There are customer use cases where sharing information via internet or uploading data somewhere is not acceptable for security reasons: this avoid the usage of some tool like the most famous Google Analytics, and prevent developers from understanding how the web application is used by the customers. I don't want to reinvent the wheel and re-implement a copy of Google Analytics, but getting inspired from it, the goal is to reuse information that we already have to extrapolate an analysis of the WebApp customer usage.

I started this project with the aim of learning a programming language where I am not so comfortable yet (python). The purpose of this Hack Week project is to bring this basic tool at a minimal stable and usable state with the purpose of analyze the usage of a WebApp in scenarios where the WebApp is used in an internal network only (offline, disconnected from the internet).

Starting from the current status of the tool at this commit, I'd like to improve it more:

  • fix the patterns finder [DONE]
    • data in UI are badly presented and grouped - [DONE]
    • the algorithm generates a pair of from-to URLs pattern ignoring they comes from a different ip/user, and the data results reflects a non-real pattern actually. This needs to be fixed. [DONE]
  • add filters for the patterns section [DONE]
  • let the table columns to be ordered
  • go through the python backend algorithms and improve [DONE]
  • provide a feature to compare and diff from a given list of URLs (a struts-config.xml for instance) which are the most used and which are never hit [DONE]

Long run roadmap:

  • let the engine keep the history of what has already been read and what not (by date and time? by log file?)
  • provide a simple optional javascript to send an AJAX request with some information (user, date and time, URL, etc) to a configured endpoint
    • this could replace the logic of reading and parsing tomcat logs
    • store this information in the database
    • run the python code against the database instead of tomcat log files

Looking for mad skills in:

log python tomcat analyzer analysis statistics web yarn reactjs

This project is part of:

Hack Week 17

Activity

  • about 1 year ago: dleidi started logorator: an offline internal analytics tool
  • about 1 year ago: LuNeves liked logorator: an offline internal analytics tool
  • about 1 year ago: dmaiocchi liked logorator: an offline internal analytics tool
  • about 1 year ago: dleidi added keyword "reactjs" to logorator: an offline internal analytics tool
  • about 1 year ago: dleidi added keyword "yarn" to logorator: an offline internal analytics tool
  • Show History

    Comments

    Be the first to comment!

    Similar Projects

    Make "salt-toaster" available to be used outside SUSE by PSuarezHernandez

    The salt-toaster (https://github.com/openSUSE...


    Porting Askbot to Python 3 by rbueker

    During the last year Askbot, a question and ans...


    Finish packaging Angr in OBS by a_faerber

    Following a FOSDEM presentation on [Angr](http:...


    Run and manage your Ansible cluster using Salt! by PSuarezHernandez

    At SUSE we've implemented a module on Salt call...


    Evaluate mirrormanager (or mirrormanager2) for download.opensuse.org mirror management by lrupp

    As there is no progress around [MIrrorPinky](ht...


    Finish packaging Angr in OBS by a_faerber

    Following a FOSDEM presentation on [Angr](http:...


    LibertyDraft.dev by hennevogel

    Discover Free Software projects that expose you...


    SVG in-depth by thutterer

    Inspired by a [conference talk](https://www.you...


    Suse Manager - SPA by LuNeves

    The experience while navigating throughout the ...