RE: [squid-users] Squid buffering slow connections

From: Arjé Cahn <Arje@dont-contact.us>
Date: Tue, 22 Oct 2002 14:52:23 +0200

Marc,

Squid works perfectly in every way - but I'm stuck with this slow connection. You don't suppose there is another way of doing this?

Arje

-----Original Message-----
From: Marc Elsen [mailto:marc.elsen@imec.be]
Sent: 22 October 2002 09:42
To: Arjé Cahn
Cc: squid-users@squid-cache.org
Subject: Re: [squid-users] Squid buffering slow connections

Arjé Cahn wrote:
>
> Hi all,
>
> My webserver is behind a slow (DSL) connection that sometimes even fails to connect. I would like to use Squid as an accelerator located at a hosting farm.
>
> Can I get Squid to do this:
>
> - Request for foo.html comes in from client
> - Squid immediately sends cached foo.html

 Will happen if the object has been fetched once, and can
be considered fresh at the time a subsequent request for that
object is made through SQUID.

> - If needed, Squid refreshes the page in the background (but keeps sending the cached version to clients while doing this)

 Squid doesn't do this. Only upon a client request for that object.

> - If the connection between Squid and the webserver fails, it keeps trying for X times
> - When the refreshed page is in, Squid overwrites previously cached version

 Squid doesn't do this, because the whole process of fetching a possible
 new version of an object is client driven.

>
> The main idea is that Squid buffers the slow or failed request BUT KEEPS SENDING cached pages. I have the feeling that this is a standard Squid concept, but I can't find it in the manuals! What do I have to do?

 Your feeling is wrong, this that is not a standard Squid concept.
 If a cache becomes an active device in the context of what it's job
is,
 it may defeat it's purpose, that is reducing Internet traffic.

 In principle the process is client driven , making the cache 'silent'
 if no objects need to be fetched.

 M.

>
> Regards,
>
> Arje

-- 
 'Time is a consequence of Matter thus
 General Relativity is a direct consequence of QM
 (M.E. Mar 2002)
Received on Tue Oct 22 2002 - 06:57:48 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:10:46 MST