From: ccie2be (ccie2be@nyc.rr.com)
Date: Fri Jan 14 2005 - 18:16:13 GMT-3
Marvin,
Thanks for pointing that out.
That certainly explains some of the problems I've had with this distance
command in the past.
Is the same true for ISIS since it's also a link state protocol?
Thanks again, Tim
----- Original Message -----
From: "marvin greenlee" <marvin@ccbootcamp.com>
To: "'ccie2be'" <ccie2be@nyc.rr.com>; <ccielab@groupstudy.com>
Sent: Friday, January 14, 2005 4:03 PM
Subject: RE: Distance vs Distance [bcc][faked-from]
> With OSPF, the router ID used is the router that is the SOURCE of the
route,
> not the router ID of the neighbor. ("show ip ospf database" will show you
> the router that is advertising the route into OSPF.)
>
> For example, with 4 OSPF routers:
>
> R1--R2--R3--R4
>
> If R1 is redistributing a route into OSPF from another protocol, and you
> want the distance for that route in R4's routing table to be modified, you
> would match the router ID of R1, since R1 is originating the route.
>
> - Marvin Greenlee, CCIE#12237, CCSI# 30483
> Network Learning Inc
> marvin@ccbootcamp.com
> www.ccbootcamp.com (Cisco Training)
>
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
> ccie2be
> Sent: Friday, January 14, 2005 12:42 PM
> To: Group Study
> Subject: Distance vs Distance [bcc][faked-from]
> Importance: Low
>
> Hi guys,
>
> Have any of you had difficulty learning when to use the different versions
> of
> this command.
>
> I have but now I think I got it and want to confirm with the pool of GS
> brain
> power.
>
> Several protocols support a version of distance that specific to that
> protocol.
>
> For example, distance eigrp, distance ospf, and distance bgp. (Rip and
isis
> don't seem to have an equivalent)
>
> In addition, there's the plain or IP version of distance <AD#> <ip addr
> mask>
> [acl] command.
>
> The way I think about these now is that the distance <protocol> version of
> the
> command is sort of "wholesale".
>
> It will change the AD for all routes in the route table that match that
> protocol (or class) within that protocol.
>
> While the distance <AD#> version of the command is more like "retail". It
> will
> affect those routes that that are either
>
> from a certain neighbor(s) or match the optional acl at the end of the
> command.
>
> Q1: Do people agree with this conceptualization?
>
> Q2: Can the same thing be done with ip version of the command that can be
> done with the protocol specific version
>
> as long as the neighbors and acl are properly defined?
>
> Q3: Has anyone come up with a simple way of knowing when using the ip
> version
> of the command which ip address to use
>
> when defining the source of the routes? I recall that sometimes it's the
> router ID of the neighbor but sometimes the physical
>
> address but can never remember which.
>
> All your thoughts and insights are appreciated.
>
> TIA, Tim
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Wed Feb 02 2005 - 22:10:23 GMT-3