Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 20 Apr 2002 20:54:21 -0600
From:      Brett Glass <brett@lariat.org>
To:        Terry Lambert <tlambert2@mindspring.com>
Cc:        "Matthew D. Fuller" <fullermd@over-yonder.net>, chat@FreeBSD.ORG
Subject:   Re: How to control address used by INADDR_ANY?
Message-ID:  <4.3.2.7.2.20020420204617.021f4470@nospam.lariat.org>
In-Reply-To: <3CC22126.9F28CE8A@mindspring.com>
References:  <4.3.2.7.2.20020420111258.021d7270@nospam.lariat.org> <4.3.2.7.2.20020419144005.0358c610@nospam.lariat.org> <4.3.2.7.2.20020419144005.0358c610@nospam.lariat.org> <4.3.2.7.2.20020420004621.02379880@nospam.lariat.org> <3CC1245C.EEE4ADE@mindspring.com> <4.3.2.7.2.20020420111258.021d7270@nospam.lariat.org> <4.3.2.7.2.20020420113621.021dfd00@nospam.lariat.org>

next in thread | previous in thread | raw e-mail | index | archive | help
At 08:17 PM 4/20/2002, Terry Lambert wrote:

>> Other options I've considered are:
>> 
>> 1) Using natd to change the souce addresses on outgoing packets
>> with a source addresses in 10.x to something routable (that is,
>> having the machine do NAT for its own internal processes). Would
>> this work?
>
>The NAT can't do block address translation, it can only do 1:N
>translation (not N:N translation).

Ah, but we only NEED to do 1:N translation. We need to translate
the source address of 10.X.Y.Z to A.B.C.1 when going outbound
on the upstream interface. I believe that ipnat is capable of
doing this with a "map" rule, because it sits outside the
kernel. But I don't know if natd (which is what I'd prefer to
use because it's able to do port-specific NAT ore gracefully)
can do this.

>> 2) Running local processes in a "jail" (assuming that this would
>> force their IP source addresses to the address assigned to the
>> "jail...." Would it? 
>
>No, it would not force the source address.

Are you sure? I haven't played much with jails, but I do note the
following on the jail(8) man page:

>     jail.socket_unixiproute_only
>          The jail functionality binds an IPv4 address to each jail, and lim-
>          its access to other network addresses in the IPv4 space that may be
>          available in the host environment.

I had always interpreted this to mean that the apps operating in
the jail were limited -- both when they listened and when they
opened outbound sockets -- to using the jail's IPv4 address. 

--Brett Glass


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4.3.2.7.2.20020420204617.021f4470>