I think useful to breakdown some aspects about SR policies: - related to candidate paths instantiation/reporting/update –via PCEP, BGP, et. al – and related extensions.. IMO, there’s some work to be done there - related path computation/formulation/validation – it’s clear SR paths are not 1:1 with RSVP hops, and SIDs are not necessarily always forwarding instructions— It is likely however a single stateful controller will have to deal with both SR policies and RSVP-TE LSPs on a single TED network to do meaningful path computation/placement. In here, I think staying in sync with TEAS would be beneficial. - specific to srv6 dataplane and nsh - specific to color based service steering
There's enough to distinguish SR policies from RSVP LSPs and it'd be good to treat/model them differently to reduce confusion and complexity. Regards, Tarek On 2018-03-22, 10:33 AM, "spring on behalf of Lizhenbin" <[email protected] on behalf of [email protected]> wrote: The SR-TE policy work has been done for a long time in SRPING WG. In addition though the name is a little misleading, I would like to remind you of the work of the draft-ietf-idr-segment-routing-te-policy-02 which has much relation with the draft-filsfils-spring-segment-routing-policy-05 in SPRING WG from which we can see the work of SR-TE policy is much beyond the scope of TEAS WG. Based on the observation I think the work should be incorporated in the recharter of SPRING WG. ________________________________________ 发件人: spring [[email protected]] 代表 stefano previdi [[email protected]] 发送时间: 2018年3月22日 16:50 收件人: Dongjie (Jimmy) 抄送: [email protected]; [email protected]; Shah, Himanshu; Alvaro Retana; Bruno Decraene; Bernier, Daniel; Jeff Tantsura; Voyer, Daniel 主题: Re: [spring] [**EXTERNAL**] Re: SPRING - rechartering discussion SPRINGers, > On Mar 19, 2018, at 3:23 PM, Dongjie (Jimmy) <[email protected]> wrote: > > Hi all, > > I totally agree with Mach, Jeff and others that there is work to be done in OAM as there are more requirements to use SR for both existing and emerging applications. > > SR-TE is another important area. The current SR-TE mainly focuses on steering traffic to particular SR paths, while TE can have a broader scope than that, for example, how to do resource partitioning (reservation) with SR needs to be discussed. I agree with the above. This is yet another aspect of TE that needs to be addressed. Also, the v6 side of SR will also require more work, e.g., in the definition of use cases that clearly benefit from extension header insertion so that (minor) changes can be proposed to the v6 specification allowing these use cases to be implemented and operated safely. s. > Actually this is already mentioned in the current charter: > > o Some types of network virtualization, including multi- > topology networks and the partitioning of network > resources for VPNs > > I’d agree with Dan that SR-TE is different from RSVP-TE, while as Himanshu said, it could be beneficial to leverage the TE expertise from TEAS. > > Best regards, > Jie > > From: spring [mailto:[email protected]] On Behalf Of Voyer, Daniel > Sent: Monday, March 19, 2018 11:42 AM > To: Shah, Himanshu; Jeff Tantsura; Bernier, Daniel; [email protected]; [email protected] > Cc: Alvaro Retana (aretana); [email protected] > Subject: Re: [spring] [**EXTERNAL**] Re: SPRING - rechartering discussion > > [DV] see inlines > > From: spring <[email protected]> on behalf of "Shah, Himanshu" <[email protected]> > Date: Sunday, March 18, 2018 at 9:23 PM > To: Jeff Tantsura <[email protected]>, Daniel Bernier <[email protected]>, Bruno Decraene <[email protected]>, "[email protected]" <[email protected]> > Cc: "Alvaro Retana (aretana)" <[email protected]>, "[email protected]" <[email protected]> > Subject: Re: [spring] [**EXTERNAL**] Re: SPRING - rechartering discussion > > Agree with Jeff, without harping on all the good reasons already stated for SPRING WG charter extensions, > I would think that it would be beneficial to leverage TE expertise from TEAS WG to > progress SR-TE there for a cohesive, uniform solution for all tunneling schemes. > > [DV] 1- SRTE is NOT a tunnel. Labels are signals straight in the IGP, as known. This is why the word “policy” was introduce with SRTE – “SRTE Policy”. > [DV] 2- According to TEAS WG charter - https://datatracker.ietf.org/wg/teas/about/: > 1. Definition of additional abstract service, link, and path > properties such as jitter, delay, and diversity. Extensions > to IGPs to advertise these properties, and extensions to > RSVP-TE to request and to accumulate these properties. > > [DV] 3- also notice in the SPRING Charter - https://datatracker.ietf.org/wg/spring/about/: > o Some types of network virtualization, including multi- > topology networks and the partitioning of network > resources for VPNs > o Network path and node protection such as fast re-route > o Network programmability > o New OAM techniques > o Simplification and reduction of network signalling > components > o Load balancing and traffic engineering > [DV] Hence I believe “SRTE policy” is a key component of the SR Architecture and should pursued as part as the Architecture definition milestone of the SPRING WG. > > Dan > > IMHO.. > > Thanks, > Himanshu > From: spring <[email protected]> on behalf of Jeff Tantsura <[email protected]> > Date: Sunday, March 18, 2018 at 3:26 PM > To: "Bernier, Daniel" <[email protected]>, "[email protected]" <[email