Lucy,

Before posting the WG version, see my comments below.

Thanks,
Marc

________________________________
From: [email protected] [mailto:[email protected]] On Behalf Of Lucy 
yong
Sent: Thursday, November 29, 2012 12:55 AM
To: Bocci, Matthew (Matthew); [email protected]
Cc: [email protected]
Subject: Re: [nvo3] Poll for adoption and IPR check: 
draft-bl-nvo3-dataplane-requirements-03.txt

I support the action, but have some comment and suggestions to this version.


*         There is no single word  about tenant system mobility. Suggest adding 
a section within section 3.9  to address tenant system mobility. For example,  
it should point out that, although the data plane learning can be used to learn 
source MAC addresses as described in section 3.2.1, the mechanism may not meet 
the tenant system mobility requirements. In this case,  when a tenant system 
moves, some frames with the destination MAC as the tenant system are forwarded 
to the old associated NVE.  Other VM mobility and reachbility can be described 
as well.

Will do so in the next revision.

*         Suggest to add a sub-section in section 3.3.2 to describe tunneling 
encapsulation. One requirement is that all the NVEs that are the member of a 
VNI MUST support a  tunneling encapsulation.  One option requirement is that 
all the NVEs that are the member of a VNI MAY support more than one tunneling 
encapsulation. In this case, an ingress NVE MUST use the egress NVE supported 
encapsulation to encapsulate the frame/packets before sending to the remote NVE.

This section specified alrady that IP tunneling is a MUST and that MPLS is a 
MAY. It is not a requirement to support multiple simultaneous encapsulations. 
If it is the case, it is up to solutions draft to address how  this would be 
negociated.

*         Section 3.6 Hierarchical NVE.  The text is very weak and without any 
requirement on it. It is not clear which function that hub NVE need to perform 
and which spoke NVE need to perform. For example, should hub NVE perform the 
encapsulation or should spoke NVE do it?  Furthermore, the last sentence does 
not apply to here at all.  Suggest either removing the section or stating out 
clearly what are the requirements for Hierarchical NVE.

Will beef up this section in a next revision.

*         Section 3.2.1 states about using data plane learning and point out 
that multiple overlay tunnels may exist.  But it does not state out any 
requirement to load balance over multiple overlay tunnels. For example, if 
there are two overlay tunnels between two NVEs, one NVE learns MAC associated 
with one overlay tunnel, can the NVE use another overlay tunnel to send the 
fame with the MAC to remote NVE? Suggest adding related requirement in the 
section.

Section 3.4.2 deals extensively with load balancing.

*         Section 3.2.2 starts with "L3 VNIs" and Section 3.2.1 starts with "an 
L2 VNI". is this intentional to illustrate the difference  between L2 overlay 
and L3 overlay or a typo?



Each section describes L2 and L3 VNIs respectively. Not sure what you mean 
here...

*         Section 3.3, suggest use tenant payload. There is no definition of 
customer in this document.

*         In section 1.2, some team already defined in the framework document. 
Should not restate here.

I would like to see these to be addressed in next version.

Cheers,
Lucy






From: [email protected] [mailto:[email protected]] On Behalf Of Bocci, 
Matthew (Matthew)
Sent: Wednesday, November 28, 2012 10:16 AM
To: [email protected]
Subject: [nvo3] Poll for adoption and IPR check: 
draft-bl-nvo3-dataplane-requirements-03.txt

This email begins a two week poll to help the chairs judge if there is 
consensus  to adopt draft-bl-nvo3-dataplane-requirements-03.txt as an NVO3 
working group draft.

Please respond to this email on the list with 'support' or 'do not support'.

Please also send any comments on the draft to the NVO3 list.

Coincidentally, we are also polling for knowledge of any IPR that applies to 
this draft, to ensure that IPR has been disclosed in compliance with IETF IPR 
rules (see RFCs 3979, 4879, 3669 and 5378 for more details).

If you are listed as a document author or contributor, please respond to this 
email whether or not you are aware of any relevant IPR. The draft will not be 
adopted until a response has been received from each author and contributor.

If you are on the NVO3 WG email list but are not listed as an author or 
contributor, then please explicitly respond only if you are aware of any IPR 
that has not yet been disclosed in conformance with IETF rules.

This poll closes on Wednesday 12th December 2012.

Regards

Matthew and Benson
_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3

Reply via email to