From: "it clown" <suse@mailbox.co.za>
To: netfilter@lists.netfilter.org
Subject: Re: limiting bandwidth with iptables or squid?
Date: Fri, 29 Oct 2004 07:54:46 +0200 [thread overview]
Message-ID: <web-493669244@mail01.infosat.net> (raw)
In-Reply-To: <417CE54B.5060107@dsl.pipex.com>
Hi All,
I am abit confused with tc.
I need a string to limit band whidth for ips or mac
addresses.
iptables -A PREROUTING -i eth0 -s x.x.x.x -t mangle -j MARK
--set-mark 1
will mark the packets for that ip.i think.
what string for tc do i need to limit x.x.x.x to say 5kb/s
download speed? eth0 is my internal network card of my
linux box.
Regards
On Mon, 25 Oct 2004 12:36:43 +0100
Andy Furniss <andy.furniss@dsl.pipex.com> wrote:
> Jason Opperisano wrote:
> > On Sun, 2004-10-24 at 10:20, it clown wrote:
> >
> >>Hi All,
> >>
> >>I would like to know how to limit bandwidth with
> iptables.I
> >>would like to limit bandwidth to ip's and mac
> address.Do i
> >>need another program to work with iptables or can
> iptables
> >>do it on its own?Does any one know where i can find
> >>something to read up on this?
> >>
> >>If iptables can not do it can i do it with squid?
> >
> >
> > first off--you can't do "inbound" traffic shaping--only
> outbound.
>
> I know your link qualifies this a bit - but I'd say you
> can do inbound (narrow end of bottleneck) traffic
> shaping. Albeit as an inperfect kludge and at the expense
> of some bandwidth. The imperfect bit being if you really
> care about latency - not bandwidth shaping, I would argue
> that I can do that almost perfectly as my ISP has a 600ms
> buffer and my shaping at 80% of 512kbit/s never looses
> control enough that packets get dropped from that.
> you
> > cannot control how fast incoming packets hit your
> machine
>
> I assume TCP - which is clocked by acks - so the rate you
> dequeue does affect the rate at which packets hit ISP
> buffer. Packets never hit my machine faster than my link
> speed of course - which makes the queue get filled in a
> burstless way.
>
> > or how big they are--
>
> mss clamping :-)
>
> Andy.
>
> you can only control how fast you allow them to leave.
> >
> > that being said--read chapter 9 of:
> >
> > http://lartc.org/howto/
> >
> > -j
> >
>
>
>
_____________________________________________________________________
For super low premiums ,click here http://www.dialdirect.co.za/quote
next prev parent reply other threads:[~2004-10-29 5:54 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-10-24 14:20 limiting bandwidth with iptables or squid? it clown
2004-10-24 15:16 ` Jason Opperisano
2004-10-24 15:35 ` it clown
2004-10-25 11:36 ` Andy Furniss
2004-10-29 5:54 ` it clown [this message]
2004-10-31 11:40 ` Andy Furniss
2004-10-25 0:15 ` Alexander Samad
2004-10-25 9:05 ` Matteo Santori
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=web-493669244@mail01.infosat.net \
--to=suse@mailbox.co.za \
--cc=netfilter@lists.netfilter.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.