Development Cycle

Drupal is a great CMS for creating websites, easy to use, extendible, great community and very powerful.
In this quick post I am going to show the reader the reader how to add a contact form to their Drupal powered
site.

First up log into your Drupal install as the “admin” user browse to Administer / Site Building / Modules in the site
back end and activate the contact module. Then be sure to click the button down the bottom of the page to save your
changes.

Now look in Administer / Site Building you will find a section named “contact form” enter this section and you can get down to the business of actually building the form.
Click on the “add category” link and add a name, I am giving mine the name “contact us”. In the next text field enter the email address that messages received through forms in this category should go to and an auto reply message for the sender if you want one. I also changed the value in the “selected” option to yes as will need one category of contact forms.

Some other options you may want to change for your form can be found in the “settings” section, such as “additional information” to display above your contact form such as instructions to the user, alternative contact details etc. Another handy option is the ability to set the maximum number of submissions a user can make through the contact form per hour to stop people spamming you.

Okay the form should now be ready to go just need to create some menu links so visitors to your site can find it. To achieve this browse to Administer / Site Building / Menus and select the “add menu item” option to create a link to your contact page. Give the link a name, maybe a description, and link to the location “/contact” click the submit button and you should now see your newly created link.
Send your self a quick test email to check its working and your done you should have a working contact form for visitors to leave feedback in under 10 minutes.

I have a eeepc 701 netbook which i use for reading ebooks, surfing the net and coding on public transport on my way to and from work. It is a great system for beating the boredom of commuting, but lately i have been having trouble writing to the 8gb sdcard i have in it for storage.

The problems seem to have come about due to my choice of file system. When i first got the card i formatted it using ext3. Logic being its pretty much the default filesystem used by Linux and i had never had any problems with it on disks before.

Seems this logic unfortunately dosn’t translate over to the world of sdcards sometimes when writing to a file, or most of the time i tried to delete data the journal seemed to chuck a wobbly and remounted the sdcard as read only.

Everything i tried to fix it, came undone and put me back at square one when i tried to delete another file. After spending some time looking around the net i now see  that although generally journaled file systems are considered superior to non-journaled filesystems, they are not generally recommended for small flash based media.

Which brings me to the process of reformatting the card to in case anyone else has made the same mistake. Its a pretty simple process but i thought it worth recording:

df -h

Note what device the sdcard is, mine was /dev/sdb1.

fdisk /dev/sdb1

With fdisk i used the “n” command first to create a new partition that occupied the whole sdcard and then the “w” option to write the table to disk.

mkfs.ext2 /dev/sdb1

Then formatted the drive with my new ext2 filesystem, hopefully saving me from any further issues. Only thing left to do was modify my /etc/fstab so the card would be mounted automatically on boot, i already had the line:

UUID=7922e6f5-eec6-4ad3-b87c-3202651318a1 /home/john/mount ext3  auto,exec,rw,async,user 0 0

But this line was going to have to change, not only had the filesystem type changed from ext3 but after being reformatted the filesystem would have a new UUID. So first i used the command:

tune2fs -l /dev/sdb1 | grep UUID

Which had the output:

Filesystem UUID:          32938a3e-9bf9-4914-b348-659492b4fea7

Armed with this information i modified the line in fstab to read:

UUID=32938a3e-9bf9-4914-b348-659492b4fea7 /home/john/mount ext2  auto,exec,rw,async,user 0 0

Rebooted, and it seems i have a writable sdcard once more.

Today i logged into a Ubuntu server which is not visited very
often to change the server where it stores backups (transferred over SSH) of the websites it hosts.

I first transferred a copy of its public key to the new backup server and used cat command to echo the key into the authorized_keys file in the .ssh directory, then tried to relogin from the original server.

But time and time again i continued to be prompted for a password, thinking i was starting to lose it i transfered the public key over again but was still greeted by the password prompt at login time.

Getting desperate i had a look in the authentication
log (/var/log/auth.log) and saw this line:

Feb 11 7:55:31 nsw-ultimo1 sshd[32598]: Public key
6f:92:15:01:ae:17:9a:d7:b6:ff:b3:3f:4c:ac:20:dd blacklisted (see
ssh-vulnkey(1))

Seems the server had not had its SSH keys updated after the discovery
last year that the random number generator in Debian’s openSSL
package was not quite random enough.

