Re:[nvo3] BFD over VXLAN: Trapping BFD Control packet at VTEP

[<[e-mail]>]

Hi Joel,






I fully agree to your analysis.


One major difference between VxLAN and Geneve (or VxLAN-GPE) is that VxLAN doesn't support multi-protocol payload, and VxLAN only supports payload of Ethernet frame. Although VxLAN specification was developed outside NVO3 WG, I believe VxLAN may also fall within the NVO3 architecture, and we try to align "BFD for VxLAN" and "BFD for Geneve".







Best Regards,


Xiao Min










原始邮件



发件人：JoelM.Halpern <[e-mail]>
收件人：Anoop Ghanwani <[e-mail]>;肖敏10093570;
抄送人：[e-mail] <[e-mail]>;[e-mail] <[e-mail]>;rtg-bfd WG <[e-mail]>;
日 期 ：2019年10月09日 06:31
主 题 ：Re: [nvo3] BFD over VXLAN: Trapping BFD Control packet at VTEP




I will add that from the point of view of VxLAN 9which is the topic), I 
would expect the MPLS packet to arrive in an Ethernet frame, and for 
VxLAN to forward that Ethernet frame.  The VTEP would not seem to even 
need to be aware that the content is MPLS.

Yours,
Joel

On 10/8/2019 6:28 PM, Anoop Ghanwani wrote:
> Hi Xiao Min,
> 
> The picture doesn't have enough information to explain why they are in 
> the same VNI, and exactly how forwarding happens between the MPLS and 
> non-MPLS parts.
> 
> Anoop
> 
> On Tue, Oct 8, 2019 at 12:31 AM <[e-mail] 
> <mailto:[e-mail]>> wrote:
> 
>     Hi Anoop,
> 
> 
>     I don't know such a draft that describes MPLS over Geneve, but I
>     believe the following figure derived from figure 1 of RFC8014 would
>     help, in the following figure Tenant System1, Tenant System2, Tenant
>     System3 and Tenant System4 are assumed belonging to the same VNI, so
>     two BFD sessions for the same VNI need to be run between NVE1 and NVE2.
> 
>                                                  +--------+
>                                             +----| Tenant |
>                                           ( ' )  | System1|
>                  ................       ( MPLS ) +--------+
>                  .              .  +--+-+ ( _ )
>                  .              .--|NVE1|---+
>                  .              .  |    |
>                  .              .  +--+-+
>                  .              .     |
>                  .  L3 Overlay  .   ( ' )
>                  .    Network   . (Ethernet)
>                  .              .   ( _ )
>                  .              .     |
>                  ................    +--------+
>                     |                | Tenant |
>                   +----+             | System2|
>                   |NVE2|             +--------+
>                   |    |--------+
>                   +----+        |
>                     |           |
>                   ( ' )       ( ' )
>                 ( MPLS )    (Ethernet)
>                   ( _ )       ( _ )
>                     |           |
>                 +--------+  +--------+
>                 | Tenant |  | Tenant |
>                 | System3|  | System4|
>                 +--------+  +--------+
> 
> 
>     Best Regards,
> 
>     Xiao Min
> 
>     原始邮件
>     *发件人：*AnoopGhanwani <[e-mail]
>     <mailto:[e-mail]>>
>     *收件人：*肖敏10093570;
>     *抄送人：*Greg Mirsky <[e-mail]
>     <mailto:[e-mail]>>;[e-mail]
>     <mailto:didutt@gmail..com> <[e-mail]
>     <mailto:[e-mail]>>;[e-mail]
>     <mailto:[e-mail]>
>     <[e-mail]
>     <mailto:[e-mail]>>;[e-mail]
>     <mailto:[e-mail]> <[e-mail]
>     <mailto:[e-mail]>>;[e-mail]
>     <mailto:[e-mail]> <[e-mail]
>     <mailto:[e-mail]>>;rtg-bfd WG <[e-mail]
>     <mailto:[e-mail]>>;Joel M. Halpern <[e-mail]
>     <mailto:[e-mail]>>;[e-mail]
>     <mailto:[e-mail]> <tsridhar@vmware..com
>     <mailto:[e-mail]>>;
>     *日 期 ：*2019年10月08日 12:15
>     *主 题 ：**Re: [nvo3] BFD over VXLAN: Trapping BFD Control packet at
>     VTEP*
>     Hi Xiao Min,
>     Is there a draft that describes MPLS over Geneve?  It sounds like
>     the NVE is an MPLS router in this case and if you're using the same
>     VNI as you switch MPLS, then it's a one-armed router.  That doesn't
>     change how BFD needs to be run between NVEs.
> 
>     Anoop
> 
>     On Mon, Oct 7, 2019 at 7:28 PM <[e-mail]
>     <mailto:[e-mail]>> wrote:
> 
>         Hi Anoop,
> 
> 
>         Sorry for the late response, I just come back from vacation.
> 
>         The use case is that the network between the VM and the NVE is
>         an MPLS network, within which the packet is forwarded basing on
