Re: [Detnet] Alissa Cooper's Discuss on draft-ietf-detnet-architecture-11: (with DISCUSS and COMMENT)

+1 on Alissa’s near-final comment:

> […] the architecture does not actually forbid the possibility of adding additional information into DetNet packets that could be used as vectors for tracking/correlation.

My understanding is that this is allowed by the DetNet architecture even though none of the data plane designs have done so to date.  In particular, I’ve seen some prior discussion of using DSCPs for to identify DetNet traffic, even though that approach is not currently being pursued.

Thanks, --David

From: detnet <detnet-bounces@ietf.org> On Behalf Of Alissa Cooper
Sent: Wednesday, April 17, 2019 9:34 AM
To: János Farkas
Cc: draft-ietf-detnet-architecture@ietf.org; detnet@ietf.org; Lou Berger; IESG; detnet-chairs@ietf.org
Subject: Re: [Detnet] Alissa Cooper's Discuss on draft-ietf-detnet-architecture-11: (with DISCUSS and COMMENT)

[EXTERNAL EMAIL]
Hi János,

On Apr 15, 2019, at 5:08 PM, János Farkas <janos.farkas@ericsson.com<mailto:janos.farkas@ericsson.com>> wrote:

Hi Alissa,

Please see in-line.
On 4/12/2019 7:08 PM, Alissa Cooper wrote:
Hi János,

On Mar 25, 2019, at 12:16 PM, János Farkas <janos.farkas@ericsson.com<mailto:janos.farkas@ericsson.com>> wrote:

Hi Alissa,

We believe that we have addressed your comments in the most recent revision: https://tools.ietf.org/html/draft-ietf-detnet-architecture-12. (https://mailarchive.ietf.org/arch/msg/detnet/utVL9ZVGcOeGtRIASRFx5WT_ErM)

Please let us know what else you would like to see done before you clear your DISCUSS.

I/we would be happy to meet with you this week if there is anything you would like to discuss.

Regards,
Janos

On 2/26/2019 2:20 PM, János Farkas wrote:
Hi Alissa,

Thank you for your review!

We can replace
"DetNet is provides a Quality of Service (QoS), and as such, does not
   directly raise any new privacy considerations."
with
"DetNet provides a Quality of Service (QoS), and as such, is not expected to
   directly raise any new privacy considerations.”

I don’t understand why this is not expected. From what I can tell, the architecture allows for the use off domain- or app-flow-specific IDs. These seem like a new potential vector for tracking, and one that not every QoS architecture requires.
As far as I understood from below, you were happy with the change to "is not expected”.

“Combined with the above removals” (about flow IDs and OAM) is what I said. Either the architecture allows for these things, in which case one might expect them, or it doesn’t.

Combined that with this discussion on the related text:
https://mailarchive.ietf.org/arch/msg/detnet/-L_wsGPMqNEOtPMgsGYaJ30nFXk

We came to the current text:
   DetNet provides a Quality of Service (QoS), and as such, is not
   expected to directly raise any new privacy considerations, the
   generic considerations for such mechanisms apply.  In particular,
   such markings allow for an attacker to correlate flows or to select
   particular types of flow for more detailed inspection.

Flow ID and associated QoS is not a new concept introduced by DetNet; therefore, does not expected to raise new concerns.

What specific further changes do you suggest to this text?

Given your exchange with Benjamin and the email today from Stewart, I would suggest:

DetNet provides a Quality of Service (QoS) mechanism, and the generic
   considerations for such mechanisms apply.  In particular, such markings
   allow for an attacker to correlate flows or to select particular types
   of flow for more detailed inspection.

My point is that the architecture does not actually forbid the possibility of adding additional information into DetNet packets that could be used as vectors for tracking/correlation. This change would also address my concern about the OAM piece, but if you or the WG want to pursue further edits regarding OAM given Greg’s mail I’m happy to review.

Best,
Alissa

This edit also doesn’t seem to cover the potential for additional privacy exposure implied by the discussion of OAM in Section 4.1.1:
Thank you for making clear that this is the text you mean under "novel flow IDs and OAM tags" below. It was not clear to me because the architecture document does not contain either new / novel   Flow ID / OAM tag.

"OAM can involve specific tagging added in the packets for tracing implementation or

   network configuration errors; traceability enables to find whether a

   packet is a replica, which DetNet relay node performed the

   replication, and which segment was intended for the replica.

This text is there since: https://tools.ietf.org/html/draft-finn-detnet-architecture-05.

Active

   and hybrid OAM methods require additional bandwidth to perform fault

   management and performance monitoring of the DetNet domain.  OAM may,

   for instance, generate special test probes or add OAM information

   into the data packet.”
This was added based on https://www.ietf.org/mail-archive/web/detnet/current/msg01577.html.

This paragraph reads to me as pretty generic text on OAM.

What specific cahnge would you like to see made to this text?

Should we, e.g., update the first sentence of the paragraph to:
   OAM can support tracing implementation or
   network configuration errors. Traceability enables to find whether a
   packet is a replica, which DetNet relay node performed the
   replication, and which segment was intended for the replica.

?

Regards,
János

Thanks,

Alissa

I'm not sure what "references to new flow IDs and OAM tags should be removed"?

Could you point to the text that should be changed?

Thank you!
Janos

On 2/20/2019 4:39 PM, Lou Berger wrote:

On 2/20/2019 10:25 AM, Alissa Cooper wrote:

On Feb 20, 2019, at 7:17 AM, Lou Berger <lberger@labn.net<mailto:lberger@labn.net>> wrote:

Hi Alissa,

Thanks for the comments - see below.

On 2/20/2019 9:54 AM, Alissa Cooper wrote:

Alissa Cooper has entered the following ballot position for
draft-ietf-detnet-architecture-11: Discuss

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)

Please refer to https://www.ietf.org/iesg/statement/discuss-criteria.html
for more information about IESG DISCUSS and COMMENT positions.

The document, along with other ballot positions, can be found here:
https://datatracker.ietf.org/doc/draft-ietf-detnet-architecture/

----------------------------------------------------------------------
DISCUSS:
----------------------------------------------------------------------

= Section 6 =

"DetNet is provides a Quality of Service (QoS), and as such, does not
   directly raise any new privacy considerations."

This seems like a false statement given the possibility that DetNet may require
novel flow IDs and OAM tags that create additional identification and
correlation risk beyond existing fields used to support QoS today.

Based on the other work in the WG, I think "is not expected" is more accurate than "does not". This is based on the WG solutions for the DetNet data plane using existing IP (v4 or 6) headers or MPLS labels for flow identification.

If that is the case then the references to new flow IDs and OAM tags should be removed from the architecture.
sounds reasonable.  Can you point to the specific offending text?
Thanks,
Lou

Would changing to "is not expected" address your concern?

Combined with the above removals, that would work for me.

Thanks,
Alissa

Thanks,

Lou

_______________________________________________

detnet mailing list

detnet@ietf.org<mailto:detnet@ietf.org>

https://www.ietf.org/mailman/listinfo/detnet