> -----Original Message-----
> From: Rick Payne [mailto:rickp@rossfell.co.uk]
> Sent: 04 July 2001 10:14
> To: Pegg Damon
> Cc: 'cisco-nsp@puck.nether.net'
> Subject: RE: Nasty iBGP design issue...
>
>
>
> --On Wednesday, July 04, 2001 10:04:50 +0100 Pegg Damon
> <Damon.Pegg@carrier1.com> wrote:
>
> > Timers (holdtime) can be adjusted from the default 180
> seconds to any
> > 32bit integer other than 0, and it is recommended that the
> keepalives are
> > readjusted in keeping to be 1/6 of the holdtime.
>
> Out of interest, where did you get the 1/6th value from?
Oops, should have been 1/3 :)
>
> > The KEEPALIVE messages are quite short (19 octets), so we
> can use the
> > formula 19*8/k, where k is keepalive frequency; hence 5
> second keepalives
> > and and 30 second holdtimer gives only 30bps i/o per neighbor.
> > Processing, according to Cisco, is negligible.
>
> However, under load, I've seen Cisco's fail to get the
> keepalives out on
> time, causing some nasty instabilities.
Exactly why I wouldn't advocate it for the core mesh, not that its needed
there. Also, I have an IGP 0/0 advertised to the rr clients for network
ingress traffic. egress to that device is admittedly tricky hence the
balancing act between stability and fallover time - Basically why I didn't
want to use it if there was any alternative...
>
> Rick
>
This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:12:43 EDT