The Labs.Com Web Lab WebTracker

The Labs - Design & Functionality For The Net

Yet Another Web Log Analysis Tool

1. Introduction
2. Download
3. Usage

WebTracker Monthly Overview Screenshot

I have been using various web analysis tools, either they are too simple and miss features I'm looking for (e.g. webalizer, visitors), or they are overly complex and a pain to get to work (e.g. awstats under linux/debian-3.1 it fails even to install).

So, I quickly started my own web tool, called WebTracker. Its features:

• one perl script
• creates HTML out of the box: yearly overview and monthly details
• incremental runs supported
• no external graphics via <img>, all bars drawn as <table>
• statistic covered:
  • day, hour, weekday
  • referer
  • ip
  • errors, not_found
  • agent
  • all, pages, script, misc (no script, no pages)
  • country
  • domain

WebTracker 2. Download

WebTracker 0.01k (latest version), License: GPLv2

Following requirements:

• perl
• dig
• whois

WebTracker 3. Usage

webtracker [-o path] hostname [log-file(s)]

examples:

• webtracker the-labs.com logs/the-labs.com-access_log
• cat logs/the-labs.com-access_log | webtracker -o stats/ the-labs.com

it creates three files:

• wt.cache: history information, human readable & editable
• whois.cache: ip (x.y.0.0) to country mapping
• domain.cache: ip (x.y.0.0) to domain mapping

Note: the log-file contains IP without HostnameLookup enabled in httpd.conf - this saves DNS lookup of apache-http.

wt.cache

This file contains all information which are kept between runs, e.g. you can delete or truncate the access_log when you ran webtracker. The file is also human readable and editable.

In case you run multiple web-sites, which most of us do, you may sym-link whois.cache and domain.cache among them, but keep wt.cache unique for each domain.

                                                                                                                                   

WebSherpa

Web Lab

JavaScript