Les, > On Dec 7, 2023, at 16:03, Les Ginsberg (ginsberg) <[email protected]> wrote: > > Folks - > > Let's be careful here. > SR-MPLS has been deployed for several years, there are multiple > implementations which have demonstrated interoperability, and clearly the > correct encoding of the SID is a key element of that interoperability. > > As a co-author, I am happy to listen to relevant feedback, but any textual > change which has the potential to even suggest that an actual change has been > made in encoding is clearly undesirable. > > John - I note you have already acknowledged any errata (or erratum š) would > be an editorial one - but given the above context and the fact that no one > over these many years has publicly voiced any concerns argues for caution. > I am sure you have more pressing issues, but as your post has already started > to cause waves, I would appreciate resolving this sooner rather than later.
Certainly no encoding change is being suggested - just a better clearer definition of the relationship between the L flag, V flag, TLV length, and TLV values (label, index, value). Thanks, Acee > > Thanx. > > Les > > > > -----Original Message----- > > From: Acee Lindem <[email protected] <mailto:[email protected]>> > > Sent: Thursday, December 7, 2023 12:44 PM > > To: John Scudder <[email protected] > > <mailto:[email protected]>> > > Cc: Hannes Gredler <[email protected] > > <mailto:[email protected]>>; > > [email protected] <mailto:[email protected]>; Les Ginsberg > > (ginsberg) <[email protected] <mailto:[email protected]>>; > > Clarence Filsfils (cfilsfil) <[email protected] > > <mailto:[email protected]>>; [email protected] > > <mailto:[email protected]>; > > [email protected] <mailto:[email protected]>; DECRAENE Bruno INNOV/NET > > <[email protected] <mailto:[email protected]>>; > > [email protected] <mailto:[email protected]>; Jeff Tantsura > > <[email protected] <mailto:[email protected]>>; Peter Psenak > > (ppsenak) <[email protected] <mailto:[email protected]>>; > > Horneffer, Martin <[email protected] > > <mailto:[email protected]>>; > > [email protected] <mailto:[email protected]>; > > [email protected] <mailto:[email protected]>; [email protected] > > <mailto:[email protected]>; > > [email protected] <mailto:[email protected]>; [email protected] > > <mailto:[email protected]>; lsr <[email protected] <mailto:[email protected]>> > > Subject: Re: [Lsr] Bug in RFC 8667 definition of SID/Index/Label > > > > Hi John, > > > > > On Dec 7, 2023, at 12:22, John Scudder <[email protected] > > > <mailto:[email protected]>> > > wrote: > > > > > > Hi Hannes, > > > > > >> On Dec 7, 2023, at 4:38 AM, Hannes Gredler > > <[email protected] > > <mailto:[email protected]>> wrote: > > >> > > >> We have used similar textblocks for the OSPFv2 and OSPFv3 SR extensions > > and I am not aware > > >> of any questions from implementators around ambiguity. > > > > > > Thanks for the pointer, Iāll take a look at those, too. > > > > > >> IMO there is clear enough language to describe proper encoding of the > > prefix-SID subTLV and > > >> I am not sure why an "erratum" is required. > > > > > > I agree that, after reconsidering the text in light of Lesās reply, itās > > > not a > > technical error (or ābugā as I put it in the subject line). However, offline > > feedback from a couple of other experienced protocol implementors indicates > > to me that Iām not the only one who finds the presentation of the > > information > > to be unclear [1] and not as helpful as it could be to someone using the > > document as a reference instead of doing an in-depth read-through. > > > > Weāll probably never BIS these RFCs but I would agree that it would be good > > for one of the RFC authors to provide a clearer definition of the > > relationship > > between the L flag, V flag, TLV length, and TLV values (label, index, > > value). > > Since it seems a single flag indicating whether the value is an MPLS label > > or > > index into an MPLS label range would have sufficed, this description would > > certainly be beneficial to those new to IGP segment routing. Also, it is > > unclear > > why an index/value ever needs to be 4 octets when the value is bounded by > > the MPLS label space itself. > > > > Thanks, > > Acee > > > > > > > > > > BTW if thereās some nuance to the quotation marks you used around > > āerratumā Iām missing it. Errata are a normal part of our process, and > > erratum > > is just the singular of errata. [2] > > > > > > Thanks, > > > > > > āJohn > > > > > > [1] This quote doesnāt quite apply, but itās a humorous way of > > > illustrating > > that information can be provided without being made available as clearly as > > it > > could be. http://hitchhikerguidetothegalaxy.blogspot.com/2006/04/beware- > > <http://hitchhikerguidetothegalaxy.blogspot.com/2006/04/beware-of-leopard-douglas-adams-quote.html> > > of-leopard-douglas-adams-quote.html > > <http://hitchhikerguidetothegalaxy.blogspot.com/2006/04/beware-of-leopard-douglas-adams-quote.html> > > > > > > [2] https://www.rfc-editor.org/errata-definitions/ > > > _______________________________________________ > > > Lsr mailing list > > > [email protected] <mailto:[email protected]> > > > https://www.ietf.org/mailman/listinfo/lsr
_______________________________________________ Lsr mailing list [email protected] https://www.ietf.org/mailman/listinfo/lsr
