[j-nsp] Bgp hold time

[Wei Keong]

Hi Steven,

Ya, agree that it could be due to the link congestion sometimes. But, we 
do observe hold time expire in non-congested links. The strange thing is 
the problem happens once in a while in M5/M20 router. Do you know if there 
is a bug reported in Junos v5.6?

Thanks,
Wei Keong


On Fri, 22 Oct 2004, Steven Wong wrote:

> Hi Wei Keong,
>
> The default hold time value being selected shouldn't be related to the
> problem you have seen coz. the session will pick the shortest one
> anyway.  With the TCP re-tx capability, even if there is a KA packet
> lost on the wire, the router should be able to re-send the KA packet
> within 90 secs hold-time period.
>
> From your description, it looks like either you have got a real
> congested link which drops the BGP KA pkts or you have seen a bug. I
> would suggest you to open a case with JTAC to check what's wrong.
>
> Regards,
> Steven
>
> -----Original Message-----
> From: juniper-nsp-bounces at puck.nether.net
> [mailto:juniper-nsp-bounces at puck.nether.net] On Behalf Of Wei Keong
> Sent: Friday, October 22, 2004 2:39 PM
> To: juniper-nsp at puck.nether.net
> Subject: [j-nsp] Bgp hold time
>
> Hi Everyone,
>
> Does anyone know why Juniper choose to have a shorter hold time compared
>
> to Cisco? Is there any advantage/disadvantage in doing so?
>
> We observed bgp flapping (juniper - juniper peering, juniper - cisco
> peering) because of Hold Timer Expire. Do you guys experience this
> before?
>
> Thanks,
> Wei Keong
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/juniper-nsp
>
>