The fix was painless enough once i had caught on to what was wrong. It was simply a case of deleting the existing keys updating the system and regenerating all of the keys from scratch using ssh-keygen.

In order to save future pain on any other neglected server i have discovered the people at Debian have created a tool named ssh-vulnkey.

So to check for vulnerable keys on other servers i simply need to log in and use the command sudo ssh-vulnkey -a and it will look for keys in all the usual places throughout the system and let me know if they are vulnerable and need replacing.

 This XKCD comic made me laugh today, Vista is the operating system everyone loves to hate.

Working on software projects is one thing but i am sure every developer has worked on at least one project in their working life that has been a nightmare project and out to be a complete disaster due  to poor management. Agile development methods attempt to make the development process easier through a predefined series of project management techniques to aid the rapid development of quality software.

Scrum is a popular approach to developing software in an agile way and has been applied to fields other than software development as iterative, incremental process for developing a product or managing work.

The video below is great to introduction for those not familiar the basic principles of scrum development in under 10 minutes. Some of the topics covered are:

• The terms involved with the scrum development methodology.
• Human roles involved a scrum based project such as the product owner, the scrum master etc.
• Basic methods which are used when scrum techniques are applied to a software development project.

The Linux kernel is one of those things i like most people tend to take for granted. I occasionally will see my system updating its kernel, read the new features in the latest Kernel release or occasionally head over the linux.org to grab the latest source if i need to compile my own for whatever reason. But thats normally as far as i look into it, until i saw this google tech talk by Greg Kroah Hartman he is a great speaker and presents a lot of interesting information about the Kernel.

Such as the Kernels rate of change i was suprised the Kernel itself consists of around 9 million lines of code, and in the period between 2007 – 2008 on average 4300 new lines of code are added to the Kernel code base everyday and that as time passes this rate of change is actually increasing dramatically.

The video is 49 minutes long and its presented very well, i think nearly anyone who has any interest in Linux would find it very informative to see who and what goes on in the core development of the operating system.

With the Christmas break many of us may be using the time away from our day jobs to work on our own pet projects. In order to assist my fellow man i have decided to post a list of some online utilities i find very useful to speed things up when building a site.

YAML

Sounds pretty boring but this is not the YAML which normally comes to mind. YAML in the case stands for Yet Another Multicolumn Layout, simply use the web based interface to design your layout and it will generate the code and CSS for you!

IconPedia

IconPedia is a nice one stop shop for free idividual icons and icon packs in PNG format.

Favicon Maker

This neat utility will create your site a favicon for your site from an uploded image, or you can create one from scratch using their web based editor.

WebScript Lab

A web based utility that allows you to generate attractive animated AJAX gif loading images. This site is also home to a few other design scripts including a favicon generator and a web 2.0 domain name generator.

FormStyle Generator

This handy online tool helps you to quickly design CSS style HTML forms that are easy on the eye.

Reflection Maker

Images with reflections are all the rage in the web 2.0 world, and this site allows you to easily add a reflection to your images so you won’t feel left out

The more that is said about the planned ISP level filtering, the worse it seems to sound. The family first nutjob Senator Steve Fielding has piped up, and said that they would like pornography and fetish material blocked by the planned filter.

This in turn seems to have inspired the South Australian independant Nick Xenophon who said should the plan go ahead he would like to block people from visiting overseas casino sites which are currently illegal to run in Australia.

Seems to back up the argument against internet censorship, just the concept of being in control of an as yet still imginary filter has them wandering about expressing their delusional ideals.

Senator Conroy and his department also seem to be getting desperate to limit debate on the matter. This article published by SMH claims that Senator Conroy’s policy advisor Belinda Dennett emailed Carolyn Dalton at the  IIA, Australia’s internet industry association asking them to try and influence internode management into pressuring Matthew Newton to shut up about the topic.

On the positive side though it seems the subject is slowly filtering out to the mainstream media outlets, which is going to make criticism of the plan a lot harder to brush under the carpet.

STOP PRESS: A new wiki has been launched to keep track of Australian senators stance on the proposed ISP filtering.

Have you ever been side tracked whilst logged into a remote server via SSH, only to find when you returned your session had timed out? A one line addition to the ssh_config can prevent this from being an issue.

