everyday

In my work as a Network Administrator for a school district, I run into problems with email click-through marketing techniques rather frequently. The specific problem is with click-through tracking of email advertising campaigns. Since we are a school district that provides internet access to minors, we tend to filter our internet traffic rather heavily. This filtering wreaks havoc with most click-through advertising emails.

For instance, an email is received inside of our network with a link to:

<a href="http://www.rs6.net/?kna76dv978y34qtib33t897jk1bt4hq3fr897&site=www.theactuallink.com">www.theactuallink.com</a>

The user thinks that when they click on theactuallink.com that they will be taken to theactuallink.com, they are instead taken to rs6.net and then bounced through to theactuallink.com. This causes a great deal of frustration with our users. We constantly receive phone calls that our internet filter is blocking theactuallink.com, when it is in fact blocking rs6.net, and rightfully so.

I imagine that this problem will only increase as more and more companies increase the amount of internet filtering they do to comply with various laws and regulations. So, what is the solution?

I would highly recommend never, ever rely on a 3rd party to bounce your links through. If you need to track the effectiveness of an email advertising campaign, then invest the time and resources to roll your own tracking. That or use a tool like Google Analytics that will not break the link even if the tracking itself gets blocked. I know that a large number of the emails are generated by non-profit organizations or businesses with limited budgets that are running their email advertising campaigns through a 3rd party, but I think the time has come to abandon the 3rd party and either hire the talent or develop it from within to avoid losing your audience entirely. There are a large number of free, open source applications that will allow you to run email advertising campaigns so the cost for software would be negligible. Hardware costs and requirements are ever decreasing as well, so the excuses not to are few.

I know that the title probably makes no sense to any one of the 2 people who read my weblog on a semi-annual basis, but I have decided to try to post more stuff related to my work as a Network Engineer. That being said…

Today I ran into a problem with our Microsoft WSUS server. Recently I upgraded the WSUS from 3.0 to 3.0 Service Pack 1 (SP1). This is the second time our WSUS has been broken by an update. Usually Microsoft’s updates are pretty solid, but WSUS updates seem to be a tad iffy. The upgrade from WSUS 2.0 to WSUS 3.0 broke completely and I had to completely rebuild the server, luckily I run the WSUS on a virtual server using VMWare’s ESX server, so rebuilding it was very simple.

The particulars of the problem are that the client machines would not download updates! Normally I don’t check on the update status of machines very often, but I was creating a new image for a laptop and the thing just would not download updates. I check the WSUS log on the client machine and noticed that there was a 0×80190194 error. A quick search on Google pointed me to a number of forum posts that state this is an HTTP 404 (Page not found) error. Unfortunately, none of these posts had any information about a solution. A bit more searching led me to a post on Microsoft’s TechNet Discussion Groups that had the solution, Re: WSUS 3 sp1 do not distribute patches to clients in Windows Server Update Services (URL is rather deep and may be broken, no warranty). The noteworthy part of the post:

1. Create a new virtual directory called ’selfupdate’
2. Set the pathname to
   %programfiles%\Update Services\selfupdate

Hopefully Microsoft will put some more resources into testing the updates for WSUS, it is a really nice product, but the updates always seem to break it!

via: Simple Bits

Some exciting news from the developers of Microsoft’s Internet Explorer 8! Basically they have reversed course on their previous decision to hamstring standards based web design by forcing IE8 to render all web pages in an identical fashion as IE7, unless explicitly told to do otherwise. Microsoft was planning on forcing web developers tell IE8 to render a web page according to web standards or else IE8 would render the page identically to IE7 (which has some shortcomings when it comes to standards).

This is definitely good news and I for one am glad to see Microsoft making a decision that makes sense for the web community at large.

I have seen a few of the comics from xkcd and have found them to be rather funny (although I don’t know what that says about me…). Anyways, check it out!

My digital library has quickly exceeded the storage capacity of my MacBook Pro. What does this mean, well I either have to get rid of files or move them off of my laptop. Since I have no desire to delete anything, I move them to a external USB drive. I received a nice little 160GB Western Digital Passport this past Christmas. While this works really well, I am loathe to store all of my stuff on one hard drive. I prefer to have data on a server that utilizes some sort of HDD redundancy to safeguard my data against HDD failure. To this end I am thinking about either a small desktop computer running Linux or perhaps a used Mac Mini. I will keep you informed of my eventual decision.

After much frustration, I think I finally have a working installation recipe for RubyOnRails on Debian Linux.

1. Install Debian
2. Install ssh and sudo

apt-get install ssh sudo

3. Remove “root” login from ssh config file
4. Create a “source” directory

cd /
mkdir source

5. Install tools for compiling from various sources

apt-get install gcc gcc-3.4-doc gcc-3.4 g++-3.4 make libc6-dev flex openssl curl wget zlib1g-dev libzlib-ruby

6. Download latest Ruby (1.8.5 as of this entry), unpack and install

cd /source
wget http://ftp.ruby-lang.org/pub/ruby/1.8/ruby-1.8.5-p2.tar.gz
tar xzvf ruby-1.8.5-p2.tar.gz
cd ruby-1.8.5
./configure
make
make install
cd ..

7. Download, unpack, compile and install zlib (not included in testing version of Ruby?)

wget http://www.blue.sky.or.jp/atelier/ruby/ruby-zlib-0.6.0.tar.gz
tar xzvf ruby-zlib-0.6.0.tar.gz
cd ruby-zlib-0.6.0
ruby extconf.rb
make
make install
cd ..

8. Download, unpack, compile and install Ruby Gems (0.9.2 as of this entry)

cd /source/
wget http://rubyforge.org/frs/download.php/17190/rubygems-0.9.2.tgz
tar xzvf rubygems-0.9.2.tgz
cd rubygems-0.9.2
ruby setup.rb

9. Install Rails

gem install rails --include-dependencies

10. Install Mongrel

gem install mongrel mongrel_cluster --include-dependencies

11. Create a dummy application

cd ~/
rails test
mongrel_rails start

12. Go to http://your.ip.address:3000/ to see if it works

Sources

• The Perfect Rails/debian/lighttpd Stack…
• Mongrel: Debian HOWTO
• RubyOnRails
• Setting up a Complete Ruby Server with Mongrel and Nginx on Debian 3.1 Part 3