[Summary-Talk] Re: Trash in log reports

["Phil St. Romain" <phil@xxxxxxxxxxxxxxx>]

Thank you, Jason.  I don't know why my server records that as a 
redirect, considering that I don't even have CGI enabled.  I've set 
up a filter for 301 codes and that helped somewhat.  Still, there's 
lots of trash, and some of it comes from 200 codes, such as the 
following:

192.168.1.51 - - [17/May/2004:07:45:31 -0500] "GET 
http://www.bidppc.com/cgi-bin/smartsearch.cgi?username=zhouyan&keywords=%20
Health%20 
HTTP/1.0" 200 23551

I have no idea why this registers as a successful hit, seeing that I 
am not hosting bidppc.com and do not have CGI enabled.  If you click 
on the link, you are taken to an error page on the bidppc.com web 
site, then you are redirected to their home page.  This seems to be a 
kind of bot, but why it registers 200 I do not know.  I'm using 
Apache 1.3.26 on Mac OS X Server 1.5 as my web server and have only 
tweaked basic settings.  I'm aware that Summary will read pretty much 
everything in access.log as a successful hit, especially 200 codes.

Is there any way to filter "GET" requests out of my log reports? 
That would do the job, I'm sure.  Checking the option to filter 
robots out of my reports doesn't work; I get no files in the reports.

Phil



>Date: Tue, 8 Jun 2004 01:32:41 -0400
>From: Jason Linhart <jason@summary.net>
>Subject: Re: [Summary-Talk] Re. Trash in log reports
>
>The status code for this request is 301, which is a redirect. Summary
>counts a 301 request as a success, the user was successfully redirected
>to another page. Apparently your server is redirecting errors to a
>separate error page, instead of serving the error directly (which would
>be the correct thing to do). Depending on what server software you are
>running, it may be possible to reconfigure the server to do the right
>thing, or it might not.
>
>This server behavior is going to make it difficult to detect bad links,
>as all of the errors will be on the error page, and not on the request
>that actually had a problem.
>
>Jason
>
>
>On 6/6/04 5:19 PM Phil St. Romain (phil@shalomplace.com) wrote:
>
>>In the access logs, however, I find the following, which is obviously
>>recorded as a hit:
>>
>>192.168.1.51 - - [17/May/2004:08:42:20 -0500] "GET http://
>>  www.kanoodle.com/clickthrough.cool?position=7001 HTTP/1.0" 301 299
>>
>  >Why is this not a failed request, I wonder?  Anything I can do about it?


-- 
Be here now in love, and all shall be well.
      http://www.shalomplace.com
-------------
Go to <http://summary.net/list.html> to update subscription info.