From: Pablo Neira Ayuso <pablo@netfilter.org>
To: Jeff Cook <jeff@deserettechnology.com>
Cc: netfilter@vger.kernel.org, kaber@trash.net
Subject: Re: Packets marked by iptables only sent to the correct routing table sometimes
Date: Wed, 31 Oct 2012 01:08:35 +0100 [thread overview]
Message-ID: <20121031000835.GA9300@1984> (raw)
In-Reply-To: <509061DC.3090606@deserettechnology.com>
On Tue, Oct 30, 2012 at 05:25:16PM -0600, Jeff Cook wrote:
> On 10/30/2012 01:16 PM, Pablo Neira Ayuso wrote:
> > On Tue, Oct 30, 2012 at 08:10:34PM +0100, Pablo Neira Ayuso wrote:
> >> On Tue, Oct 30, 2012 at 11:21:01AM -0600, Jeff Cook wrote:
> >>> Hello.
> >>>
> >>> I am trying to route packets generated by a specific user out over a
> >>> VPN. I have this configuration:
> >>>
> >>> $ sudo iptables -S -t nat
> >>> -P PREROUTING ACCEPT
> >>> -P OUTPUT ACCEPT
> >>> -P POSTROUTING ACCEPT
> >>> -A POSTROUTING -o tun0 -j MASQUERADE
> >>>
> >>> $ sudo iptables -S -t mangle
> >>> -P PREROUTING ACCEPT
> >>> -P INPUT ACCEPT
> >>> -P FORWARD ACCEPT
> >>> -P OUTPUT ACCEPT
> >>> -P POSTROUTING ACCEPT
> >>> -A OUTPUT -m owner --uid-owner guy -j MARK --set-xmark 0xb/0xffffffff
> >>>
> >>> $ sudo ip rule show
> >>> 0: from all lookup local
> >>> 32765: from all fwmark 0xb lookup 11
> >>> 32766: from all lookup main
> >>> 32767: from all lookup default
> >>>
> >>> $ sudo ip route show table 11
> >>> 10.8.0.5 dev tun0 proto kernel scope link src 10.8.0.6
> >>> 10.8.0.6 dev tun0 scope link
> >>> 10.8.0.1 via 10.8.0.5 dev tun0
> >>> 0.0.0.0/1 via 10.8.0.5 dev tun0
> >> ^^^^^^^^^
> >>
> >> 23.1.17.194, this doesn't go through tun0
> >
> > Sorry, I meant: 23.1.17.194, this goes through tun0.
> >
> >> 209.68.27.16, this doesn't go through tun0
> >>
> >> Address & CIDR => 209.68.27.16 & 128.0.0.0 => 128.0.0.0
> >>
> >> Then: 128.0.0.0 != 0.0.0.0, then go to default route, likely to be
> >> eth0.
>
> Thanks very much, I can verify that adding a route for 128.0.0.0/1 to
> table 11 fixes things.
>
> Apologies for asking a naive question, but could you please inform me
> where 128.0.0.0/1 comes from and why it's ANDed against external IP
> addresses? I've tried to find info on Google about 128.0.0.0 and CIDR
> and unfortunately have not been able to find anything thus far that
> enlightens me as to why that route is necessary. I'd really like to
> understand, so if you spend some time explaining it to me I'd appreciate it.
Your mask is wrong. Using CIDR notation 0.0.0.0/1 matches networks
from 0.0.0.0 to 127.255.255.255.
I'd suggest to add some default route to that table to get everything
through tun0 instead of adding 128.0.0.0/1
Regards.
prev parent reply other threads:[~2012-10-31 0:08 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-30 17:21 Packets marked by iptables only sent to the correct routing table sometimes Jeff Cook
2012-10-30 19:10 ` Pablo Neira Ayuso
2012-10-30 19:16 ` Pablo Neira Ayuso
2012-10-30 23:25 ` Jeff Cook
2012-10-30 23:45 ` Ed W
2012-10-31 0:08 ` Pablo Neira Ayuso [this message]
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=20121031000835.GA9300@1984 \
--to=pablo@netfilter.org \
--cc=jeff@deserettechnology.com \
--cc=kaber@trash.net \
--cc=netfilter@vger.kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).