All,

This is to confirm that the CTT has already been updated to use the All OCF
Nodes.  It looks like the only thing left to do (other than getting the ATG
to approve this CR) is to have the PF participants update their devices
with the patch, right?

Mitch

-----Original Message-----
From: [email protected]
[mailto:architecture_tg at openconnectivity.org] On Behalf Of Wouter van der
Beek (wovander)
Sent: Tuesday, July 12, 2016 2:07 AM
To: Mitch Kettrick; '???(Uze Choi)'; architecture_tg at openconnectivity.org
Cc: iotivity-dev at lists.iotivity.org; Jacek Hryszkiewicz
Subject: RE: [architecture_tg] IPv6 Multicast Address assignment for "All
OCF Nodes"

Comment #00108 - change ?All CoAP Nodes? into 'All OCF Nodes'

https://workspace.openconnectivity.org/apps/org/workgroup/architecture_tg/vi
ew_comment.php?comment_id=108&add_comment=1


Kind Regards,
Wouter

-----Original Message-----
From: [email protected]
[mailto:architecture_tg at openconnectivity.org] On Behalf Of Wouter van der
Beek (wovander)
Sent: 12 July 2016 09:45
To: Mitch Kettrick <cpm at openconnectivity.org>; '???(Uze Choi)'
<uzchoi at samsung.com>; architecture_tg at openconnectivity.org
Cc: iotivity-dev at lists.iotivity.org; Jacek Hryszkiewicz
<jacek.hryszkiewicz at comarch.com>
Subject: RE: [architecture_tg] IPv6 Multicast Address assignment for "All
OCF Nodes"

I am about to write the CR (since Stephane is on holiday...)

The lines in section 2047 ( v1.1.0) that references All CoAP nodes will be
replaced with All OCF nodes...

a)      All advertising or publishing Devices shall join the ?All CoAP
Nodes? multicast group, i.e. FF0X::FD for IPv6 and listen on the port
?5683?.
b)      The Client that wants to discover resources shall first join the
?All CoAP Nodes? multicast group. 
c)      Then the Client shall send a discovery request (GET request) to the
multicast group ?All CoAP Nodes? and port 5683 where the URI in the
request shall be /oic/res.

So we are not using All Coap Nodes multicast group anymore... so we are not
seeing CoAP multicast messages anymore and CoAP does not see the OCF
multicast messages anymore.

Kind Regards,
Wouter

-----Original Message-----
From: [email protected]
[mailto:architecture_tg at openconnectivity.org] On Behalf Of Mitch Kettrick
Sent: 12 July 2016 09:38
To: '???(Uze Choi)' <uzchoi at samsung.com>;
architecture_tg at openconnectivity.org
Cc: iotivity-dev at lists.iotivity.org; Jacek Hryszkiewicz
<jacek.hryszkiewicz at comarch.com>
Subject: RE: [architecture_tg] IPv6 Multicast Address assignment for "All
OCF Nodes"

Hi,

Comarch can make this change to the CTT but they asked for one
clarification.  Should the CTT change from the CoAP multicast address to
the OCF multicast address and completely forget about the CoAP address?  In
other words, once we're using the OCF multicast address (and this patch is
included in all devices that are certified), would the CoAP multicast
address ever be needed?

Thanks
Mitch

-----Original Message-----
From: [email protected]
[mailto:architecture_tg at openconnectivity.org] On Behalf Of ???(Uze Choi)
Sent: Monday, July 11, 2016 8:04 PM
To: architecture_tg at openconnectivity.org
Cc: iotivity-dev at lists.iotivity.org
Subject: RE: [architecture_tg] IPv6 Multicast Address assignment for "All
OCF Nodes"

Dwarka question has assumed that we may not apply this update into this PF
device.
But If we can do this by today or tomorrow, we don't need to concern both
compatible.
I think we can finalize all related tasks soon.

BR, Uze Choi
-----Original Message-----
From: [email protected]
[mailto:architecture_tg at openconnectivity.org] On Behalf Of Thiago Macieira
Sent: Tuesday, July 12, 2016 11:46 AM
To: architecture_tg at openconnectivity.org
Cc: Stephane Lejeune (stlejeun); Dwarkaprasad Dayama; iotivity-
dev at lists.iotivity.org
Subject: Re: [architecture_tg] IPv6 Multicast Address assignment for "All
OCF Nodes"

On segunda-feira, 11 de julho de 2016 23:45:10 PDT Stephane Lejeune
(stlejeun)
wrote:
> > 2. How do we plan to handle backward compatibility for devices being 
> > certified in in this week @ Poland?
> I have a couple ideas but prefer having this discussion after we 
> actually know we do have certified devices.

Suggestion: add a patch that keeps the All CoAP Nodes compatibility by
default, but has a compile-time option to turn it off (just All OCF Nodes). 
That way, we can choose whether we certify requiring both or just the new
address.

If we decide to certify only requiring the new address, it becomes the
user's decision whether to keep compatibility or not. They wouldn't have
to, for certification. But they may want to if they have pre-certified
devices that they want to keep compatibility with.

--
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center




-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2016.0.7640 / Virus Database: 4613/12581 - Release Date: 07/08/16


-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2016.0.7640 / Virus Database: 4627/12599 - Release Date: 07/11/16

Reply via email to