.org)
  3. # v 0.4, 2008-04-28 (look for latest version at http://www.jirp.nl)
  4. # Copyright (c) 2008, Raymond Mul (raymond at mul.nl)
  5. # Released under the MIT License (http://www.opensource.org/licenses/mit-license.php)
  6.  
  7. # Description:
  8. # The latest information and a complete manual regarding this script look for it @ http://www.jirp.nl
  9.  
  10. # In short, this script let's you create multiple mediawiki's from one source directory, backup them, archive themand list them.
  11.  
  12. # This script, at the moment, has three functions:
  13. # 1) create : this allows you to setup a mediawiki based on one source directory.
  14. #   – create directory in /var/www/;
  15. #   – create symbolic links to the source;
  16. #   – create images directory for uploading specific for that wiki;
  17. #   – create database, user with a strong password and grant specific rights;
  18. #   – create wiki tables, import sql-dump;
  19. #   – create LocalSettings.php and changes it so it comes wiki specific;
  20. #   – create reports such as database connection string, copy of sql-importand saves a copy of LocalSettings.php;
  21.  
  22. # 2) archive: at some time you may want to terminate a wiki and archive it, this option will just do that.
  23. #   – archive runs first the backup function to do a last minute backup;
  24. #   – archive cleans-up the wiki directory;
  25. #   – archive drops the database and the database user;
  26.  
  27. # 3) list: gives you a listing of running wiki's created with this script.
  28.  
  29. # Usage:
  30. # /script/path/wiki-admin create               # create wiki
  31. # /script/path/wiki-admin archive              # archive wiki and clean up
  32. # /script/path/wiki-admin list [wiki-name|*]              # list info about a wiki or without a name shows all info about all wiki's
  33.  
  34. # Requirements
  35. # – makepasswd                                            # for generating strong passwords
  36. # – mysql, mysqldump                                      # part of mysql-client package
  37. # – mysql server                                          # localhost or somewhere else you need ofcourse a mysql-server

Apache in combination with CNAMES pointing to VirtualHosts is no problem and very easy to implement.
For instance, say you have the following cnames :
wiki-1.somesite.org
wiki-2.somesite.org

  • First step, create the wiki’s, just follow the script.
  • Next, create the VirtualHost file in /etc/apache2/sites-available/ for wiki-1 and wiki-2 (see example below)
    notice that I redirect tcp/80 directly to tcp/443. If you don’t have a certificate then remove the first VirtualHost entry and change *:443 to *:80
  1. <VirtualHost *:80>
  2.         ServerName wiki-1.somesite.org
  3.         ServerAdmin admin@somesite.org
  4.  
  5.         Redirect permanent / https://wiki-1.somesite.org/
  6.          
  7. </VirtualHost>
  8.  
  9. <VirtualHost *:443>
  10.         ServerName wiki-1.somesite.org
  11.         ServerAdmin admin@wiki-1.somesite.org
  12.  
  13.         SSLEngine On
  14.         SSLCertificateFile /etc/apache2/ssl/law.pem
  15.         SSLCertificateChainFile /etc/apache2/ssl/sureserverEDU.pem
  16.  
  17.         DocumentRoot /var/www/wiki-1
  18.         <Directory />
  19.                 Options FollowSymLinks
  20.                 AllowOverride None
  21.         </Directory>
  22.         <Directory /var/www/wiki-1/>
  23.                 Options Indexes FollowSymLinks MultiViews
  24.                 AllowOverride AuthConfig
  25.                 Order allow,deny
  26.                 allow from all
  27.                 # This directive allows us to have apache2's default start page
  28.                 # in /apache2-default/, but still have / go to the right place
  29.                 # RedirectMatch ^/$ /apache2-default/
  30.         </Directory>
  31.  
  32.         ErrorLog /var/log/apache2/wiki-1-error.log
  33.  
  34.         # Possible values include: debug, info, notice, warn, error, crit,
  35.         # alert, emerg.
  36.         LogLevel warn
  37.  
  38.         CustomLog /var/log/apache2/wiki-1-acces.log combined
  39.         ServerSignature Off
  40.  
  41. </VirtualHost>
  • Enable site with: a2ensite wiki-1 && a2ensite wiki-2
  • /etc/init.d/apache2 reload
  • Now go to /var/www/wiki-1 and change in LocalSettings.php the following parameter:
  1. original : $wgScriptPath       = "/wiki-1";
  2. change: $wgScriptPath       = "";

This needs to be done because your VirtualHost defines your wiki location.