Simply open a terminal window on your Ubuntu machine and type:

sudo pico /etc/ssh/ssh_config

And add the line:

ServerAliveInterval 60

Then save your changes, this change will send a small “keep-alive” packet to the server every 60 seconds, making it think that the connection is actively being used, so your connection will no longer time out due to inactivity.

Awstats i think has lost the lime light in the last few years thanks to Google Analytics, but that dosent mean its useless. Its quick and unlike Google Analytics the latest stats are not a few days old, well they arent if your cron jobs run on regular basis anyhow. This post aims to show you how to get awstats up and running on your Ubuntu based server, with the stats of a virtual host pulled straight from the Apache log files.

First login in to your server via SSH and become root:

sudo su

Update your package list:

apt-get update

Install the awstats program:

apt-get install awstats

Create a new configuration file for the virtual host you would like to gather statistics on:

cp /etc/awstats/awstats.conf /etc/awstats/awstats.mydomain.tld.conf

You will need to create a separate configuration file for each virtual host you would like to gather statistics about. Of course when the virtual host is setup in Apache it will need its own log files.

To see where our visitors hail from we will also need to install a PERL GeoIP library:

apt-get install libnet-ip-perl libgeo-ipfree-perl

Now its actually time to edit the awstats file for you virtual host so it contains the appropriate settings, lines you may need to look at are:

Line 51 - This needs to point to the access log of the virtual host

Line 122 - Make sure the log type reflects the format Apache is writing

Line 153 - Change this to the domain name of your site

Line 168 - Aliases etc that can be used to access the virtual host

Line 329 - To protect access to the sites statistics

Line 340 - “”    “”    “”    “”    “”

Line 350 - “”    “”    “”    “”    “”

Line 362 - Should Awstats create a directory for storing reports if it dosent already exist

Line 372 - Should Awstats use a text or XML file to build the stats database

Line 382- XHTML or HTML statistic reports

Line 413 - Should Awstats empty the access log after a parsing it can save having to worry about rolling it.

Line 428 - Tells Awstats if you would like to archive the logs which it has already parsed

Line 450 - The default name of the report file

Line 466 - Clients that should be skipped in the reports i.e your own IP address

Line 479 - Skip certain user agents i.e wget etc

Line 497 - Dont count certain files on your site

Line 732 - Turn on Awstats debugging for more verbose output

Line 1255 - If you want to enable extra toolips in your reports

Line 1271 - Reverse lookups for IPV6 IP address’s

Line 1302 - Use of a free GeoIP service for country to IP lookups

After the configuration is finished for our virtual host you will need to configure Apache so you can access your statistics.
Create a new file for Apache with:

pico /etc/apache2/conf.d/awstats

In this file put the content:

Alias /awstatsclasses “/usr/share/awstats/lib/”
Alias /awstats-icon/ “/usr/share/awstats/icon/”
Alias /awstatscss “/usr/share/doc/awstats/examples/css”
ScriptAlias /cgi-bin/ /usr/lib/cgi-bin/
ScriptAlias /awstats/ /usr/lib/cgi-bin/
Options ExecCGI -MultiViews +SymLinksIfOwnerMatch

Now restart the Apache service so you changes to the configuration take effect:

/etc/init.d/apache2 restart

Now add a line to your crontab to update your statistics every so often:

pico /etc/crontab

And add the following line to update your stats every 20 minutes:

*/20 * * * * root /usr/lib/cgi-bin/awstats.pl -config=mydomain.tld -update >/dev/null

Save the changes and just to be on the safe side try and update the stats drom the command line with:

/usr/lib/cgi-bin/awstats.pl -config=mydomain.tld -update

Note: If you get an error that looks like

Error: SiteDomain parameter not defined in your config/domain file. You must edit it for using this version of AWStats.
Setup (‘/etc/awstats/awstats.conf‘ file, web server or permissions) may be wrong.
Check config file, permissions and AWStats documentation (in ‘docs’ directory).

Just delete the /etc/awstats/awstats.conf file and it will read the one you are trying to specify instead.

Now to see the statistics for the site you just generated type the hostname of your server into a browser followed by /awstats/awstats.pl?config=mydomain.tld i.e

http://www.myserver.com/awstats/awstats.pl?config=mycooldomain.com

With any luck you should now have access to a webpage with some pretty detailed statistics on your domain.