Hi DJ,
Appreciate it.
Thx
Jorge
From: Dhananjaya Rao (dhrao) <dh...@cisco.com>
Date: Thursday, March 25, 2021 at 6:20 PM
To: Rabadan, Jorge (Nokia - US/Mountain View) <jorge.raba...@nokia.com>,
draft-dskc-bess-bgp-...@ietf.org <draft-dskc-bess-bgp-...@ietf.org>,
bess@ietf.org <bess@ietf.org>
Subject: Re: [bess] About draft-dskc-bess-bgp-car
Hi Jorge,
Thank you once again for reviewing the draft and for your feedback.
We will update the security considerations in a newer version.
The use of transposition here is to be efficient in bytes consumed in each NLRI
when multiple routes have shared information such as locator. It provides
flexibility to encode the common locator portion of SRv6 SIDs in the attribute.
We will clarify with an example in the newer version.
Regards,
-Dhananjaya
On 3/10/21, 7:03 AM, "BESS on behalf of Rabadan, Jorge (Nokia - US/Mountain
View)" <bess-boun...@ietf.org on behalf of jorge.raba...@nokia.com> wrote:
DJ, Swadesh,
Thanks for your time about these two drafts during the BESS session.
I just wanted to follow up on the two comments I made:
1) About this and the security section
“The indication of the key length enables BGP Speakers to determine
the key portion of the NLRI and use it along with the NLRI Type field
in an opaque manner for handling of unknown or unsupported NLRI
types. This can help Route Reflectors (RR) to propagate NLRI types
introduced in the future in a transparent manner.”
As discussed, this brings a potential security risk, since, for unknown
route types, the RR only validates the key length without understanding if the
content is wrong or not for the route type. I think DJ agreed this has to be
discussed into the Security section (which by the way does not exist in the
document in rev 01 __).
2) About the srv6 tlv and transposition:
"o SRv6 SID Information: field of size as indicated by the length
that either carries the SRv6 SID(s) for the advertised color-aware
route as one of the following:
* A single 128-bit SRv6 SID or a stack of 128-bit SRv6 SIDs
* A transposed portion (refer [I-D.ietf-bess-srv6-services]) of
the SRv6 SID that MUST be of size in multiples of one octet and
less than 16."
@Swadesh, in the srv6-services draft, transposition is used for efficient
packing, which *should not* be an issue here since the srv6 tlv is part of the
NLRI (at least for 1 SID). I 'suspect' the use case here is to *save* some
bytes when a stack of SIDs need to be advertised with the BGP CAR route, and
those SIDs have a few common bytes in the locator part? it would be good if you
can clarify please.
Thank you!
Jorge
_______________________________________________
BESS mailing list
BESS@ietf.org
https://www.ietf.org/mailman/listinfo/bess
_______________________________________________
BESS mailing list
BESS@ietf.org
https://www.ietf.org/mailman/listinfo/bess
