[mpls] Re: AD review: draft-ietf-tsvwg-ecn-mpls

[Lars Eggert <[e-mail]>]

Hi, Bruce,

On 2007-9-12, at 17:04, ext Bruce Davie wrote:
>  Thanks for the comments. Before I respond to them, I wanted to  
> mention one change that Bob and I want to make in the next revision  
> in addition to the changes you request. It is just a small point of  
> clarification, but we don't want to surprise anyone. The proposed  
> new text, which would appear at the end of Section 6, is included  
> below:
>
> We note that in a network where L-LSPs are used, ECN marking SHOULD  
> NOT cause packets from the same microflow but with different ECN  
> markings to be sent on different LSPs. As discussed in [RFC3270],  
> packets of a single microflow should always travel on the same LSP  
> to avoid possible misordering. Thus, ECN marking of packets on L- 
> LSPs SHOULD only affect the EXP value of the packets.

thanks for the heads-up. This clarification seems fine to me.

(I'm cutting things from your reply that I agree with.)

> On Sep 11, 2007, at 5:40 AM, Lars Eggert wrote:
>> (1) Section 8.3 doesn't really discuss an "example use", which is  
>> what the title of Section 8 is. Suggest to cut it or move the  
>> content elsewhere.
>
> In my mind, section 8.3 is an example of how you can use ECN in  
> MPLS networks. I think the problem here is that we didn't say  
> enough to make that apparent. At the same time, I don't really see  
> any harm in dropping the section.

I'd prefer clarifying the example over removing it. If that's doable,  
great.

>> (3) My recollection was that there was an agreement to move the  
>> discussion of how one would use this to support PCN to an  
>> appendix. Some of this has happened, but Section 8.4 still exist  
>> in the main body of the document, and is longer than all the other  
>> example uses. I suggest to replace the content of Section 8.4 with  
>> a pointer to the appendix, and move the PCN details there.
>
> We certainly moved all the protocol details around PCN support to  
> an appendix as previously agreed. I have no problem moving most of  
> this section to an appendix too.

I'm not saying you should move all of it. Leave a sentence or two in  
Section 8.4, but move more of the details to the appendix. My basic  
concern was that there is still a lot of emphasis on the PCN usage  
example compared to the other ones, which could make it look as if  
PCN was _the_ intended usage, instead of one possible use case.

(I might be nitpicking here. If so, feel free to let me know.)

I tagged this as "Revised ID Needed" in the tracker. I fully expect  
the next revision that fixes these nits to directly go to IETF last  
call.

Lars

_______________________________________________
mpls mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/mpls