Re: [MEXT] weird idea about flow bindings

[Behcet Sarikaya <behcetsarikaya at yahoo.com>]

Hi Marcelo,
  The problem with 3697 seems to be not its contents but its deployment status.
I would agree with your proposed modification but how do we address the problem of getting it used? In mobile environments?

Regards,

Behcet



----- Original Message ----
From: marcelo bagnulo braun <marcelo at it.uc3m.es>
To: Charles E. Perkins <charles.perkins at earthlink.net>
Cc: mext <mext at ietf.org>
Sent: Wednesday, May 20, 2009 8:02:38 AM
Subject: Re: [MEXT] weird idea about flow bindings

Charles E. Perkins escribió:
> 
> Implementation of what?  Implementation of Mobile IP?  Implementation of IPv6?
> I guess the application would have to use something to indicate that the flow
> label should NOT be zero.  I still am not really seeing what's wrong with that.
> 
> In brief, since applications today do not use the flow label, I don't see what's
> wrong with the flow label being zero.  When applications are built that need to
> use the flow label, they'll have to have operating systems support to do it.
> 

I agree that if the application wants to define a flow i.e. set the flow label, that should be honored. So no argument here.

The question is whether a lower part of the stack could also have an idea of what a flow is.

My suggestion is that packets belonging to a connected socket (either TCP or UDP) by default form a flow.

So, i would like to change rfc3697 in the follwoing way:
If the application sets the flow label, then it must be honored.
If the application does not set the flow label, then the stack assign a rando flow label value for each connected socket and packets from that socket must carry the selected flow label.

would that make sense to you?

Regards, marcelo


> Thanks for any light you can shed on this.
> 
> Regards,
> Charlie P.
> 
> 

_______________________________________________
MEXT mailing list
MEXT at ietf.org
https://www.ietf.org/mailman/listinfo/mext