1 router, 2 ISP's running BGP. How to prefer one received default route from ISP over the other?

[pcweber]

I have an ASR 1000. 2 ISP's and gettign partial routes from each. Also, gettign default routes from each. OF course the one I prefer is never the populated default route unless I shut the BGP session down to the other ISP. So this proves I am gettigna  Def route from both. I tried setting up a route map and applying it to the inbound BGP. Does not seem to change anything. Here is what I have for related config.

router bgp 15263

 no bgp transport path-mtu-discovery

 bgp log-neighbor-changes

 network 198.175.140.0

 network 198.175.141.0

 timers bgp 12 40

 neighbor 41.14.12.221 remote-as 3356

 neighbor 41.14.12.221 description **L3 BGP**

 neighbor 41.14.12.221 prefix-list prefixfilter#2 out

 neighbor 41.14.12.221 route-map PREPEND#2 out

 neighbor 169.40.253.129 remote-as 7029

 neighbor 169.40.253.129 description WINDSTREAM BGP PEER*****

 neighbor 169.40.253.129 soft-reconfiguration inbound

 neighbor 169.40.253.129 prefix-list prefixfilter#2 out

neighbor 169.40.253.129 route-map PREPEND#3 out

 neighbor 169.40.253.129 route-map from-WS in (this is the BGP neighbor I wish to have the default route be less preferred. And any revived routes for that matter)

Below is the route-map that is applied inbound. As you see I am trying to add metric, local pref, weight, and even AS prepends. Nothing seems to impact the inbound received Def route. Also, is there a way to affirm the received route details like you can with outbound bgp routes by using the "sh ip bgp nei x.x.x.x advertised routes?

route-map from-WS permit 10
set metric 99999
set local-preference 88888
set weight 777
set as-path prepend 7029 7029 7029 7029

[Jon Marshall]

The weight on it's own should be enough. 

What does the output of "sh ip bgp" look like ? 

Did you do a "clear ip bgp 169.40.253.129 soft in" after applying your route map. 

Jon

[paul driver]

Hello

Do the default have a assigned weight value already if so you may need to increase the weight value you have already applied to the route-map?

sh ip bgp | in 0.0.0.0

[Georg Pauwen]

Hello,

on a side note, I assume you do not use all these values in the same route map ? The below should be sufficient:

route-map from-WS permit 10
set weight 777
!

route-map from-WS permit 20

!

neighbor 169.40.253.129 route-map from-WS in

This will ensure that routes from this neighbor (169.40.253.129) are preferred.

[Giuseppe Larosa · Accepted solution]

Hello @pcweber ,

>> neighbor 169.40.253.129 route-map from-WS in (this is the BGP neighbor I wish to have the default route be less preferred. And any revived routes for that matter)

but your route-map is doing the opposite rising weight and local preference !

all you need is to assign a less then 100 (default value)  local preference to received routes

The highest local pref is preferred

route-map from-WS permit 10

set local-preference 80

alternatively you can increase the weight on the other neighbor.

Hope to help

Giuseppe

[Jon Marshall]

Giuseppe

Good spot, I missed the bit about less preferred.

Jon

[pcweber]

OK, my alerts to the forum posts is not working so I did not realize there were replies. The route- map below achieved the goal. The one I had was the result of incrementally adding things to see if I could get it to affect the received def. route. Thank you and this did it as I had the preference value backwards.

route-map from-WS permit 10
set local-preference 88