[c-nsp] Sup1a MSFC2 Policy Based Routing

[Jason Ford]

Trent,

Well it is good to know that we aren't the only the ones that has this 
problem. We are not running MLS on our 6500's as far as I know unless it 
is on by default (which I think it is). We have a few ports in 
switchport mode but other than that, everything else is routing with 
BGP. We generally have a simple configuration.

I will send you an email if we can get it worked out or figure out a 
work around, but as it stands right now we can't get it to work.

Regards,

jason

Trent Lloyd wrote:
> Hi Jason,
>
> One of the guys here was fiddling with this on our 6500 w/SUP1A-MSFC2
> and was seeing some similar problems, what was worse is a few random
> packets from *other interfaces* did hit the route-map and that caused
> some extra funkyness, not sure whats going on there but we haven't
> visited it again since -- do let me know if you find a solution.
>
> Are you guys running MLS?
>
> Cheers,
> Trent
>
> On Tue, Dec 26, 2006 at 06:18:24PM -0500, Jason Ford wrote:
>   
>> Richard,
>>
>> Tried setting it to 4.4.4.9 (which would be the ip address assigned to 
>> my interface on g1/1) and still no luck. It just seems that the BGP 
>> route is overshadowing the PBR. It doesn't forward the traffic out as 
>> the route-map tells it to. This is why I was perplexed in the first place :)
>>
>> Any other ideas or suggestions would be greatly appreciated.
>>
>> Regards,
>>
>> jaosn
>>
>> Richard Gallagher wrote:
>>     
>>> Try setting the next-hop to the directly connected address that 
>>> 4.4.4.10  is reachable via.
>>>
>>> Jason Ford wrote:
>>>       
>>>> Hey all,
>>>>
>>>> I am having some problems getting policy based routing (using a 
>>>> access-list to identify source) working on a 6500 with a Sup1a / 
>>>> MSFC2 card. Background of the setup is the following:
>>>>
>>>> 1. Have 4 BGP peers using equal weight in and out for routing choices.
>>>> 2. 6500 is running in Native IOS with version 
>>>> c6sup12-ps-mz.121-26.E3.bin and c6msfc2-boot-mz.121-26.E3.bin.
>>>>
>>>> I am trying to use source based routing to control one of my subnets 
>>>> outbound traffic to use a lower cost BGP peer over the other 3. The 
>>>> source address in this case we will use 1.1.1.0/28 and is directly 
>>>> connecting to f 3/10. The BGP peer router I want the traffic to go 
>>>> out is 4.4.4.10 and is directly connected to g 1/1. So I setup the 
>>>> following route-map and identify the source network via an 
>>>> access-list with the following configuration.
>>>>
>>>> access-list 180 permit ip 1.1.1.0 0.0.0.16 any
>>>>
>>>> route-map peer-out permit 10
>>>>  match ip address 180
>>>>  set ip next-hop 4.4.4.10
>>>>
>>>> This is where I get a bit lost. I assigned this route map to the 
>>>> interface in which the 1.1.1.0 network was attached with this.
>>>>
>>>> interface f 3/10
>>>> ip policy route-map peer-out
>>>>
>>>> I don't see the policy routing matches increasing when I do a show 
>>>> route-map peer-out. On top of that, if I traceroute from any server 
>>>> in the 1.1.1.0/28 network to the outside world, it still uses the BGP 
>>>> preferred route from the route table. Is there something I am doing 
>>>> wrong here to override the outbound traffic for this subnet to go 
>>>> over a desired BGP peer?
>>>>
>>>> Any help would be greatly appreciated.
>>>>
>>>> Regards,
>>>>
>>>> jason
>>>>
>>>>
>>>> _______________________________________________
>>>> 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/
>>>>         
>> _______________________________________________
>> 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/
>>