Re: Transperant HTTP Redundancy.

FYI: The Inktomi Traffic Server does implement a failover
mechanism very much like what you describe. Also, via
products like the Alteon switch there are additional
failover options with our product.

One of the other benifits of Traffic Server is that the
pair of machines can load balance the traffic during the
normal (non-failure) case. Because TS supports true
clustering, a two node system provides a unified cache which
is truely 2X as large as either cache alone. Naive load
balancing will cause a large number of pages to become
replicated inside both caches, wasting valuable cache space.

Both the fault tolerance feature and the true unified
cluster cache work in both transparent and non-transparent
configurations. This is not true of any other products of which
I am aware.

Let me know if I can provide you with any more specific details.

Paul

-- 
Paul Gauthier,                                   (650)653-2800
CTO, Inktomi Corporation                  gauthier@inktomi.com
Javier Puche. CSIC RedIRIS wrote:
> 
> Hi,
> 
>  A very simple way of getting the redundancy you are looking for is:
> 
> - Set just one next-hop IP at the router. Let's say we have IP x.x.x.x
> for the proxy-cache service.
> - Have two or more machines in the same ethernet segment and IP subnet.
> Let's call them 1 and 2.
> - Have both machines with IP addresses different than x.x.x.x but with
> the ability to listen to x.x.x.x with a virtual interface when needed.
> - When machine 1 attending x.x.x.x falls down, have the other machine
> (2) detect that situation and set up a virtual interface for attending
> x.x.x.x , do it also the other way around.
> - Make sure that if either machine goes up again doesn't listen to IP
> x.x.x.x before checking that the other machine is not doing that
> already.
> 
>  I know how to set up virtual interfaces with Solaris and Linux via
> ifconfig, I suppose most other unix flavors can also do it.
> 
> Regards,
> 
> Javier Puche.