protected]>, "[email protected]" <[email protected]> > Cc: Alvaro Retana <[email protected]>, "[email protected]" <[email protected]> > Subject: [**EXTERNAL**] Re: [spring] SPRING - rechartering discussion > > Hi, > > I'm not going to repeat all the valid reasons to continue mentioned beforehand. > There's definitely work to be done in architecture and O&M areas as well as co-ordination of various activities across IETF. > > Cheers, > Jeff > On 3/18/18, 13:23, "spring on behalf of Bernier, Daniel" <[email protected] on behalf of [email protected]> wrote: > > Hi, > > I echo the need to continue the SPRING work on service-chaining. There is a growing interest to have a mechanism that operates at the forwarding plane level using source routing as an alternative to a dedicated service overlay. This will surely generate other related work such as automated service discovery, inter-domain chaining policies, parallelism versus sequential chaining, various control-plane implementations, etc. > > Secondly, since there is a tight relation to SR chaining and TE policies, I believe there will is a lot of opportunities related to Path Awareness which is currently running in IRTF. Opportunities like, intent translation to SR policies, Policy requests or announcements between domains and host (probably app) level TE policy requests (e.g. how can an app receive a proper policy based on its requirements) ? > > My humble operator 0.02 cents. > > Daniel Bernier | Bell Canada > ________________________________________ > From: spring <[email protected]> on behalf of [email protected] <[email protected]> > Sent: Monday, March 5, 2018 11:59 AM > To: [email protected] > Cc: Alvaro Retana (aretana); [email protected] > Subject: [spring] SPRING - rechartering discussion > > Hello WG, > > now that nearly all the core documents are in the hands of IESG or beyond, we think it is time to (re)discuss rechartering. > We brought up that question few meetings ago and the feedback, at that time, was that the WG at least needs to be maintained to discuss the extensions following deployment feedback. > > But we need also identify technical directions. > > In order to initiate the discussion we are proposing some high level items but we'd like to make clear a few points before: > * these are only proposals; what might end-up as the next steps for SPRING will be what the WG is willing to work on (which includes having cycles for that). > * what the WG might be rechartered to do is not necessarily limited to that; so other proposals are welcome. > > So, we thought of the following: > > * general architectural work / extensions > there are still few items on our plate and we expect that some might need to be progressed, and we should maybe allow for others to come. > > * service chaining > last meeting there were proposals discussed in SPRING to realize some form of service chaining. any work in that space would require close coordination with SFC and maybe other WG. > > * yang > we are a bit behind here and there is definitely work to do. > > > So please comment on these and propose additional items. > > We'll likely have a dedicated slot in London but we'd like to progress before that. > > Thank you, > --Martin, Rob, Bruno > > > -----Original Message----- > > From: Martin Vigoureux [mailto:[email protected]] > > Sent: Wednesday, March 29, 2017 4:25 PM > > To: [email protected] > > Cc: [email protected]; Alvaro Retana (aretana) > > Subject: Next steps for SPRING? > > > > WG, > > > > in the session we have opened the discussion on the future of the WG, > > putting all options on the table (recharter/close/sleep). > > As a foreword, we still have few WG Documents that we need to -and will- > > push towards IESG (and a greater number that need to reach RFC status), > > but with those we'll have reached most if not all of our milestones, > > thus the question on what's next. > > > > So, we think we have heard during the session that closing wasn't > > desired and one reason for that is to have a home to share and discuss > > deployment considerations as the technology gets deployed. > > There are also a few individual documents knocking at the door, and some > > of them were presented during the session. > > > > To reach out to everyone, we are thus asking the question on the list. > > We would like to hear from you all what the working group should be > > focussing on. > > > > Note, the expectation is that future items should not be use-cases but > > rather be technology extensions/evolutions. > > > > Thank you > > > > Martin & Bruno > > _________________________________________________________________________________________________________________________ > > 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 > [email protected] > https://www.ietf.org/mailman/listinfo/spring > _______________________________________________ > spring mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/spring > > > > _______________________________________________ > spring mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/spring > > _______________________________________________ > spring mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/spring _______________________________________________ spring mailing list [email protected] https://www.ietf.org/mailman/listinfo/spring _______________________________________________ spring mailing list [email protected] https://www.ietf.org/mailman/listinfo/spring
