Showing results for 
Search instead for 
Did you mean: 

Output drops and BGP flaps


Hi all,

I have been having a discussion with our Telco as we intermittently losing connectivity with their BGP peer.  The Telco are saying that they are not always seeing replies coming back from us when they send a BGP keepalive and this is due to output drops that we see on our interface to them and I am drawing a bit of a blank so any help would be appreciated.

Our WAN to the Telco never sees over 2Mb in our out and this is a 4Mb link so cannot be an overutilisation problem.

I have added BGP to our QOS policy so this will be getting a better share of the bandwith available.

Routing updates are given priority are they not? so the interface would have to be 100% utilised for BGP to drop and in any case we can go days without a lost BGP session.

I have attached my router interface / BGP and QOS config and again I would appreciate any help.

Thanks in advance,


1 Reply 1


Hi Jonathan,

you have bgp peer on gig interface, which should not be genrally causing any bottleneck,

to troubleshoot this further i would perform following task to further isolate the issue.

1. check if the CPU is highly ustilised when the telco reports any missing keepalive, when cpu is high i may missout to process the control packets even if its high priority

2. control plane traffic is by default prioritized over the data traffic either by pak_priority or ip precedence

3. also it will be great to have wireshark capture on this gig interface which will prove the point that if bgp keepalive are sent out, this will be thw proving point

hope this helps


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:

Recognize Your Peers