<Keepalives are temporarily in throttle due to closed TCP window>

Richard A Steenbergen ras at e-gerbil.net
Thu Sep 17 18:11:11 UTC 2009


On Thu, Sep 17, 2009 at 09:17:00AM -0500, Michael Ruiz wrote:
> Oh you guys are going to love this...Before I could send out the maintenance notification for tonight to make changes,  The session has been up for 21 hours.  This is before I could put a packet capture server on the segment.  <Sigh> 
> 
> <SNIP>
>   BGP state = Established, up for 21:29:25
>   Last read 00:00:24, last write 00:00:02, hold time is 180, keepalive interval is 60 seconds
>   Neighbor capabilities:

You don't need to use an external sniffer, you can use "debug ip packet"
to see traffic being punted to the control plane, or in the case of the
6500 you can use ELAM or ERSPAN (though this is probably a little bit on
the advanced side). If this was an MTU mismatch a sniffer wouldn't
reveal anything other than missing packets anyways, which you could just
as easily deduce from a debug or looking at the retransmit counters on
the bgp neighbor.

http://cisco.cluepon.net/index.php/Using_capture_buffer_with_ELAM
http://cisco.cluepon.net/index.php/6500_SPAN_the_RP

My money is still on MTU mismatch. Assume the simplest and most likely 
explanation until proved otherwise.

-- 
Richard A Steenbergen <ras at e-gerbil.net>       http://www.e-gerbil.net/ras
GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)




More information about the NANOG mailing list