[Gen-art] Re: Latest updates : draft-ietf-l2tpext-pwe3-ethernet-08.txt

[Cullen Jennings]

My apologies - I did not realize there was text for me in this email till I got the ping today :-) More inline ... On Aug 4, 2006, at 12:03 PM, Maria A. Dos Santos ((mariados)) wrote: ########################################################################### *Cullen Jennings:* *Discuss:*    *[2006-07-05]* I may not understand the congestion control technique but    it looks like it says,  "in case of congestion, pull the plug". I don't    believe anyone will implement this congestion control suggestions - and    if they do, I don't believe the resulting systems will be usable - they    will be impossible to manage not to mention the incredible DOS    opportunities. No advice on the threshold is provided. I think this    needs to be updated to either be clear no congestion control is provided.   From what I understand, the draft is inline with the current state of congestion control for emulation technologies, and PWE3 WG is currently conducting further study on the topic.  Once we get the result of that, this draft and all other pwe drafts (both l2tp and mpls) will leverage from that mechanism.  So at this point, the goal of the "Congestion Control" section in this draft is only to emphasize the possibility of congestion and provide some general best-effort recovery mechanisms.  Would it be better if I emphasize the goal? I'd have no problem with something along the lines of:  This specification offers no congestion control and is not TCP friendly <add ref to where TCP Friendly is define>. Future works, such as <insert non normative reference that will not block your document>, may extend this to provide congestion control.   I am not sure what general default threshold is appropriate given that the threshold will vary according to network size and average traffic load of individual network.  That's why it is left undefined and left to administrator to experiment. Hmm - I know something about networks and I have no idea how I would set up the parameters such that they would work and not mess up other protocols on the same network. I understand that how you might want the parameters on a satellite link, vs a corporate switched LAN, to be somewhat different. I view the problem with selecting numbers here is that no matter what you pick, it is clear many people will disagree with it and there will be cases where it does not work. That is typically a sign that algorithm might be less than ideal regardless of the parameters. However, if we are agreed that for practical deployments and implementation, we expect that PWE will not have automated congestion control, then I have no real problem with the text you have here and we can just ignore trying to figure out if this provides enough advice that anyone will be able to get something that works.       The section titled "Applicability Statement" gives me no idea where this    is applicable and where it is not.   Would you mind elaborating what information are you expecting here?   I might have this wrong but talking to people I get the idea that the expectation is that this protocol will be used across the public internet  as well as private networks and will be used for many things including forming virtual network connection between remote offices for enterprises.

_______________________________________________
Gen-art mailing list
Gen-art@ietf.org
https://www1.ietf.org/mailman/listinfo/gen-art