However the use case does drive the design of the protocol, e.g. for SDN case, you might want separate every part of Ipsec (or other type of encryption) tunnel config so that controller has the flexibility to provision each of them differently on different node, and if you want to use BGP as the provision protocol, then that’s a quite lot a extensions to BGP; While in other non SDN case, user might want a solution with least changes to BGP and least information advertised by each node, so you might want group multiple parts of tunnel configuration into a single part (e.g. the color sub-TLV in my draft)
From: Xialiang (Frank, Network Standard & Patent Dept) <[email protected]> Sent: Tuesday, April 2, 2019 2:21 AM To: Robert Raszuk <[email protected]> Cc: Hu, Jun (Nokia - US/Mountain View) <[email protected]>; Fernando Pereñíguez García <[email protected]>; Linda Dunbar <[email protected]>; Roman Danyliw <[email protected]>; idr wg <[email protected]>; [email protected]; [email protected]; [email protected]; Gabriel López Millán <[email protected]>; Yoav Nir <[email protected]>; Alvaro Retana <[email protected]>; [email protected] WG <[email protected]>; Benjamin Kaduk <[email protected]>; Rafa Marin Lopez <[email protected]>; Paul Wouters <[email protected]> Subject: 答复: [I2nsf] 答复: [IPsec] using BGP signaling to achieve IPsec Tunnel configuration (draft-hujun-idr-bgp-ipsec): potential conflict with the I2NSF's Controller facilitated IPsec configuration Hi Robert, Thanks for further clarification, it helps for me. Again, I think the key point is not the original use case, but the function gaps each draft can fill in. B.R. Frank 发件人: Robert Raszuk [mailto:[email protected]] 发送时间: 2019年4月2日 16:32 收件人: Xialiang (Frank, Network Standard & Patent Dept) <[email protected]<mailto:[email protected]>> 抄送: Hu, Jun (Nokia - US/Mountain View) <[email protected]<mailto:[email protected]>>; Fernando Pereñíguez García <[email protected]<mailto:[email protected]>>; Linda Dunbar <[email protected]<mailto:[email protected]>>; Roman Danyliw <[email protected]<mailto:[email protected]>>; idr wg <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]>; Gabriel López Millán <[email protected]<mailto:[email protected]>>; Yoav Nir <[email protected]<mailto:[email protected]>>; Alvaro Retana <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> WG <[email protected]<mailto:[email protected]>>; Benjamin Kaduk <[email protected]<mailto:[email protected]>>; Rafa Marin Lopez <[email protected]<mailto:[email protected]>>; Paul Wouters <[email protected]<mailto:[email protected]>> 主题: Re: [I2nsf] 答复: [IPsec] using BGP signaling to achieve IPsec Tunnel configuration (draft-hujun-idr-bgp-ipsec): potential conflict with the I2NSF's Controller facilitated IPsec configuration Hi Frank, This draft does not talk about distributing any security related parameters. Maybe the name is a bit confusing as for some it means to be IPSec related. We have discussed the draft in Prague and agreed to also extend it with other types of secure encap. I have not discussed it with other authors but perhaps a much proper name and clearly less controversial would be: draft-hujun-idr-encrypted-transport-autodiscovery or draft-hujun-idr-eta (for short) or something along those lines to reflect what this work is really about and how it differs from other proposals. Thx, R. On Tue, Apr 2, 2019 at 3:52 AM Xialiang (Frank, Network Standard & Patent Dept) <[email protected]<mailto:[email protected]>> wrote: Hi Jun, My personal view is no matter which use cases (SDN-based or BGP-based) you are for, the basic goal is to configure/distribute the IPSec parameters between the associated peers, for next step IKEv2 session negotiation. That is why all these related drafts should be aligned in certain way. B.R. Frank 发件人: I2nsf [mailto:[email protected]<mailto:[email protected]>] 代表 Hu, Jun (Nokia - US/Mountain View) 发送时间: 2019年4月2日 6:22 收件人: Fernando Pereñíguez García <[email protected]<mailto:[email protected]>>; Linda Dunbar <[email protected]<mailto:[email protected]>> 抄送: Roman Danyliw <[email protected]<mailto:[email protected]>>; idr wg <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]>; Gabriel López Millán <[email protected]<mailto:[email protected]>>; Yoav Nir <[email protected]<mailto:[email protected]>>; Alvaro Retana <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> WG <[email protected]<mailto:[email protected]>>; Benjamin Kaduk <[email protected]<mailto:[email protected]>>; Rafa Marin Lopez <[email protected]<mailto:[email protected]>>; Paul Wouters <[email protected]<mailto:[email protected]>> 主题: Re: [I2nsf] [IPsec] using BGP signaling to achieve IPsec Tunnel configuration (draft-hujun-idr-bgp-ipsec): potential conflict with the I2NSF's Controller facilitated IPsec configuration Again, Linda, as discussed with you multiple times, my draft is really about extending current draft-ietf-idr-tunnel-encaps<https://datatracker.ietf.org/doc/draft-ietf-idr-tunnel-encaps/> to cover IPsec tunnel and other encryption tunnel like DTLS in next revsion (based on the feedback I got from Prague); My draft is not intended to address SDN for IPsec use case and it does not require a central controller, and there are use cases where a central controller is not needed or can’t be used, my draft is intended for those cases; So I really don’t see any conflict here From: IPsec <[email protected]<mailto:[email protected]>> On Behalf Of Fernando Pere?