Dear Michael, I agree with you. "Best way to handle transit links is for prefixes to only be suppressed from Link LSAs".
Above solution will help us In a partial deployment scenario, where upgraded routers (NON-DR with prefix hiding configured) and not-yet-upgraded routers(DR) may coexist. Thanks & Regards Rajesh -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Michael Barnes Sent: Thursday, February 23, 2012 2:26 AM To: [email protected] Subject: Re: [OSPF] OSPF WG Last Call for "Hiding Transit-only Networks in OSPF " - <draft-ietf-ospf-prefix-hiding-02.txt> Folks, To expand on Marek's comment, please consider the scenario of a transit link with routers A and B (and perhaps others) where A is the DR and configured with prefix suppression but B has a global prefix which needs to be advertised. In this scenario A must include B's prefix in its intra-area-prefix LSA for the Link. Therefor I believe the best way to handle transit links is for prefixes to only be suppressed from Link LSAs. The DR in this case will not have put its own global prefix in its Link LSA and that will keep it out of the intra-area-prefix LSA but the DR prefix LSA can still be built according to RFC5340 section 4.4.3.9 Regards, Michael On 02/22/2012 06:45 AM, Marek Karasek (mkarasek) wrote: > Hi Acee, > > one can argue that if prefix-suppression is configured on DR, DR will > not produce Intra-Area-Prefix-LSA corresponding to the transit network. > To some extend it's equivalent of v2 behavior, prefix suppression has to > be configured on DR in order to hide the subnet on the broadcast > segment. > > In any case, it would be good if draft specify: > - if prefixes are suppressed from Intra-Area-Prefix-LSA only, or also > from link LSA. > - if prefix-suppression is configured on DR, will be suppressed only > own prefixes, or also prefixes learnt from BDRs/DROTHERs via link LSAs. > > I read current draft such that prefixes are suppresses only from > Intra-Area-Prefix-LSA and if suppression configured on DR, DR will not > produce Intra-Area-Prefix-LSA > corresponding to the transit network. I'm OK with it, but other models > are acceptable too. > > Thanks marek > >> -----Original Message----- >> From: [email protected] [mailto:[email protected]] On Behalf > Of >> Acee Lindem >> Sent: Wednesday, February 22, 2012 2:54 PM >> To: Yi Yang (yiya) >> Cc: OSPF List >> Subject: Re: [OSPF] OSPF WG Last Call for "Hiding Transit-only > Networks >> in OSPF " -<draft-ietf-ospf-prefix-hiding-02.txt> >> >> Hi Yi, >> The global addresses do need to be suppressed from OSPFv3 Link-LSAs >> since they are used by the DR to produce the Intra-Area-Prefix-LSA >> corresponding to the transit network >> Thanks, >> Acee >> >> On Feb 22, 2012, at 8:21 AM, Yi Yang wrote: >> >>> Hi Rajesh, >>> >>> Thanks for your comments. I agree that prefix-hiding must not be >> configured on VL interface. But link-LSAs are not being used in SPF >> calculations. >>> >>> Yi >>> >>> >>> >>> On Feb 20, 2012, at 10:22 PM, Rajesh wrote: >>> >>>> Dear Acee, >>>> >>>> Some more thoughts. >>>> >>>> 1) In section 3 of this draft " Hiding IPv6 Transit-only Networks > in >> OSPFv3" >>>> We can also mention about LINK LSA. If the prefix hiding is >> configured on a >>>> interface, then no need to mention global prefixes in Link LSA. >>>> >>>> 2) For prefix hiding configuration option, may be we need to > support >> OSPF >>>> and OSPFv3 MIBs as well. >>>> >>>> Thanks >>>> Rajesh >>>> >>>> >>>> -----Original Message----- >>>> From: Acee Lindem [mailto:[email protected]] >>>> Sent: Monday, February 20, 2012 7:49 PM >>>> To: Rajesh >>>> Cc: OSPF List >>>> Subject: Re: [OSPF] OSPF WG Last Call for "Hiding Transit-only >> Networks in >>>> OSPF " -<draft-ietf-ospf-prefix-hiding-02.txt> >>>> >>>> Hi Rajesh, >>>> >>>> On Feb 20, 2012, at 12:55 AM, Rajesh wrote: >>>> >>>>> Dear Acee, >>>>> >>>>> We can add one statement in section 3 of this draft [which is for >> OSPFv3] >>>>> "LA-bit prefix we advertise in intra area prefix LSA for Virtual >> LINK IPv6 >>>>> address discovery, must not be taken from the interface where the >> prefix >>>>> hiding is configured" >>>> >>>> It is somewhat obvious that you wouldn't choose an address that is >> not >>>> advertised. However, I guess it wouldn't hurt to add that the RFC >> 5340, >>>> Section 4.4.3.9. Intra-Area-Prefix-LSAs, selection an IPv6 address >> to >>>> terminate virtual links will be modified to exclude interfaces with >> the >>>> prefix hiding configured. >>>> >>>> Thanks, >>>> Acee >>>> >>>> >>>> >>>>> >>>>> Please check. >>>>> >>>>> Thanks& Regards >>>>> Rajesh >>>>> >>>>> >>>>> >>>>> -----Original Message----- >>>>> From: [email protected] [mailto:[email protected]] On >> Behalf Of >>>> Acee >>>>> Lindem >>>>> Sent: Wednesday, February 08, 2012 10:37 PM >>>>> Cc: OSPF List >>>>> Subject: [OSPF] OSPF WG Last Call for "Hiding Transit-only > Networks >> in >>>> OSPF >>>>> " -<draft-ietf-ospf-prefix-hiding-02.txt> >>>>> >>>>> As I have heard no objections, I'm beginning the 2 week OSPF >> Working Group >>>>> last call for draft-ietf-ospf-prefix-hiding-02.txt. >>>>> Please review the draft and post your last call comments prior to >> 12:00 AM >>>>> PDT on February 23nd, 2012. >>>>> Here is a URL for your convenience: >>>>> >>>>> http://www.ietf.org/id/draft-ietf-ospf-prefix-hiding-02.txt >>>>> >>>>> Thanks, >>>>> Acee >>>>> >>>>> On Jan 26, 2012, at 11:19 AM, Acee Lindem wrote: >>>>> >>>>>> As WG co-chair, I have reviewed this document and believe it is >> ready for >>>>> OSPF WG last call. Any other opinions? >>>>>> There is at least one implementation. Here is a URL for you >> convenience: >>>>>> >>>>>> http://www.ietf.org/id/draft-ietf-ospf-prefix-hiding-01.txt >>>>>> >>>>>> There is an IPR disclosure on this draft: >>>>>> >>>>>> http://datatracker.ietf.org/ipr/1423/ >>>>>> >>>>>> I will start WG last call next week if I don't hear any >> objections. >>>>>> >>>>>> Thanks, >>>>>> Acee >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> OSPF mailing list >>>>>> [email protected] >>>>>> https://www.ietf.org/mailman/listinfo/ospf >>>>> >>>>> _______________________________________________ >>>>> OSPF mailing list >>>>> [email protected] >>>>> https://www.ietf.org/mailman/listinfo/ospf >>>>> >>>> >>>> _______________________________________________ >>>> OSPF mailing list >>>> [email protected] >>>> https://www.ietf.org/mailman/listinfo/ospf >>> >> >> _______________________________________________ >> OSPF mailing list >> [email protected] >> https://www.ietf.org/mailman/listinfo/ospf > _______________________________________________ > OSPF mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/ospf > _______________________________________________ OSPF mailing list [email protected] https://www.ietf.org/mailman/listinfo/ospf _______________________________________________ OSPF mailing list [email protected] https://www.ietf.org/mailman/listinfo/ospf