>         MPLS label, but not Ethernet MAC address and/or 802.1Q VLAN.
>         When two such kind of MPLS networks need to communicate with
>         each other, through a Geneve tunnel, the encap I illustrated
>         would be used.
> 
> 
>         Best Regards,
> 
>         Xiao Min
> 
>         原始邮件
>         *发件人：*AnoopGhanwani <[e-mail]
>         <mailto:[e-mail]>>
>         *收件人：*肖敏10093570;
>         *抄送人：*Greg Mirsky <[e-mail]
>         <mailto:[e-mail]>>;[e-mail]
>         <mailto:[e-mail]> <[e-mail]
>         <mailto:[e-mail]>>;[e-mail]
>         <mailto:[e-mail]>
>         <[e-mail]
>         <mailto:[e-mail]>>;[e-mail]
>         <mailto:[e-mail]> <[e-mail]
>         <mailto:[e-mail]>>;[e-mail]
>         <mailto:[e-mail]>
>         <[e-mail]
>         <mailto:[e-mail]>>;rtg-bfd WG
>         <[e-mail] <mailto:[e-mail]>>;Joel M. Halpern
>         <[e-mail]
>         <mailto:[e-mail]>>;[e-mail]
>         <mailto:[e-mail]> <[e-mail]
>         <mailto:[e-mail]>>;
>         *日 期 ：*2019年09月28日 05:36
>         *主 题 ：**Re: [nvo3] BFD over VXLAN: Trapping BFD Control
>         packet at VTEP*
>         Hi Xiao Min,
>         Thanks for the details about the encap but the use case is not
>         clear.  It might help if you explain why its necessary to map a
>         physical Ethernet port and/or 802.1Q VLAN to the same VNI as an
>         MPLS packet without an L2 header.
> 
>         Thanks,
>         Anoop
> 
>         On Thu, Sep 26, 2019 at 7:50 PM <[e-mail]
>         <mailto:[e-mail]>> wrote:
> 
>             Hi Anoop,
> 
> 
>             Due to the fact that a variety of Tunnels could be used
>             under the NVO3 architecture, as an example, below figure
>             illustrates the format of MPLS packet over Geneve Tunnel.
> 
>                  0                   1                   2                   3
>                  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
>                 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>                 |                                                               |
>                 ~                      Outer Ethernet Header                    ~
>                 |                                                               |
>                 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>                 |                                                               |
>                 ~                        Outer IPvX Header                      ~
>                 |                                                               |
>                 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>                 |                                                               |
>                 ~                        Outer UDP Header                       ~
>                 |                                                               |
>                 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>                 |                                                               |
>                 ~                          Geneve Header                        ~
>                 |                                                               |
>                 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<-+
>                 |                                                               |  |
>                 ~                         MPLS Label Stack                      ~  M
>                 |                                                               |  P
>                 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+  L
>                 |                                                               |  S
>                 |                                                               |
>                 ~                             Payload                           ~  P
>                 |                                                               |  K
>                 |                                                               |  T
>                 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<-+
>                 |                               FCS                             |
>                 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> 
> 
>             Note that in NVO3 working group Greg and I have submitted an
>             individual draft draft-xiao-nvo3-bfd-geneve, which is used