Download mediawiki-setup-0.4.tgz Version 0.4

After download verify with: sha1sum mediawiki-setup-0.4.tgz

f2678d81aceb9d72e58d7e51c1460750376a3447 mediawiki-setup-0.4.tgz (2008/04/29)

Untar it in /root for example with: tar xvzf mediawiki-setup-0.4.tgz

If you like it, drop me a note, or if you have any questions…. drop them to.

Cheers,

Raymond.

Post to Twitter

Ubuntu on Dell XPS M1330

Latest update: 4 aug 2008

For recapping my findings, I wrote up a small article on how to install Linux Ubuntu 8.04 LTS Desktop Edition (32bit) on a Dell XPS M1330.

Goals:
I switch from 64bit back to 32bit for compatibility reasons. There are to many apps that don’t work and are required to fit my needs. So I set up a boot partition (350MB – unencrypted) and a encrypted partition (100GB). Then I used that partition to set up LVM2 and prepared a root (10 GB), home (86 GB) and a swap partition (4 GB).

Download @ www.ubuntu.com and make sure to check “alternate cd” checkbox. It gives you the opportunity to install lvm2 on a encrypted partition.

Intro: Normally I run OSX os on Macbook Pro’s. However, since these are very pricey, and I smashed mine to smithereens,  I’ve decided to buy a cheap Dell.

  • Hardware
  • Partitioning harddisk (Dual boot: Ubuntu & OSX)
  • Installing OSX
  • Installing Ubuntu


Hardware :
Fast procs and 4 GB of Memory…

Partitioning harddisk:
describe….

Installing OSX:
I uses iAtkos_v4. This is version 10.5.2 and it installed without any problems. However I did change the BIOS setting from AHCI to ATA. Somehow after installing OSX it could not start giving me an error that is could not find its root device. Changing this after installing it booted just fine. Currently I don’t have wifi working ( Network controller: Intel Corporation PRO/Wireless 4965 AG or AGN Network Connection (rev 61). I probably by an other truemobile pci card.

Installing Ubuntu:
Perform regular installation.

  • Grub multiboot (Ubuntu Heron & OSX 10.5.2 Leopard)
    sudo vim /boot/grub/menu.lst
    add `#`  before hiddenmenu

    add menu option …
    title                Leopard OSX 10.5.2
    rootnoverify   (hd0,0)
    makeactive
    chainloader +1

  • thinkfinger
    sudo apt-get install build-essential libtool libusb-dev libpam0g-dev pkg-config
    sudo apt-get install pam-thinkfinger thinkfinger-tools
    sudo tf-tool --acquire
    sudo tf-tool --verify
    sudo cp ~/.thinkfinger.bir /etc/pam_thinkfinger/your_login.bir
    sudo vim /etc/pam.d/common-auth (change it like following)
    auth	sufficient	pam_thinkfinger.so
    auth	requisite	pam_unix.so try_first_pass nullok_secure
    auth	optional	pam_smbpass.so migrate missingok
    
    echo uinput >> /etc/modules (this will make sure that the modules loads when you reboot)
    sudo modprobe uinput (to load it for now)
    Restart gnome Ctrl + Alt + Backspace
    
    For more detailed information : https://wiki.ubuntu.com/ThinkFinger
  • Mucking-up desktop

    http://openswitch.org/2008/04/28/how-to-enable-emerald-in-hardy-heron-804/

  • sudo apt-get install awn-manager-trunk awn-extras-applets-trunk
    Here is my screenshot

Post to Twitter

Howto install vpn on Asus eee pc series

For a colleague of mine at the University of Tilburg (Netherlands) I connected his new gadget to the internet. He wrote the following on his webblog

[communitysense.wordpress.com]

I have to work a lot on Tilburg University campus, and continuously need to access the Internet, which is one main reason I bought my ASUS EEE laptop. However, to get access on campus one needs a VPN connection. On Windows machines, it is no problem to install the Cisco VPN client, but, alas, not so for Linux machines like mine. Raymond Mul, ICT manager at the Faculty of Law, and avid Linux enthusiast, has been so kind to write a short manual (in Dutch) explaining how to make the university network open up for Linux lovers as well.

