hi all, my AS is a multihomed non-transit ( connected to 2 differents AS).
with 1 AS (PRIMARY) i used IP address of physical interface to peer EBGP session, with other AS (BACKUP), i used loopback address to peering EBGP session.
when i bring BACKUP EGBP session up, this disturb the primary link and i have many trafic loss on the primary link, until i shutdown backup bgp session.
i d'ont know why, ?
ALL traffic (incoming and outgoing) pass through PRIMARY AS, BACKUP AS are use only when PRIMARY fail. also my AS is not transit, i have filter BGP update between these 2 AS and only localy generate route are advertise to these peer using
neighbor <neighbor ip> route-map Permit_local_route out
ip as-path access-list 13 permit ^$
route-map Permit_local_route permit 10
match as-path 13
. with primary AS bandwidth is 32M, with secondary AS bandwidth is 7M.
problem is that when a bring up bgp session with secondary AS, PRIMARY link slow down at max 9M. this appen until i shutdown bgp session with seondary AS.
i make bgp session with secondary AS using loopback, and this loopback is reacheble using static route through this AS.
it looks like that the "secondary" eBGP session is preferred when it is setup.
You can do the following:
-use AS path prepending of your own AS number towards secondary ISP
- advertise more specific prefixes to primary ISP and a less specific prefix to secondary ISP (feasible only if your address space is made of multiple /24: advertising prefixes more specific then /24 is usually a problem)
- contact the secondary ISP and ask them if they use BGP communities to allow you to tell them the level of preference they should apply to routes received from your AS.
In case this option is possible you need to set the BGP community to the expected value and you need to send the community to the neighbor with
neigh secondary-isp send-community
the BGP community can be set on the route-map applied to the secondary ISP neighbor.
- if none of the above works you can consider to use BGP conditional advertising to advertise your prefixes to secondary ISP only when primary BGP session fals
Starting from NFVIS 3.12 versions, the deploy option does not depict all the SR-IOV VFs(Virtual Functions) available in a physical interface. This change is introduced as (i) the number of VFs of ENCS platform on LANs side is increased to 24 and (ii) the...
Community Live- Getting to know Cisco SD-WAN
(Live event - formerly known as Webcast- Wednesday December 11, 2019 at 10 am Pacific/ 1 pm Eastern / 7 pm Paris)
This event will have place on Wednesday 11th, December 2019 at 10hrs PDT
Hi alli have 40 spots (40 Ethernet cables for computers coming out from switch) and i want each of these spots to have fix IP which means if i swap the computer the IP of certain spot remain the same.example : at spot 30 i have IP address of 192.168.22.40...
Cisco DNA Center nodes lost network connectivity. Cannot SSH to nodes. Cluster and Enterprise port connected to Cisco Nexus Switches.
Cisco DNA Center kernel logs showing hung queue error messages. "sudo cat /var/log/kern.log"
Cisco Digital Network Architecture Center Modules(Design Module)Wireless Part.In this article, we are going to talk about Cisco Digital Network Architecture Center design Module, Wireless Part.Cisco DNA Center gives us the flexibility and scalability to c...