Re: [squid-users] Caching and tracking or 2-level cache rule?

From: Henrik Nordstrom <hno@dont-contact.us>
Date: Thu, 20 Dec 2001 16:27:26 +0100

What you describe cannot easily be acheived using Squid, but you may be able
to do something close to it by using a redirector which performs the tracking.

Also, using Session URL's generally totally screw up caching, as caching is
performed on the URL as a whole. There is no way Squid can know that the
"/SID/" component of the URL is not actually part of the URL. You can use a
redirector to remove the /SID/ component, but then it won't get forwarded to
the origin servers either except as part of Referer...

Regards
Henrik Nordström
MARA Systems AB

On Thursday 20 December 2001 14.13, Ulf Wendel wrote:
> Hi,
>
> I'm wondering if there's a way to use a reverse proxy without loosing
> full (session-based) user tracking performed by standard CGI (PHP)
> scripts. What I want to do is this:
>
> 1.) Request URL/SID/somepage.php
>
> 2.) Squid
>
> 2.1.) If URL/somepage.php is cached return cached page _and_
> 2.2.) request URL/tracker.php?sid=SID&page=somepage.php
>
> Every request to the webserver that matches a certain pattern is served
> by the reverse proxy (2.1) and causes the reverse proxy to request a
> certain page on a (dedicated) tracking server.
>
> I know I should be able to do this with the Apache. But I'm just curious
> if there's a way to do this with the squid as well? Can I do so using
> the config file? Can it be done using a hook into the squid (C-module)?
>
> Thanks!
>
> Ulf

-- 
MARA Systems AB, Giving you basic free Squid support
Customized solutions, packaged solutions and priority support
available on request
Received on Thu Dec 20 2001 - 08:27:05 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:05:26 MST