Download asus-3e-uvt-vpn.pdf Version 0.1

If you find this howto useful, please leave a note. If the Dutch characters look scary to you, purchase me to write this howto in English. ;o)

Post to Twitter

Howto Aladdin eToken ssh key authentication

Start with installing the software for your mac. The current release is v4.55 [2008-03-27].When installed there are to programs “~/Applications/eToken/”. PKIMonitor, which will be available on your “menu bar”, and an application called eToken Properties.

If you followed the installation procedure then you are at the point for inserting the eToken. While looking at the eToken a red led should lid up. That’s a good thing!

First thing you need to do is “initialize” your eToken.Open up “Open eToken Properties”, click eToken , click “Advanced”, click “Initialize eToken”, Set User and Administrator passwords (both different and strong”, click “Advanced” and check the box “Load 2048-bit RSA key support”. Click Oke, Oke, Oke, I’ve stopped counting….

At this point your eToken is ready for use. However the software provided does not give you the opportunity to add a ssh private key.

Therefore, we need to install some additional software. So, go fetch OpenSC @ http://www.opensc-project.org/ . If you are like me, running “Leopard” on your mac, then you need the experimental build which you can get here : http://www.opensc-project.org/files/sca/experimental/ Currently I’m using version sca-0.2.3-pre2.dmg.

Follow the instructions. When your done, you can find that it lives in /Library/OpenSC

Change directory to /Library/OpenSC/bin

2008-04-15 – I’m sorry to announce that my macbook broke! I’ve taken it back to the mac store but I can not finish this article at the moment. To be continued!

2008-04-11 – Follow-up within a few days

Post to Twitter

Slamming vw type2

Finally at least! My vw type 2 from 1970 has been slammed. Friday, 21 December, I had an arrangement with Erwin from http://www.volksprojects.com. I left Hilvarenbeek, a small village nearby Tilburg, @ 7:00 and arrived at 8:30u.

Since it was pretty cold that day and my butt was freezing off, I kindly grabbed the hot offered coffee by Erwin. After having the basic needs we started on “the” project. With Erwin as guru, everything went pretty smoothly. The new front-beam was ready to build underneath, prepared with raster plates nicely weld on the beam.

Almost with the same ease as the original beam, it was replaced underneath. Then we build it up and the front was done. BTW when I mention “we” most of the time I’ll refer to Erwin ;-) The adapters, converting 5×112 to 5×130, where placed and also the 6j Fuchs 16″ wheels. In front, the tyre size is 185/50/16. This setup gives me about a small inch between the tire and shield.

Then we started on the back-end of the bus. First we dropped it by 2 levels. Immediately giving installation problems to put on the 16″ wheels. With a hinge jack we were able to assemble the wheels. When we finished the job there was just one conclusion to make. It looked ass-kicking! wh00t! BUT the suspension plates did not have enough free space. It left us no other choice then bring it back up by one level. Now the 7j wheels fitted nicely without any problems. Choice of tyres in the back, 205/50/16. Also this setup works great. I have about 1/4 inch between tyre and carrosserie!

At the end we had a few little problems like euh…. failing gas cable.. fixed, thanks to Erwin. And then, Moment Supreme, yep battery dead. GGRRRRRR. With a little help from starting cables and a “working” car, the t2a came alive.
When I left Erwin’s property to drive home, it was 22:00 o clock. About one hour later, frozen again, I reached Hilvarenbeek.

If you want more info, drop me an e-mail.

Post to Twitter

 

Twitter links powered by Tweet This v1.6.1, a WordPress plugin for Twitter.