LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

Re: Route through rather than connect to possible?

To: Kyle Sparger <ksparger@xxxxxxxxxxxxxxxxxxxx>
Subject: Re: Route through rather than connect to possible?
Cc: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
From: Horms <horms@xxxxxxxxxxxx>
Date: Thu, 14 Sep 2000 15:54:28 -0400
On Thu, Sep 14, 2000 at 10:36:52AM -0400, Kyle Sparger wrote:
> > the client sends a connect request to the VIP:port on the director which 
> > is forwarded to the real-server, where the connect request is accepted.
> 
> Right, in this case, the director must have the IP address on the local
> machine.
> 
> In the case I was describing, the director would in fact be a gateway.
> The upstream router would send the packet to the appropriate VIP, through
> the director.  The director would intercept it enroute, and rewrite the
> frame/header and then push it back to one of the real servers.
> 
> It's like NAT, except rather than doing an address translation from
> non-routable network to routable network, it's doing an address
> translation to send it to one of the real servers.

Absolutely, this should work. In fact I was (and still am) planning on
incorporating documentation in Ultra Monkey on how to use routing protocols
to effect fail-over of LVS servers as an alternative to heartbeat.

In both of the following setups, fwmark rules aside, the Linux Directors
are acting as routers of sorts, that is traffic for the VIPs is being
routed through them, it just happens that as the Linux Directors see this
traffic it is dealt with as if it were traffic for a VIP.

http://ultramonkey.sourceforge.net/ultramonkey-1.0.1/topologies/example-ha-lb-4-fwmark.html

http://ultramonkey.sourceforge.net/ultramonkey-1.0.1/topologies/example-ha-lb-5-fwmark.html

-- 
Horms


<Prev in Thread] Current Thread [Next in Thread>