íguez García Sent: Monday, April 1, 2019 3:05 PM To: Linda Dunbar <[email protected]<mailto:[email protected]>> Cc: Roman Danyliw <[email protected]<mailto:[email protected]>>; idr wg <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]>; Gabriel López Millán <[email protected]<mailto:[email protected]>>; Yoav Nir <[email protected]<mailto:[email protected]>>; Alvaro Retana <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> WG <[email protected]<mailto:[email protected]>>; Benjamin Kaduk <[email protected]<mailto:[email protected]>>; Rafa Marin Lopez <[email protected]<mailto:[email protected]>>; Paul Wouters <[email protected]<mailto:[email protected]>> Subject: Re: [IPsec] using BGP signaling to achieve IPsec Tunnel configuration (draft-hujun-idr-bgp-ipsec): potential conflict with the I2NSF's Controller facilitated IPsec configuration Hi Linda, We have revised draft-hujun-idr-bgp-ipsec and, to the best of our understanding, we do not see any conflict with our draft being discussed in I2NSF. The IPsec attributes configured through BGP are only the peer’s tunnel address and local/remote subnet prefixes (that are used for the traffic selectors). The rest of the IPsec configuration (AH/ESP, cryptographic algorithms, keys, etc.) are obtained via a “color mapping”, which is something not covered by the draft since it assumes routers are somehow pre-provisioned with this information. Thus, we do not see this draft is also facing the task of formalizing the complete configuration of an IPsec device. We appreciate any clarification in case we are wrong. Best regards, Fernando.. El jue., 28 mar. 2019 a las 16:01, Linda Dunbar (<[email protected]<mailto:[email protected]>>) escribió: Just to reiterate the concerns and issues I raised during IDR Thurs session discussion on using BGP signaling to achieve IPsec Tunnel configuration (draft-hujun-idr-bgp-ipsec). Copy I2NSF WG because there is similar discussion for over a year. Copy IPsecme WG as the group has many experts on the IPsec configuration. 1. I2NSF WG has an on-going discussion on Controller facilitated IPsec configuration which has been discussed for over a year. Even though the I2NSF’s IPsec Configuration is between Controller and devices, whereas the BGP signaling IPsec Configuration proposed by draft-hujun-idr-bgp-ipsec is between peers, the configuration parameters to the devices are for the same purpose, therefore, should be aligned to avoid future conflicts to the industry. 2. When using IPsec Tunnel between two peers, usually they are separated by untrusted domain. If Router “A” is allowed to gets the IPsec tunnel configurations from peers across untrusted domain (instead of the today’s practice of from administrators), then many issues come up, for example: How can a router “A” trust the Traffic Selection policy from a remote peer B? If the router “A” already has its Traffic Selection policy configured for a specific IPsec tunnel, but different from the Traffic Selection policy from remote peer B, which policy should Route A enforce for the IPsec Tunnel? If the router “A” doesn’t have Traffic Selection policy specified, there are two remote nodes B & C signaling the “A” with different Traffic Selection policy, what should A do? 3. RFC5566 only specifies a simple indication of IPsec Encap, but doesn’t do any of the IPsec configuration portion. As indicated by BESS WG chair, there are multiple drafts addressing IPsec in BESS, IDR, and WGs in Security Area, involved Chairs and ADs may need to agree where is the home for continuing the discussion to avoid future conflicts. Cheers, Linda Dunbar _______________________________________________ IPsec mailing list [email protected]<mailto:[email protected]> https://www.ietf.org/mailman/listinfo/ipsec -- ---------------------------------------------------------------------------------------------------- Fernando Pereñíguez García, PhD Department of Sciences and Informatics University Defense Center, (CUD), Spanish Air Force Academy, MDE-UPCT C/ Coronel Lopez Peña, s/n, 30720, San Javier, Murcia - SPAIN Tel: +34 968 189 946 Fax: +34 968 189 970 ------------------------------------------------------------------------------------------------------ _______________________________________________ I2nsf mailing list [email protected]<mailto:[email protected]> https://www.ietf.org/mailman/listinfo/i2nsf
_______________________________________________ I2nsf mailing list [email protected] https://www.ietf.org/mailman/listinfo/i2nsf
