RE: Starting point for squid plug-in?

From: Robert Collins <robert.collins@dont-contact.us>
Date: Fri, 29 Jun 2001 15:55:49 +1000

> -----Original Message-----
> From: herbert farmer [mailto:hffarmer@hotmail.com]
> Sent: Friday, June 29, 2001 3:46 PM
> To: anil@recoil.org; squid-dev@squid-cache.org
> Subject: Re: Starting point for squid plug-in?
>
>
> "Anil Madhavapeddy" <anil@recoil.org> wrote:
> >herbert farmer wrote:
> > >
> > > Does anyone really expect an external ICAP server to
> > > perform well enough to handle tasks as resource intensive as
> > > content filtering??
> >
> >Read the spec ... that's exactly the point of iCAP - to vector
> >resource-intensive tasks to a separate server farm. If you
> have 10,000
> >concurrent requests and one of them has to be virus-scanned,
> why impact
> >the other 9,999 requests by hogging CPU doing this? Instead, pay the
> >cost of a TCP connection and have external farms do the hard
> work, and
> >keep your cache simple.
> >
>
> On the other hand, if you have 10,000 concurrent requests, and you
> need to check EVERY ONE of them to see if the request should be
> allowed or not, a TCP connection and ICAP request exchange seems just
> a bit excessive for the processing of a simple boolean allow/deny
> facility.

A simple boolean test doesn't really fit the "resource intensive" target
for iCAP.
 
> Does squid have any hooks for such things?? Where should I go
> looking??

Yes! The stock squid has ACL's - the code is quite modular, you simply
define a new enum, and then code in acl.c to parse and process your acl.
I have written a fully modularised variant of this, allowing extra ACL
type's to be added with no source change to squid, but that's not in
HEAD. (See the generic.modules sourceforge branch if you want to fiddle
- no HTTP or caching changes are made in that branch, just code
arrangement).
 
Rob

> Thanks in advance,
>
> herbert farmer
> hffarmer@hotmail.com
>
> _________________________________________________________________
> Get your FREE download of MSN Explorer at http://explorer.msn.com
>
>
Received on Fri Jun 29 2001 - 00:05:53 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:14:04 MST