cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1772
Views
0
Helpful
17
Replies

BGP does not advertise to ISP

JD87440
Level 1
Level 1

Haven't had to do this in a number of years and cannot get our new ISP to see any advertised routes for BGP.

Tried hacking away at prefix-lists, etc. but just cannot seem to get advertisements out.

It's got to be something staring me in the face but I just can't see it after a couple of days. Any ideas?

 

We're using this:

router rip
version 2
redistribute bgp 123456
network AAA.0.0.0
!
router bgp 123456
bgp router-id BBB.BBB.BBB.2
bgp log-neighbor-changes
network AAA.AAA.AAA.0 mask 255.255.255.0
redistribute rip
neighbor BBB.BBB.BBB.1 remote-as 174
!
ip route 0.0.0.0 0.0.0.0 BBB.BBB.BBB.1
!
!
ip prefix-list DEFAULT-ROUTE seq 5 permit 0.0.0.0/0
!
ip prefix-list PREPEND-LIST seq 5 permit AAA.AAA.AAA.0/24
logging trap debugging
access-list 30 deny 224.0.1.35 log
access-list 30 deny 224.0.1.60 log
access-list 30 deny 224.0.1.3 log
access-list 30 deny 224.0.1.2 log
access-list 30 deny 224.0.1.22 log
access-list 30 deny 224.0.1.24 log
access-list 30 remark Multicast Filtering ACL
access-list 30 deny 224.0.0.0 0.0.0.255 log
access-list 30 deny 239.0.0.0 0.255.255.255 log
access-list 30 permit 224.0.0.0 15.255.255.255 log
access-list 101 permit ip any any
!
route-map PREPEND-MAP-B permit 10
match ip address prefix-list PREPEND-LIST
set as-path prepend 123456 123456 123456 123456
!
route-map PREPEND-MAP-B permit 20
!
route-map DEFAULT-ONLY permit 10
match ip address prefix-list DEFAULT-ROUTE

 

 

17 Replies 17

Muhammad Awais Khan
Cisco Employee
Cisco Employee

Hi,

 

It seems with your configuration that you are advertizing, you can verify it with below command to see what has been advertized:

 

show ip bgp neighbor bbb.bbb.bbb.1 advertised-routes

 

Thank  you for the reply.

I get the following results when issuing the show ip bgp neighbor bbb.bbb.bbb.1 advertised-routes command -

 

Network Next Hop Metric LocPrf Weight Path
*> AAA.AAA.AAA.0/25 0.0.0.0 0 32768 i
*> BBB.BBB.BBB.0/29 0.0.0.0 0 32768 i

Total number of prefixes 2

 

I'm surprised by the weight.

Francesco Molino
VIP Alumni
VIP Alumni
Hi

To see what you're advertising, you can use the command show ip bgp neighbor x.x.x.x advertised-route.

I see you prefix-list and route-map.
For the rmap doing as-path prepend, you need to attach it to your neighbor statmenent like neighbor x.x.x.x route-map PREPEND-MAP-B out.

What is the goal of your default route route-map?

Thanks
Francesco
PS: Please don't forget to rate and select as validated answer if this answered your question

Thank  you for the reply.

I get the following results when issuing the show ip bgp neighbor bbb.bbb.bbb.1 advertised-routes command -

 

Network Next Hop Metric LocPrf Weight Path
*> AAA.AAA.AAA.0/25 0.0.0.0 0 32768 i
*> BBB.BBB.BBB.0/29 0.0.0.0 0 32768 i

Total number of prefixes 2

 

I'm surprised by the weight.

 

What is the goal of my default route route-map? I am just trying to manipulate a configuration of my predecessor and would gladly simplify if I could. It's been years since I've had to do anything BGP even for classes and definitively out of practice.

 

The big difference and challenge for me with this one is that there is only one /24 range allotted to us by this provider. I've only worked with two separate /24 subnet ranges previously. So I'm sure the pre-pend is part of our problem here.

I've tried to divide it in half and use a /25 range on each circuit. Not sure how forgiving or specific the upstream neighbor will be.

 

I don’t think Cogent will accept a /25 and certainly not a /29. 

 

Jon

You were right about the providor not accepting anything less than the full /24. Not sure how I'm supposed to configure the other circuit now.

 

It's not clear whether you only have one /24 for both circuits or a /24 from each provider ?

 

What exactly do you want to do with your addressing ?

 

Jon

The weight is normal because a local prefix you're advertising.
ISPs will block any subnet lower than a /24 because this is the lowest subnet advertised on the internet.

What are you trying you achieve exactly?

Thanks
Francesco
PS: Please don't forget to rate and select as validated answer if this answered your question

We migrated to a new ISP. Two circuits.
Previously had two circuits as well but had two ranges assigned.

We'd like to create fault tolerance between the two circuits (each having a
different head end location).

This provider said that they can do it with one range of /24.

Don't know how to achieve that yet. Dividing in half with /25s didn't work.

 

If you just want fault tolerance advertise /24 out of both circuits and AS prepend on one circuit so all traffic uses the other circuit unless it is unavailable. 

 

Or advertise out /24 as is out of both and let the routing take care of it although you are more likely to get asymmetric traffic. 

 

Jon

Thanks. I'll try that.

The providor asked that we use a community string with the providor's AS &
(weight?)
ex:
route-map providername permit 10
set community 124:125

(default is 130 so this was supposed to be less priority)
What I ended up with the the config was what looked like the old style

route-map providername permit 10
set community 11403389

So when using "sh ip bfp nei XXX.XXX.XXX.XXX advertised-routes" I get the
same results on both routers, the one with the community string and the one
without -

Network Next Hop Metric LocPrf Weight Path
*> XX.XX.XX.0/24 0.0.0.0 0
32768 i


Ok now i got your point. If you want to implement redundancy then you can advertise the same /24 and add a route-map to increase the as-path using the as-path prepend command on your 2nd link.
This way the return traffic from anyways come to your 1st link and if it fails it will go through the second link.

Thanks
Francesco
PS: Please don't forget to rate and select as validated answer if this answered your question

Thanks. The providor asked that we use a community string with the
providor's AS & (weight?)
ex:
route-map providername permit 10
set community 124:125

(default is 130 so this was supposed to be less priority)
What I ended up with the the config was what looked like the old style

route-map providername permit 10
set community 11403389

So when using "sh ip bfp nei XXX.XXX.XXX.XXX advertised-routes" I get the
same results on both routers, the one with the community string and the one
without -

Network Next Hop Metric LocPrf Weight Path
*> XX.XX.XX.0/24 0.0.0.0 0
32768 i


To have the new community format, you need to enter the command:
ip bgp-community new-format

Sometimes ISPs give you some community to let them know which one is primary vs secondary.
So on site 1 you set a community A and on site 2 community B. With that information they will always advertise A as primary and B as secondary.
If in addition to that you set the as-path on site2, you'll make sure it'll always be advertised as secondary, no matter what the ISP is doing on their side with communities.

Thanks
Francesco
PS: Please don't forget to rate and select as validated answer if this answered your question
Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: