Re: per user throttling

The short answer is that this isn't for the faint of heart.  It can be done, 
but you'll need to do a good bit of coding and have a good understanding of 
iptables, packet marking, NAT, and the order in which all these processes 
occur.

I managed to get this working, but it was basically a complete rewrite of 
the throttling part, and also involved changing almost every reference to 
packet marking in iptables.  Here's an old message I sent explaining how I 
accomplished this.  It should at least provide direction on one working path 
to accomplish this:

http://freenetworks.org/pipermail/pebble-linux/2004-October/001019.html

----- Original Message ----- 
From: <[email protected]>
To: <[email protected]>
Sent: Tuesday, April 19, 2005 11:33 AM
Subject: [NoCat] per user throttling


> Hi,
>
> in the archive of the mailing list I found two approaches dealing with per
> user throttling. One the "throttle-per-public-user-04.patch" limiting 
> based
> upon iptable filters for the public user class and the posts from Derk van
> de Velde and Sameer Verma
> "http://lists.nocat.net/pipermail/nocat/2004-March/004629.html".
>
> So far I could not make either of them working correctly. When applying 
> the
> changes from Derk/Sameer redirection returns to the authentication page
> rather than the originally requested page...
>
> anybody has any recent experience with throttling per user and can walk me
> through this?
>
> do the available patches effect each other? all patches (found a nice
> summary at http://www.juneauwireless.net/NoCatAuth.html) claim to be run
> against 0.82. do I have to worry about the order of applying patches or
> comparing from hand rather than patch -px <...?
>
> thanks,
> regards
> jj
>
> _______________________________________________
> NoCat mailing list
> [email protected]
> http://lists.nocat.net/mailman/listinfo/nocat
>