rejecting counters: a regexp question

From: Stefan Monnier <monnier+lists/squid/news/@dont-contact.us>
Date: 16 Dec 1997 20:25:04 -0500

So I tried to setup a counter-rejection regexp.
It seems to work fine, but for some reasons, some requests still go through.
Here is the part of my squid.conf:

acl counter1 urlpath_regex [Cc]ount(er|)(|\.(cgi|gif|exe))(|/.*)$
acl counter2 url_regex ^http://(www\.|)(i|)count[a-z0-9]*(\.[a-z0-9]*|)\.com/
acl counter2 url_regex ^http://[a-z0-9]*\.starcounter\.com/
http_access deny counter1
http_access deny counter2

I have no "http_access" directive earlier in my squid.conf (even though I do
have some later, fo course) and there's no other "acl counter1" or "acl
counter2". And yet, the log shows that http://c2.xxxcounter.com/counter.cgi
has gotten through (with a TCP_HIT/302). Strangely enough, if I tried to access
this URL from my browser, squid correctly rejects the request.
What could cause this behavior ? It doesn't seem to be related to TCP_HIT
since I also got some TCP_MISS and other variants for pages that are supposed
to be denied.

I'm running squid-1.1.18 (with some patches to cache cookies as well as
temporary redirects) on Linux-2.0.32 with glibc.

        Stefan
Received on Tue Dec 16 1997 - 17:28:56 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:37:58 MST