Re: Clarification in size of Rtr and network LSA

Mike Fox <mjfox@US.IBM.COM> Fri, 13 June 2003 13:17 UTC

Received: from cherry.ease.lsoft.com (cherry.ease.lsoft.com [209.119.0.109]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02517 for <ospf-archive@LISTS.IETF.ORG>; Fri, 13 Jun 2003 09:17:19 -0400 (EDT)
Received: from PEAR.EASE.LSOFT.COM (209.119.0.19) by cherry.ease.lsoft.com (LSMTP for Digital Unix v1.1b) with SMTP id <10.00A12D70@cherry.ease.lsoft.com>; Fri, 13 Jun 2003 9:17:17 -0400
Received: from PEACH.EASE.LSOFT.COM by PEACH.EASE.LSOFT.COM (LISTSERV-TCP/IP release 1.8e) with spool id 45535321 for OSPF@PEACH.EASE.LSOFT.COM; Fri, 13 Jun 2003 09:17:15 -0400
Received: from 32.97.110.133 by WALNUT.EASE.LSOFT.COM (SMTPL release 1.0i) with TCP; Fri, 13 Jun 2003 09:17:15 -0400
Received: from westrelay04.boulder.ibm.com (westrelay04.boulder.ibm.com [9.17.193.32]) by e35.co.us.ibm.com (8.12.9/8.12.2) with ESMTP id h5DDHC2R213748 for <OSPF@PEACH.EASE.LSOFT.COM>; Fri, 13 Jun 2003 09:17:13 -0400
Received: from d03nm118.boulder.ibm.com (d03av02.boulder.ibm.com [9.17.193.82]) by westrelay04.boulder.ibm.com (8.12.9/NCO/VER6.5) with ESMTP id h5DDH90m152340 for <OSPF@PEACH.EASE.LSOFT.COM>; Fri, 13 Jun 2003 07:17:11 -0600
X-Mailer: Lotus Notes Release 5.0.11 July 24, 2002
X-MIMETrack: Serialize by Router on D03NM118/03/M/IBM(Release 6.0.1 [IBM]|May 21, 2003) at 06/13/2003 07:17:10
MIME-Version: 1.0
Content-type: text/plain; charset="US-ASCII"
Message-ID: <OF98E391D3.19DFF603-ON85256D44.0047AE33-85256D44.00489EAA@us.ibm.com>
Date: Fri, 13 Jun 2003 09:17:08 -0400
Reply-To: Mailing List <OSPF@PEACH.EASE.LSOFT.COM>
Sender: Mailing List <OSPF@PEACH.EASE.LSOFT.COM>
From: Mike Fox <mjfox@US.IBM.COM>
Subject: Re: Clarification in size of Rtr and network LSA
To: OSPF@PEACH.EASE.LSOFT.COM
Precedence: list

Krishna Rao wrote:

Hi,
> Usually we allocate
> (Max MTU size - (OSPF header size  + IP header size +
> MD5 authentication size) --> 1500 - (28 + 20 + 16)= 1436
> and start filling the LSA. For Point to point interface we add two
> links and need 24 bytes for each point to point interface. That results
> in supporting only 59 interfaces in a single area. Is this an acceptable
argument?
> or Should we design such that OSPF sends a LSA more than MTU size and
gets fragmented in IP.
> What is the scalabilty figure for Number of interfaces in a area for
popular routers?

Krishna,

We have been facing this problem for years.  Our platform supports virtual
interfaces (VIPA) and they can run into the hundreds on one host.  We have
found ourselves having to build router LSAs that are much larger than the
largest MTU size.  While you like to avoid IP fragmentation when you can,
in this case you simply can't.  However, as RFC 2328 makes clear in
Appendix A.1, Encapsulation of OSPF Packets, you can build any OSPF packet
including the router LSA up to 65535 bytes (including all headers) in size.
In our experience not all the OSPF implementations support this properly,
but I think the word is getting out and more of them are as time goes by.

Also note that this problem is solved in OSPFv3, where router LSAs are not
required to flow as one packet.

Mike
-----------------------------------------------------------------------
Enterprise Network Solutions
-----------------------------------------------------------------------
Research Triangle Park, NC  USA