>             to address BFD over Geneve.
> 
>             The intention is to make the two drafts draft-ietf-bfd-vxlan
>             and draft-xiao-nvo3-bfd-geneve aligned, that is to say, we
>             try to define the identical mechanism for the common part of
>             BFD over VxLAN Tunnel and BFD over Geneve Tunnel. For the
>             common part, draft-xiao-nvo3-bfd-geneve would reference to
>             draft-ietf-bfd-vxlan, and for the other part specific to
>             Geneve, we'll define the specific mechanism in
>             draft-xiao-nvo3-bfd-geneve.
> 
> 
>             Hope that clarifies.
> 
> 
>             Best Regards,
> 
>             Xiao Min
> 
>             原始邮件
>             *发件人：*AnoopGhanwani <[e-mail]
>             <mailto:[e-mail]>>
>             *收件人：*肖敏10093570;
>             *抄送人：*Greg Mirsky <[e-mail]
>             <mailto:[e-mail]>>;[e-mail]
>             <mailto:[e-mail]> <[e-mail]
>             <mailto:[e-mail]>>;[e-mail]
>             <mailto:[e-mail]>
>             <[e-mail]
>             <mailto:[e-mail]>>;[e-mail]
>             <mailto:[e-mail]> <[e-mail]
>             <mailto:[e-mail]>>;[e-mail]
>             <mailto:[e-mail]>
>             <[e-mail]
>             <mailto:[e-mail]>>;rtg-bfd WG
>             <[e-mail] <mailto:[e-mail]>>;Joel M. Halpern
>             <[e-mail]
>             <mailto:[e-mail]>>;[e-mail]
>             <mailto:[e-mail]> <[e-mail]
>             <mailto:[e-mail]>>;[e-mail]
>             <mailto:[e-mail]> <[e-mail]
>             <mailto:[e-mail]>>;
>             *日 期 ：*2019年09月26日 23:16
>             *主 题 ：**Re: [nvo3] BFD over VXLAN: Trapping BFD Control
>             packet at VTEP*
>             Hi Xiao Min,
>             I think we would need more detail around the use case
>             below.  What does the MPLS packet over Tunnel look like?
> 
>             Thanks,
>             Anoop
> 
>             On Wed, Sep 25, 2019 at 11:37 PM <[e-mail]
>             <mailto:[e-mail]>> wrote:
> 
>                 Hi Anoop,
> 
> 
>                 Thanks for your comments.
> 
>                 Considering a scenario where TS1 has an MPLS access
>                 (i.e. MPLS-Packet over Tunnel between NVEs) to VNI1, TS3
>                 has an Ethernet access (i.e. MAC-Frame over Tunnel
>                 between NVEs) to VNI1, then how can TS1 and TS3 share
>                 one VAP?
> 
> 
>                 Best Regards,
> 
>                 Xiao Min
> 
>                 原始邮件
>                 *发件人：*AnoopGhanwani <[e-mail]
>                 <mailto:[e-mail]>>
>                 *收件人：*肖敏10093570;
>                 *抄送人：*Greg Mirsky <[e-mail]
>                 <mailto:[e-mail]>>;[e-mail]
>                 <mailto:[e-mail]> <[e-mail]
>                 <mailto:[e-mail]>>;[e-mail]
>                 <mailto:[e-mail]>
>                 <[e-mail]
>                 <mailto:[e-mail]>>;[e-mail]
>                 <mailto:[e-mail]> <[e-mail]
>                 <mailto:[e-mail]>>;[e-mail]
>                 <mailto:[e-mail]>
>                 <[e-mail]
>                 <mailto:[e-mail]>>;rtg-bfd WG
>                 <[e-mail] <mailto:[e-mail]>>;Joel M.
>                 Halpern <[e-mail]
>                 <mailto:[e-mail]>>;[e-mail]
>                 <mailto:[e-mail]> <[e-mail]
>                 <mailto:[e-mail]>>;[e-mail]
>                 <mailto:[e-mail]> <[e-mail]
>                 <mailto:[e-mail]>>;
>                 *日 期 ：*2019年09月26日 08:36
>                 *主 题 ：**Re: [nvo3] BFD over VXLAN: Trapping BFD
>                 Control packet at VTEP*
>                 _______________________________________________
>                 nvo3 mailing list
>                 [e-mail] <mailto:[e-mail]>
>                 https://www.ietf.org/mailman/listinfo/nvo3
> 
>                  >>>
>                 Some people may argue that all Tenant Systems connecting
>                 to the same Virtual Network MUST share one VAP, if
>                 that's true, then VAP1 and VAP3 should merge into one
>                 VAP and my explanation doesn't work. Copying to NVO3 WG
>                 to involve more experts, hope for your clarifications
>                 and comments.
>                  >>>
> 
>                 I would be one of those that would argue that they MUST
>                 share on VAP if they connect to the same Virtual
