Linux Netfilter discussions
 help / color / mirror / Atom feed
From: Pablo Neira Ayuso <pablo@netfilter.org>
To: seba@econ.uba.ar
Cc: netfilter@vger.kernel.org
Subject: Re: VRF + Netfilter deployment - multitenancy filtering box
Date: Thu, 27 Oct 2016 21:43:55 +0200	[thread overview]
Message-ID: <20161027194355.GB1257@salvia> (raw)
In-Reply-To: <c847aa653f58c75da5717819ff0f861b.squirrel@minkevie.econ.uba.ar>

On Mon, Oct 10, 2016 at 11:45:05AM -0300, seba@econ.uba.ar wrote:
> Hi, I'm trying to deploy VRFs + Netfilter in a such a way I could maintain
> different filtering rules while attaining isolated traffic at Layer 3, and
> perhaps some overlapping IPs functionality.
> 
> Even though there's some VRF + NF Kernel documentation, I could use some
> help from the netfilter community; and I think lack of information
> regarding this topic is due to only recent kernels (>4.6 will do the
> sucessful magic!) aren't massive today, but it'll be a hot issue when this
> kernel reaches production level.
> 
> Here's what I'm looking after:
> 
>                            +
>                         ^  |
>                         |  |
>                         |  v
>                         |
>                      +--+------+
>                      |         | 172.17.0.0/16
>                      | eth0    |
>                      | 	       |
>                      +---^-----+
>        +---------------^ |  ^-----------+
>        |                 |              |
>        |                 |              |
>        |                 |              |
>        |                 |              |
>    +---v------+     +----v-----+     +--v--------+
>    |          |     |          |     |           |
>    | RED      |     | BLUE     |     | GREEN	 |
>    | eth1     |     | eth2     |     | eth3      |
>    |          |     |          |     |           |
>    +----------+     +----------+     +-----------+
> 10.0.0.0/24         10.0.0.0/24      10.0.0.0/24
> 
> 
> Suppose you have an shared & physical interface which leads traffic inside
> and outside our box, serving three customers RED, GREEN & BLUE through
> dedicated physical internal interfaces, at the same time we're mantaining
> their traffic isolated from each other at layer 3 using VRFs (and maybe,
> later permit Overlapping IPs between customers if possible).
> 
> I'm considering creating three VRF devices and attaching each different
> customer interfaces to them at the same time I attach eth0 to all the VRF
> devices, then deploy NAT rules through iptables in order to get the
> traffic flow in the desired direction...
> 
> How would it be your filtering strategy in order to achieve some sort of
> "VRF/multitenant" filtering ability? Filtering rules in each vrf device
> it's the only way or I'm missing something?

Did you consider using conntrack zones for this?

See iptables-extensions(8) manpages, more specifically the CT target.
You can set the zone from raw/prerouting chain.

  reply	other threads:[~2016-10-27 19:43 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-10 14:45 VRF + Netfilter deployment - multitenancy filtering box seba
2016-10-27 19:43 ` Pablo Neira Ayuso [this message]
2016-10-28  1:05   ` seba

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=20161027194355.GB1257@salvia \
    --to=pablo@netfilter.org \
    --cc=netfilter@vger.kernel.org \
    --cc=seba@econ.uba.ar \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox