Hi Tarek, More inline
-----Original Message----- From: Tarek Saad (tsaad) [mailto:ts...@cisco.com] Sent: Tuesday, July 21, 2015 23:45 To: LITKOWSKI Stephane SCE/IBNF; cbow...@juniper.net Cc: spring@ietf.org Subject: Re: New Version Notification for draft-bowers-spring-adv-per-algorithm-label-blocks-01.txt Hi, On 2015-07-21, 5:04 PM, "stephane.litkow...@orange.com" <stephane.litkow...@orange.com> wrote: >Hi Tarek, > >I think per-prefix granularity is still possible. >Advertising a SRGB for a particular algorithm does not mean that all >the prefixes advertised will be advertised for that particular algorithm. > >I mean considering you have a network with node A ,B, C, D, each one >advertising a loopback. >You want to run algo 0 between all routers, so each router advertises a >SRGB for algo 0, and loopback address with SID is advertised with Algo >0 also. >Now you want to run algo 2 but only A and B requires communication >between each other using algo 2. As C and D may be on the path from A >to B, all routers requires supports of algo 2 and a new SRGB is >provisioned [TS]: Yes, if a prefix is to be reachable via alg 2, it¹s understandable every other node a will allocate/program label for alg2. >on all routers corresponding to algo 2. But in term of SID >advertisement, you may configure only A and B to advertise Algo2 for >Loopback address (Algo is encoded in PrefixSID), so there would be only >2 new MPLS [TS]: quote from draft: "In Option 2 each node advertises a single node index and a unique label block for each algorithm. " 1) it is not clear (from proposal) if alg support is to be explicitly enabled per prefix, and how? If it is to be explicitly enabled per alg2 for a prefix/loopback, wouldn't the overhead be comparable to configuring an index per algorithm? [SLI] IMO, it's implementation dependant.Regarding the overhead, IMO, as a service provider, the overhead is different. The overhead in option#1 is in managing the unicity of the SID and so a strong requirement of interaction with the OSS (for example) to retrieve a unique value => so automation is harder. Here it's just another CLI command to add which is generic for all routers, or at least no requirement to interact with an external system to retrieve informations. 2) also, not clear how a node announces support for multiple algs for same SID index value. Is multiple prefix-SID sub-TLVs announced with same SID index and for every algorithm it supports? If so, wondering how would a node withdraw support for a specific alg? [SLI] IMO, multiple prefix-SID sub-TLVs are attached to the prefix, one for each algorithm (algorithm field changes). If you want to withdraw, remove a subTLV. Regards, Tarek >forwarding entries instead of 4. > >Does it make sense ? > >Stephane > > >-----Original Message----- >From: spring [mailto:spring-boun...@ietf.org] On Behalf Of Tarek Saad >(tsaad) >Sent: Tuesday, July 21, 2015 15:05 >To: cbow...@juniper.net >Cc: spring@ietf.org >Subject: Re: [spring] New Version Notification for >draft-bowers-spring-adv-per-algorithm-label-blocks-01.txt > >Hi Chris, > >A possible limitation of advertizing per-alg SRGB is that a node will >always assign/maintain as many SRGB-alg labels (on possibly all nodes) >for a prefix without being able to control this on per-prefix if needed. >Are there are any considerations for this in your proposal? > >Regards, >Tarek > >-----Original Message----- >From: internet-dra...@ietf.org [mailto:internet-dra...@ietf.org] >Sent: Wednesday, June 24, 2015 6:03 PM >To: Hannes Gredler; Pushpasis Sarkar; Chris Bowers; Chris Bowers; >Pushpasis Sarkar; Hannes Gredler >Subject: New Version Notification for >draft-bowers-spring-adv-per-algorithm-label-blocks-01.txt > > >A new version of I-D, >draft-bowers-spring-adv-per-algorithm-label-blocks-01.txt >has been successfully submitted by Chris Bowers and posted to the IETF >repository. > >Name: draft-bowers-spring-adv-per-algorithm-label-blocks >Revision: 01 >Title: Advertising Per-Algorithm Label Blocks >Document date: 2015-06-24 >Group: Individual Submission >Pages: 7 >URL: >https://www.ietf.org/internet-drafts/draft-bowers-spring-adv-per-algori >thm >- >label-blocks-01.txt >Status: >https://datatracker.ietf.org/doc/draft-bowers-spring-adv-per-algorithm- >lab >e >l-blocks/ >Htmlized: >https://tools.ietf.org/html/draft-bowers-spring-adv-per-algorithm-label >-bl >o >cks-01 >Diff: >https://www.ietf.org/rfcdiff?url2=draft-bowers-spring-adv-per-algorithm >-la >b >el-blocks-01 > >Abstract: > Segment routing uses globally-known labels to accomplish destination- > based forwarding along shortest paths computed using Dijkstra's > algorithm with IGP metrics. This draft discusses how to use segment > routing to accomplish destination-based forwarding along paths > computed using other algorithms and metrics. In particular, the > draft contrasts two different options for associating labels with > different algorithms for computing forwarding next-hops. > > > > > >Please note that it may take a couple of minutes from the time of >submission until the htmlized version and diff are available at >tools.ietf.org. > >The IETF Secretariat > >_______________________________________________ >spring mailing list >spring@ietf.org >https://www.ietf.org/mailman/listinfo/spring > >_______________________________________________________________________ >___ _______________________________________________ > >Ce message et ses pieces jointes peuvent contenir des informations >confidentielles ou privilegiees et ne doivent donc pas etre diffuses, >exploites ou copies sans autorisation. Si vous avez recu ce message par >erreur, veuillez le signaler a l'expediteur et le detruire ainsi que >les pieces jointes. Les messages electroniques etant susceptibles >d'alteration, Orange decline toute responsabilite si ce message a ete >altere, deforme ou falsifie. Merci. > >This message and its attachments may contain confidential or privileged >information that may be protected by law; they should not be >distributed, used or copied without authorisation. >If you have received this email in error, please notify the sender and >delete this message and its attachments. >As emails may be altered, Orange is not liable for messages that have >been modified, changed or falsified. >Thank you. _________________________________________________________________________________________________________________________ Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration, Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci. This message and its attachments may contain confidential or privileged information that may be protected by law; they should not be distributed, used or copied without authorisation. If you have received this email in error, please notify the sender and delete this message and its attachments. As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified. Thank you. _______________________________________________ spring mailing list spring@ietf.org https://www.ietf.org/mailman/listinfo/spring
