Re: Doubt in rfc 1990..
Fred Baker <[email protected]>
| Newsgroups | gmane.ietf.pppext |
|---|---|
| Message-ID | <[email protected]> |
it is perfectly legal to send some packets in PPP headers and some packets in Multilink headers. You'll find my name in the Acknowledgments; I (among others) spent a lot of time and email in discussion with the authors getting this right. When we designed PPP Multilink, we recognized that there were applications that might send IP packets that were small enough to not need fragmentation and for which the delays imposed by the fragmentation and reassembly procedures might be detrimental - voice on IP being the obvious example. We explicitly chose to permit them to be sent around the multilink channel. I think you will find several implementations that literally do that. Cisco's is called "Low Latency Queuing" by the marketing folks. On Oct 11, 2005, at 2:25 AM, Hariharasubramanian CS wrote: > I went through RFC 1990 (MP)... > > It doesn't state clearly if IP packets MUST be encapsulated in > Multilink Headers.... > > There is roundabout statement in the RFC : > > "Note that network protocols that are not sent using multilink headers > cannot be sequenced. (And consequently will be delivered in any > convenient way). " > > > What is the ietf's stand ? > Can somebody cast some light? > > Thanks > Hari. > > _______________________________________________ > Pppext mailing list > [email protected] > https://www1.ietf.org/mailman/listinfo/pppext > _______________________________________________ Pppext mailing list [email protected] https://www1.ietf.org/mailman/listinfo/pppext