Skip site navigation (1)Skip section navigation (2)
Date:      30 Apr 2001 17:38:21 -0400
From:      Chris Shenton <chris@shenton.org>
To:        "Kulraj Gurm (bosa.ca account)" <kulraj@bosa.ca>
Cc:        "Ken Bolingbroke" <hacker@bolingbroke.com>, "Ted Mittelstaedt" <tedm@toybox.placo.com>, <questions@FreeBSD.ORG>
Subject:   Re: Redundant Internet connections
Message-ID:  <87y9si9igi.fsf@thanatos.shenton.org>
In-Reply-To: "Kulraj Gurm's message of "Sun, 29 Apr 2001 08:06:02 -0700"
References:  <Pine.BSF.4.21.0104290057220.87921-100000@fremont.bolingbroke.com> <002901c0d0bd$e8466de0$64c8a8c0@asknet.com>

next in thread | previous in thread | raw e-mail | index | archive | help
"Kulraj Gurm (bosa.ca account)" <kulraj@bosa.ca> writes:

> I am a little confused however; maybe just my lack of understanding, but all
> your discussion on redundancy has been focussed on out bound traffic - where
> I can conceptually see it working. But what are the DNS implications?

It's not DNS, per se. Rather: how do packetss coming to your network
know which path to take?

If you have connections to ISP-1 and ISP-2, and ISP-1 dies, how does
traffic on the net know how to reach you via ISP-2, its network, and
its network provider?

If your host has IP 1.2.3.4 and that IP address space belongs to
ISP-1, I don't see how a connection to ISP-2 is going to help when
your connection to ISP-1 dies.



Now perhaps you could have a box with an IP for ISP-1 and one for
ISP-2, and NAT your internal network so neither ISP sees your real LAN
addresses. Then have DNS list both and when ISP-1 dies, change the IPs
in your DNS and quickly push them out... This seems kind of a hack
though.


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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?87y9si9igi.fsf>