[c-nsp] RE: [nsp] Two Routes to destination - failover

[Ejay Hire]

Hello.

A couple of options.

Normal static routes won't work, because the wireless
interface won't go down/down even if the link fails, so they
are out the window.

A route map that uses a service assurance agent or cdp to
verify reachability would work.

A routing protocol would work.  If this is inside your
network, and the remote site is small, RIP would be okay.
The only problem with RIP would be that it won't notice if
the link is dropping 50% of the packets, it only cares if
the hello's get through.  Same problem with Rip v2, OSPF,
BGP...  But, EIGRP has the capability of monitoring the
"reliability" of a link. 

The catch is, EIGRP doesn't monitor reliability by default.
You have to turn it on.  The documentation mumbles something
about K values and doing this.

Good luck.

To change the k's, 
Router eigrp xxx
metric weights tos k1 k2 k3 k4 k5

-Ejay


> -----Original Message-----
> From: cisco-nsp-bounces at puck.nether.net 
> [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of
Paul Stewart
> Sent: Monday, July 12, 2004 8:03 PM
> To: cisco-nsp at puck.nether.net
> Subject: [nsp] Two Routes to destination - failover
> 
> I"m looking for a simple approach to an upcoming project.
> 
> We currently have a remote site that is working fine.  For

> redundency we would
> like to take a separate physical route to this location
using 
> wireless and
> make this new connection the "primary" connection and the 
> existing connection
> a "backup" connection.
> 
> Can this be done using two route statements and metic
values? 
>  I"m looking for
> a really simple solution that will push all traffic via a 
> primary link and
> fallback to the secondary link if there's a failure..
> 
> Thanks,
> 
> Paul
> 
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/