[c-nsp] BGP convergence with jumbo frames

[Tony Li]

Pete,

ip mtu does affect both transit and originated traffic.

If you did not change the IP MTU, then you are causing
all of your TCP segments to undergo IP fragmentation.  If
one of those fragments is delayed or lost, for whatever
reason, you will trigger TCP retransmissions and kill
performance that way.

Tony


On Aug 2, 2004, at 8:22 AM, Pete Kruckenberg wrote:

> Thanks for all the responses. Some clarifications:
>
> The interface MTU (on both ends of the point-to-point fiber)
> was set to 9000 bytes for all tests.
>
> I adjusted "ip tcp mss" from 536 to 8636 (which IIRC was the
> "max data segment" shown in the "show ip bgp neighbor"
> output prior to fiddling with anything). I double-checked
> the "max data segment" on the BGP neighbor after each MSS
> change to make sure it matched up.
>
> MTU path discovery was on, and I double-checked the max data
> segment under the show ip bgp neighbor. Though in this case
>
> I didn't play with "ip mtu" because I couldn't determine
> from Cisco documentation if that affects just traffic
> originating on the router or if it also affects traffic
> passing through the router. These links have to support
> 9000-byte jumbo frames. The documentation for "ip tcp mss"
> was more clear that it only affects router-originated
> traffic.
>
> Pete.
>
> On Mon, 2 Aug 2004, Tantsura, Jeff wrote:
>
>> Date: Mon, 2 Aug 2004 08:52:11 +0200
>> From: "Tantsura, Jeff" <jeff.tantsura at capgemini.com>
>> To: Pete Kruckenberg <pete at kruckenberg.com>, cisco-nsp at puck.nether.net
>> Subject: RE: [c-nsp] BGP convergence with jumbo frames
>>
>>
>> IP MTU ?
>>
>> -----Original Message-----
>> From: cisco-nsp-bounces at puck.nether.net
>> [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Pete 
>> Kruckenberg
>> Sent: Monday, August 02, 2004 7:42 AM
>> To: cisco-nsp at puck.nether.net
>> Subject: [c-nsp] BGP convergence with jumbo frames
>>
>> Spent some time recently trying to tune BGP to get convergence down as
>> far as possible. Noticed some peculiar behavior.
>>
>> I'm running 12.0.28S on GSR12404 PRP-2.
>>
>> Measuring from when the BGP session first opens, the time to transmit
>> the full (~128K routes) table from one router to another, across a
>> jumbo-frame (9000-bytes) GigE link, using 4-port ISE line cards (the
>> routers are about 20 miles apart over dark fiber).
>>
>> I noticed that the xmit time decreases from ~ 35 seconds with a 
>> 536-byte
>> MSS to ~ 22 seconds with a 2500-byte MSS.
>>> From there, stays about the same, until I get to 4000, when
>> it beings increasing dramatically until at 8636 bytes it takes over 2
>> minutes.
>>
>> I had expected that larger frames would decrease the BGP converence
>> time. Why would the convergence time increase (and so significantly) 
>> as
>> the MSS increases?
>>
>> Is there some tuning tweak I'm missing here?
>>
>> Pete.
>>
>>
>> _______________________________________________
>> 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/
>>
>>
>> Our name has changed.  Please update your address book to the 
>> following format: "recipient at capgemini.com".
>>
>> This message contains information that may be privileged or 
>> confidential and is the property of the Capgemini Group. It is 
>> intended only for the person to whom it is addressed. If you are not 
>> the intended recipient,  you are not authorized to read, print, 
>> retain, copy, disseminate,  distribute, or use this message or any 
>> part thereof. If you receive this  message in error, please notify 
>> the sender immediately and delete all  copies of this message.
>>
>
> _______________________________________________
> 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/
>