Re: Suggestions wanted for Squid Log "data mining" script

On Mon, 6 Dec 1999, Miguel A.L. Paraz wrote:

]I'm planning to write an Open Source "data mining" script for Squid logs.
]This will go through access.log (and maybe even store.log), parse them and
]store the relevant info in a SQL database. To create reports we go through
]this SQL data to build "custom views" of the data. Another possibility is to
]let it easily accumulate data from multiple logfiles (different dates or
]from different caches).
]
]Any suggestions?
]I plan to make it very extensible (in other words: I won't write it in a hurry)
]so that other folks can customize it easily.

Yes, something akin to it is already being done, see:

        http://www.cache.dfn.de/DFN-Cache/Development/Seafood/

Only the graphical web interface is missing. The project will terminate
this month and the final program and documentation are going to be put on
the above mentioned place. <smug>The parser is quite speedy, I
believe.</smug>

The project's proposal refers to the action of "correlating data" in order
to show trends not immediately obvious from the basic tables. Even though
the project will not achieve all goals of data mining, it is believed that
seafood provides a managable multitude of basic data as a foundation.

Le deagh dhùrachd,
Dipl.-Ing. Jens-S. Vöckler (voeckler@rvs.uni-hannover.de)
Institute for Computer Networks and Distributed Systems
University of Hanover, Germany; +49 511 762 4726
Received on Mon Dec 06 1999 - 02:54:20 MST