Re: [core] CoAP over TCP

[Ben Kinsella <[e-mail]>]

Hi.

Apologies for resurrecting this thread, but I haven't seen any updates on this topic since November.

>> 3) CoAP over TCP.  draft-bormann-core-coap-tcp-01 is a summary of the
>> design choices. draft-tschofenig-core-coap-tcp-tls selects one option
>> without explaining why.
>
>draft-tschofenig-core-coap-tcp-tls describes what we (ARM and Zebra)
>have implemented and deployed.

draft-bormann-core-coap-tcp expired in January, and draft-tschofenig-core-coap-tcp-tls is due to expire next week.
Please excuse my ignorance of how the IETF process works, but my understanding was that the IETF would standardise on one approach for CoAP over TCP.
Is this still going to happen?

My interest is in the use of CoAP and LWM2M over cellular networks, and I have some concerns over the use of UDP in scenarios which do not support VPNs or private APNs.
Can anyone provide real-world examples of the successful use of CoAP over UDP in this type of environment?
For example, my understanding is that LWM2M already makes a slight modification to CoAP in order to get around the cellular NAT issue: the device initiates the connection to the head-end.
Is UDP as the transport simply not an issue, and therefore there is no real motivation for CoAP over TCP?

I was hoping that IETF standardisation of CoAP over TCP would also trigger changes in OMA's LWM2M spec.
For example, if I adopt the ARM approach outlined in draft-tschofenig-core-coap-tcp-tls, I presume I cannot use a standard LWM2M server?
In what scenarios is the ARM/Zebra approach used?

Alternatively, I know that there are also efforts underway to add an MQTT binding to LWM2M, as an alternative to CoAP.
Which is likely to reach standardisation first: CoAP over TCP, or LWM2M over MQTT?

Regards,
Ben.

Ben Kinsella | Senior Engineer
B+B SmartWorx | Westlink Commercial Park, Oranmore, Co. Galway, Ireland
Office: [phone] | 
Email: [e-mail] | bb-smartworx.com
Skype: ben.kinsella | ie.linkedin.com/in/benkinsella/


Hannes Tschofenig | 11 Nov 00:33 2014
Re: CoAP over TCP

Hi Carsten,

thanks for the write-up.

On 10/24/2014 06:11 PM, Carsten Bormann wrote:
> In alternative transports, I believe we have three hot topics:
> 
> 1) General issues.  draft-silverajan-core-coap-alternative-transports
> is a good basis for this and we will do the call for adoption soon.
> 
> 2) CoAP over SMS.  This is also related to the question of DTLS over
> SMS.  We have a good draft for the former, which is mostly waiting
> for (1).  draft-fossati-dtls-over-gsm-sms merits some discussion.

Thanks for mentioning draft-fossati-dtls-over-gsm-sms; I am happy to
briefly describe what we had been working on.

> 
> 3) CoAP over TCP.  draft-bormann-core-coap-tcp-01 is a summary of the
> design choices. draft-tschofenig-core-coap-tcp-tls selects one option
> without explaining why.

draft-tschofenig-core-coap-tcp-tls describes what we (ARM and Zebra)
have implemented and deployed.

Our use case is firewall traversal in enterprise networks. Our goal was
to use our existing CoAP over UDP implementation. Doing optimizations to
CoAP when we already add all the complexity of TCP didn't seem
worthwhile to us.

Ciao
Hannes