>                 Network.  IMO, the NVO3 arch doc should have been
>                 clearer about this.
> 
>                 Thanks,
>                 Anoop
> 
>                 On Tue, Sep 24, 2019 at 7:40 PM <[e-mail]
>                 <mailto:[e-mail]>> wrote:
> 
>                     Hi Santosh,
> 
> 
>                     With regard to the question whether we should allow
>                     multiple BFD sessions for the same VNI or not, IMHO
>                     we should allow it, more explanation as follows...
> 
>                     Below is a figure derived from figure 2 of RFC8014
>                     (An Architecture for Data-Center Network
>                     Virtualization over Layer 3 (NVO3)).
> 
>                                          |         Data Center Network (IP)        |
>                                          |                                         |
>                                          +-----------------------------------------+
>                                               |                           |
>                                               |       Tunnel Overlay      |
>                                  +------------+---------+       +---------+------------+
>                                  | +----------+-------+ |       | +-------+----------+ |
>                                  | |  Overlay Module  | |       | |  Overlay Module  | |
>                                  | +---------+--------+ |       | +---------+--------+ |
>                                  |           |          |       |           |          |
>                           NVE1   |           |          |       |           |          | NVE2
>                                  |  +--------+-------+  |       |  +--------+-------+  |
>                                  |  |VNI1 VNI2  VNI1 |  |       |  | VNI1 VNI2 VNI1 |  |
>                                  |  +-+-----+----+---+  |       |  +-+-----+-----+--+  |
>                                  |VAP1| VAP2|    | VAP3 |       |VAP1| VAP2|     | VAP3|
>                                  +----+-----+----+------+       +----+-----+-----+-----+
>                                       |     |    |                   |     |     |
>                                       |     |    |                   |     |     |
>                                       |     |    |                   |     |     |
>                                -------+-----+----+-------------------+-----+-----+-------
>                                       |     |    |     Tenant        |     |     |
>                                  TSI1 | TSI2|    | TSI3          TSI1| TSI2|     |TSI3
>                                      +---+ +---+ +---+             +---+ +---+   +---+
>                                      |TS1| |TS2| |TS3|             |TS4| |TS5|   |TS6|
>                                      +---+ +---+ +---+             +---+ +---+   +---+
> 
>                     To my understanding, the BFD sessions between NVE1
>                     and NVE2 are actually initiated and terminated at
>                     VAP of NVE.
> 
>                     If the network operator want to set up one BFD
>                     session between VAP1 of NVE1 and VAP1of NVE2, at the
>                     same time another BFD session between VAP3 of NVE1
>                     and VAP3 of NVE2, although the two BFD sessions are
>                     for the same VNI1, I believe it's reasonable, so
>                     that's why I think we should allow it.
> 
> 
>                     Of course, in RFC8014 it also says:
> 
>                     "Note that two different Tenant Systems (and TSIs) attached to a common NVE can share a VAP (e.g., TS1 and TS2 in Figure 2) so long as they connect to the same Virtual Network."
> 
>                     Some people may argue that all Tenant Systems
>                     connecting to the same Virtual Network MUST share
>                     one VAP, if that's true, then VAP1 and VAP3 should
>                     merge into one VAP and my explanation doesn't work.
>                     Copying to NVO3 WG to involve more experts, hope for
>                     your clarifications and comments.
> 
> 
>                     Best Regards,
> 
>                     Xiao Min
> 
> 
> 
> 
> 
> 
> _______________________________________________
> nvo3 mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/nvo3
>