[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Mention TE in draft-ietf-shim6-proto-01.txt?
- To: Brian E Carpenter <brc@zurich.ibm.com>
- Subject: Re: Mention TE in draft-ietf-shim6-proto-01.txt?
- From: Pierre Baume <pierre@baume.org>
- Date: Sun, 23 Oct 2005 19:40:48 +0200
- Cc: shim6@psg.com
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=Lfjr7nndd7VWBmoSXRC2dxOIyEq3tCT/mhoiXI/Wcd4iVSFBOkrpZxcFcS53uWDQOLxbMqdtdNY/+RmRJrUTqXYL/OgkUWnFoSQmJ05PnlHG0X1lVbLiv9KpO13g3JtPn4RrZq1MynbZXenvm/1rovxQMGf9slv6Kp1yVpYsaCk=
- In-reply-to: <435796C5.80702@zurich.ibm.com>
- References: <435796C5.80702@zurich.ibm.com>
Hi Brian and all,
On 10/20/05, Brian E Carpenter <brc@zurich.ibm.com> wrote:
[...]
> There's quite a lot in the draft about how preferences are communicated
> and very little about where they come from. draft-ietf-shim6-arch-00.txt
> doesn't help with this either. Somewhere, I think we need a list
> of possible sources of these preferences, and dynamic or static TE policy
> needs to be in that list.
>
> Some discussion of traffic engineering in the architecture document
> would also seem necessary.
>
> Brian
Here's some text, below. It could be inserted in
draft-ietf-shim6-arch-00.txt after the 'Endpoint Identity' section.
Local Policy Settings
How does a host know which locator addresses can be used, in what
circumstances and how?
A Shim6-aware host needs a set of administrative and traffic
engineering policies:
- Locator addresses that can be used interchangeably, subject to
possible further constraints.
- Restrictions for the usage of some of the locator addresses, if
needed. They can be based on the ULP or on the availability of other
(preferred) located addresses.
- Recommandations for the proportions of traffic to send from
different locator addresses.
These parameters will be used in the Shim6 capability negociations
with other Shim6-aware protocol elements.
I hope that this is going in the right direction. :-)
Pierre.