RE: [squid-users] Squid delay pools using DS/Precedence field to limit BW

From: David Wilson <dave@dont-contact.us>
Date: Fri, 23 Nov 2001 11:28:55 +0200

Hi Henrik,

Great ! Thanks for the help, I wonder how I got about getting the feature
coded into Linux-2.4.x ? ;-) ..or coding a patch for it ?

-----Original Message-----
From: hno@hem.passagen.se [mailto:hno@hem.passagen.se]On Behalf Of
Henrik Nordstrom
Sent: 23 November 2001 09:54
To: David Wilson
Cc: Squid-Users@Squid-Cache. Org
Subject: Re: [squid-users] Squid delay pools using DS/Precedence field
to limit BW

David Wilson wrote:

> So from what you are saying it's almost impossible for me to build ACL's
> that can discriminate between the value's of the DS/precedence bit field ?

Not impossible, but first the OS needs to be extended so there is a way
that Squid may get access to the information.

> I wonder if looking at the source of "tc" (Linux shaper) would shed some
> light on how to implement the same feature into Squid ?

Probably not. The traffic shaper runs in the kernel, and is packet
based, so it does have access to the per packet DS information quite
naturally.

> I also wonder how guys that produce hardware caches manage to get it
right,
> like Netcache, Allot etc etc. ?

Well.. they have control over the OS as well as the cache application.

> Anyone got a "Learn c in 1 week" book for me, sheez I wish I knew how this
> stuff worked so I could strip it to pieces and at least make an attempt at
> building the feature in ? :)

The coding required in Squid is quite small, once there is a mechanism
that Squid may use to get the DS information from the kernel on TCP
traffic..

Regards
Henrik
Received on Fri Nov 23 2001 - 02:29:47 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:04:27 MST