Re: Ospf redistrib into eigrp - 2 locations

Thanks Carlos for info

Aaron

On Mar 6, 2011, at 5:20 AM, Carlos G Mendioroz <tron_at_huapi.ba.ar> wrote:

> Ah, now I see some things I missed before.
>
> First, couple of global thing: EIGRP only carries one type of external
> routes, like OSPF's E1, so internal EIGRP cost does change the initial
> cost at redistribution point (cost seed). But there is no dynamic
> translation of original cost (OSPF) into EIGRP metrics, so it does not
> matter how good/bad OSPF metrics were at the redistribution points.
>
> Now, when R1 shows the route to R3, now R3 has 2 competing sources for
> this network: OSPF and EIGRP. Do metrics matter ? At a different level,
> because you can not compare OSPF's with EIGRP's. So administrative
> distance is used and EIGRP wins. Now, R3's initial metric is larger
> (and if you have split horizon, it won't even tell to R2).
>
> -Carlos
>
>
> Aaron @ 06/03/2011 02:23 -0300 dixit:
>> Thanks Carlos, That might be, however it didn't exactly seem that way in
>> what I saw...
>> Router 1 and 3 where the two routers redistributing ospf into
>> eigrp.....router 2 was the one router solely running eigrp. Router 1, 3 and
>> 2 share a vlan with one another.
>> if I didn't tell you before, router 1 and 3 are both connected to the same
>> WAN/OSPF side...learning much the same routes as each other....so point is
>> that they should be redis'ing the same set of routes into eigrp to that
>> router 2 that I mention. Also, I'm running a route-map upon redis at router
>> 1 and 3 and setting same metrics to be fed into eigrp to router 2.
>> When I redistributed from both routers 1 and 3 simultaneously I wanted to
>> see dual equal cost routes in router 2. I didn't. so, I removed
>> redistribution on router 1....then looked at router 2's metric for a route
>> learned via router 3. Say it was 12345678. Then I removed redis from
>> router 3 and enabled redis on router 1....then looked at router 2's metric
>> for a route learned via router 1. Say it was 12345678. ...point is, it
>> seems that when singularly redis'ing from 1 or 3, 2 sees same metric.
>> That's why it didn't appear that the network router 2 was seeing was further
>> away (metrically) via 1 compared to 3 or vice versa.
>> I'm still undergoing my learning in the area of redistributing routing info
>> in such a way that it could be back-fed via another router on same LAN
>> segment (BMA) or NBMA network perhaps which is connected back into the area
>> of the network that the redis'ing router is....so I don't know the details
>> enough to appreciate what exactly was going on, but it would seem that you
>> wouldn't want router 1 to send routes to router 3 or vice versa that they
>> already know about since they are both connected to the same WAN..... with
>> that in mind, I did the following to filter out route info on router 1 and 3
>> so they wouldn't learn anything from each other, but still be able to redis
>> their ospf info into the vlan shared with router 2....
>> "distribute-list in f1/1" on router 1 and 3. F1/1 is the interface on both
>> of those that connects to the vlan that they both share with router 2.
>> Now router 2 sees equal cost routes.
>> I don't really know why not having a distribute list filtering out route
>> info from router 1 and 3 would've caused router 2 to NOT have the dual equal
>> cost routes.....but this fixed it.
>> Aaron
>> -----Original Message-----
>> From: Carlos G Mendioroz [mailto:tron_at_huapi.ba.ar] Sent: Saturday, March 05, 2011 11:57 AM
>> To: Aaron
>> Cc: ccielab_at_groupstudy.com
>> Subject: Re: Ospf redistrib into eigrp - 2 locations
>> Cause the distance from that router to the redistribution places is different ?
>> -Carlos
>> Aaron @ 05/03/2011 13:36 -0300 dixit:
>>> I'm trying to redistrib ospf into eigrp at 2 locations; and at those 2
>> locations i set metrics to be 1 1 1 1 1
>>> There is a third router in the eigrp AS , purely eigrp. Wondering why
>> that third router is not seeing 2 equal cost routes in its routing table
>> (rib) from each of my 2 redistrib locations.?
>>> Aaron
>>>
>>>
>>> Blogs and organic groups at http://www.ccie.net
>>>
>>> _______________________________________________________________________
>>> Subscription information may be found at: http://www.groupstudy.com/list/CCIELab.html
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>
> --
> Carlos G Mendioroz <tron_at_huapi.ba.ar> LW7 EQI Argentina

Blogs and organic groups at http://www.ccie.net
Received on Sun Mar 06 2011 - 21:09:29 ART