[c-nsp] Operational experiences of aggressive bgp keepalive timers in private-IP (non-internet) networks?

[P C]

Thanks guys.  Maybe I'll compromise on 1/4 and see how it works at a
few sites, and monitor the logs for hold timer expired, etc.

I'd love to do BFD, but:

For internal links, Cisco chooses to license it with the "DATA"
license on ISRs, limiting it's adoption on CE equipment.  The
economics just aren't there for a basic IP-in IP-out box.  It's really
a simple feature that shouldn't be with licensing for stuff like MPLS,
L2TPV3, DECNET, IPX, etc... but that's just my opinion.

and

For external links, my SP can't suppose it on all their equipment,
they tell me due to control plane CPU issues if all customers used it.




On Mon, Nov 7, 2011 at 12:43 AM, David Hughes <David at hughes.com.au> wrote:
>
> Not specifically on ISR's, but we ran BGP timers of 1/5 on iBGP peers for years without issue.  That included LAN, metro dark fibre, and interstate managed ethernet attached devices.  In the mix of devices were various generations of 7200's which would have far less control plane processing power than current ISR's
>
>
>
> David
> ...
>
>
> On 04/11/2011, at 11:39 AM, P C wrote:
>
>> What experiences have you had using very aggressive BGP timers on
>> ISR's connecting to a service provider IP VPN/MPLS services on T1 and
>> Ethernet links?
>>
>> Assuming the proper QOS is in place, have values as low as 1/3 or 2/6
>> proven reliable in production operations?
>> _______________________________________________
>> 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/
>
>