05-22-2023 07:26 AM - last edited on 06-22-2023 03:28 AM by Translator
Hello,
I have two site-to-site VPN tunnels coming off two FTD 2110s using BGP to a third party. One tunnel is primary, with the other being backup. The third party's documentation suggests that in order to control tunnel selection, I must do the following: Set a higher local preference, which I have done, and
prepend AS_Paths
to force a lower preference. I'm stuck on the
AS_Path
, as I'm not really sure where this is done. Under General Settings->BGP, then General again, I tried editing the "Number of AS numbers in the
AS_Path attribute of received routes
from none to 1 on my primary and 10 on my secondary. This did not resolve my asymmetric routing issue, as my requests are leaving my network and going out of the primary VPN as it is designed to do, but my third party is sending responses back to my backup VPN. This was working fine for 2-3 weeks, then something must have changed on their side, and they see my backup as the preferred route for some reason. If someone could share where I go to
prepend the AS_Paths
,it would be greatly appreciated, or confirm that the section I modified was correct and I can go back to the third party.
Thanks,
Solved! Go to Solution.
05-24-2023 01:03 AM - last edited on 06-22-2023 04:18 AM by Translator
Hello
Are they within the same bgp ASN, as such do they have and IBGP peering with each other, if so then id would use local preference instead of weight and yes use the
route-map for as-prepending
Example2:
FW1
route-map LP_rm
set local-preference 10000
router bgp 10
neigbour <Fw2> remote-as 10
neigbour <Fw2> next-hop self
neigbour <primary isp> route-map LP in
FW2
route-map PrePend_rm
set as-path prepend 10 10 10
router bgp 10
neigbour <Fw1> remote-as 10
neigbour <Fw1> next-hop self
neigbour <backup isp> route-map PrePend_rm OUT
05-22-2023 07:39 AM - last edited on 06-07-2023 02:13 AM by Translator
the Local Perference control the outbound traffic
the
AS_PATH
control the inbound traffic
check this
Enter the Number of AS numbers in
AS_PATH
attribute. An
AS _PATH
attribute is a sequence of intermediate AS numbers between source and destination routers that form a directed route for packets to travel. Valid values are between 1 and 254. The default value is None.
05-22-2023 07:58 AM - last edited on 06-22-2023 03:01 AM by Translator
Hello @dcanady55,
To control how to enter in your AS for a particular subnet you can configure
AS prepend
. On the other hand, to influence how to exit from your AS for a particular Subnet (outbound) you configure Local-Preference (iBGP conserve the local pref value, not eBGP).As concerned,
AS_PATH
attribute you can increase the value on the "secondary" side. Note, that some provider dont accept more than 10 value. Add twice your
AS_PATH
should be good to influence as you wish. To have symetric flow, configure the local pref consequently. Default value is 100, greater is this value, better choice you have.
05-22-2023 02:51 PM - last edited on 06-22-2023 04:23 AM by Translator
Hello
@dcanady55 wrote:
Hello,
I have two site-to-site VPN tunnels coming off two FTD 2110s using BGP to a third party. One tunnel is primary, with the other being backup.
I assume then these two boxes are in a HA configuration as such it will be seen as a single logical unit with two EBGP peering towards the ISPs.
If this is the case then would suggest:
Egress traffic
use weight attribute towards your isps peers (higher value most preferred)
Ingress traffic
- use
AS-Path prepending
out towards the least preferred isp (backup) peer)
Example:
route-map PrePend_rm
set as-path prepend 10 10 10 router bgp 10
neigbour <primary isp> weight 60000
neigbour <backup isp> weight 40000
neigbour <backup isp> route-map PrePend_rm OUT
clear ip bgp * soft
05-23-2023 05:28 AM
Hi Paul,
No, these two FTDs are in separate DCs. They both are a part of an HA group within their respective DC though. Would you still recommend a route map?
05-23-2023 05:38 AM - last edited on 06-22-2023 02:16 AM by Translator
the issue is using
AS_PATH
prepend
or it missing from FMC/FDM ?
05-23-2023 05:49 AM - last edited on 06-22-2023 02:19 AM by Translator
The issue was I couldn't find anything on the FMC under BGP related to
prepending
the
AS_Path
. I know now from Paul's post that this is found within a
route map
.
05-23-2023 06:14 AM - last edited on 06-22-2023 03:20 AM by Translator
Yes but I think you need to use
route-map
with flexconfig.
or there is other
route-map
?
did you check as
attribute
?
05-24-2023 01:03 AM - last edited on 06-22-2023 04:18 AM by Translator
Hello
Are they within the same bgp ASN, as such do they have and IBGP peering with each other, if so then id would use local preference instead of weight and yes use the
route-map for as-prepending
Example2:
FW1
route-map LP_rm
set local-preference 10000
router bgp 10
neigbour <Fw2> remote-as 10
neigbour <Fw2> next-hop self
neigbour <primary isp> route-map LP in
FW2
route-map PrePend_rm
set as-path prepend 10 10 10
router bgp 10
neigbour <Fw1> remote-as 10
neigbour <Fw1> next-hop self
neigbour <backup isp> route-map PrePend_rm OUT
06-03-2023 01:32 PM - last edited on 06-22-2023 03:24 AM by Translator
Hi Friend, Other Cisco community member face same issue with BGP, he need to use
route-map
in flexconfig to modify MED/
AS-Prepend
can you share the flexconfig you use.
Thanks
MHM
06-12-2023 07:22 AM - last edited on 06-22-2023 03:26 AM by Translator
No flexconfig needed. I built a standard ACL then created a
route map
attaching the ACL to it. Inside that
route map
is where you can
prepend the AS path
under set clauses. Then you go back into BGP under routing and under neighbor you add the
route map out
.
06-12-2023 07:35 AM
thanks a lot I get the idea,
have a nice day
MHM
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide