Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 30 Mar 2004 11:47:21 -0800
From:      "Rick Duvall" <rduvall@onlinehighways.net>
To:        "Marc G. Fournier" <scrappy@hub.org>, <freebsd-questions@freebsd.org>
Subject:   Re: BGP On Host
Message-ID:  <012001c4168f$d13592a0$f901a8c0@ws21>
References:  <00f801c4168b$05aebf20$f901a8c0@ws21> <20040330192619.GA6498@ergo.nruns.com> <010801c4168d$f595ede0$f901a8c0@ws21> <20040330154244.R1005@ganymede.hub.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Kind of, except both machines are on 2 different ISP, in different states.
At ApacheCon, I described how I have a server in Oregon and a server in
Florida, and I have a monitoring program that does a dynamic DNS update when
one of the hosts goes down.  The individual described that what they are
doing is something to do with BGP, in which they have multiple servers in
different countries, all with the same IP address.  Traffic is routed to the
nearest logical server, until one goes down, then the traffic is routed to
the nearest logical server that is still up.  That is what I am wanting.
Since I didn't get the person's contact information, I am trying to figure
out how to do it myself.

Sincerely,

Rick Duvall
----- Original Message ----- 
From: "Marc G. Fournier" <scrappy@hub.org>
To: "Rick Duvall" <rduvall@onlinehighways.net>
Cc: <jan.muenther@nruns.com>; <freebsd-questions@freebsd.org>
Sent: Tuesday, March 30, 2004 11:43 AM
Subject: Re: BGP On Host


>
> sounds like you are describing a load balancing switch ... two seperate
> boxes behind the switch, with a single "public" IP in front that sends a
> heartbeat to the boxes behind it ...
>
> On Tue, 30 Mar 2004, Rick Duvall wrote:
>
> > I wasn't sure if it was BGP or if it was something else.  Definetly
between
> > routers would be using BGP.  But, I heard at an apache conference
somebody
> > was doing something where the machine would send a keepalive to the
directly
> > connected Cisco router, and if the router didn't receive the keepalive
> > signal, BGP would re-route the traffic to the other host.  Both hosts
are on
> > different ISP, but have the same IP address.  Traffic is routed from the
> > requester to the closest logical server.  I think UltraDNS does this
with
> > their DNS servers as well.
> >
> > Anyway, I don't know what the host uses to send the keepalive to the
Cisco
> > router, or even how to configure the BGP to make it work.  I was
wondering
> > if somebody on the list has set up the same configuration on a couple of
> > fault tolerant FreeBSD boxes.
> >
> > Sincerely,
> >
> > Rick Duvall
> > ----- Original Message -----
> > From: <jan.muenther@nruns.com>
> > To: "Rick Duvall" <rduvall@onlinehighways.net>
> > Cc: <freebsd-questions@freebsd.org>
> > Sent: Tuesday, March 30, 2004 11:26 AM
> > Subject: Re: BGP On Host
> >
> >
> > > > (mirrored).  If both hosts are up, the traffic is routed to the
closes
> > > > server to the person making the request.  Otherwise, if one server
is
> > down,
> > > > traffic is automatically re-routed to the other box.
> > >
> > > That is not what BGP is made for. It's an exterior routing protocol
for
> > > routes between AS.
> > >
> >
> > _______________________________________________
> > freebsd-questions@freebsd.org mailing list
> > http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> > To unsubscribe, send any mail to
"freebsd-questions-unsubscribe@freebsd.org"
> >
>
> ----
> Marc G. Fournier           Hub.Org Networking Services
(http://www.hub.org)
> Email: scrappy@hub.org           Yahoo!: yscrappy              ICQ:
7615664
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?012001c4168f$d13592a0$f901a8c0>