Re: [sipcore] WG adoption Request

Hi!

In my view this document is far from ready for being a workgroup item. It assumes a specific audience, which from start is wrong for an IETF RFC. It assumes that REFER is in a specific type of dialog - a call.

If we start the discussion from the problem statement we may be able to find a way forward and restart with a different document.

Why would a REFER transaction need a specific failure code for failures? What is the specific problem with the existing set of response codes?

Is it a failure because of a specific policy that prevents the transaction from succeeding? 
Could this apply to other methods than REFER?

Think more generic.

Cheers,
/O

> On 25 Sep 2023, at 20:52, Ranjit Avasarala <ranjitkav12@gmail.com> wrote:
> 
> Hi Shrawan
> 

> while a new SIP response for call transfer failure may make sense, but adding it to existing SIP implementations would be quite overwhelming and kind of unnecessary.  Also like Brian suggests,  history-info header would be useful to know what happened to the call - while transfer was being attempted.
> 
> Also e.g. A calls B and B sends REFER to A with Refer-To: C, then A calls C,  then a call from A to C is essentially a new call.  So why do we really need 487?
> in another scenario, when A calls B and B transfers the call to C,  then say transfer to C fails, then here again, the call from B to C is like a regular call.  so here too I don't see the need for 487.
> 
> comments?
> 
> 
> On Mon, Sep 25, 2023 at 1:26 PM Shrawan Khatri <skhatri@qti.qualcomm.com <mailto:skhatri@qti.qualcomm.com>> wrote:
>> Hello Roland,
>> 
>>  
>> 
>> The UE’s subsequent behavior is defined by carriers based on the SIP response code, not based on the accompanying History-Info header or cause in the Reason header that comes with along with the SIP Response code, and the UE’s behavior for existing SIP response codes has already been defined by carriers. None of the existing SIP response codes are suitable to handle the case of call transfer failure. For instance, receiving SIP response code 403 might cause the UE to perform IMS re-registration, which would be pointless if the call transfer failed due to the carrier’s policy. The purpose of the draft is to enable the definition of a specific UE behavior for this new type of failure to meet new automotive use cases.
>> 
>>  
>> 
>> Note that in the past, similar drafts were submitted to introduce new SIP response codes for new use cases, in particular:
>> 
>>  
>> 
>>                                                 - RFC 8197: A SIP Response Code for Unwanted Calls (From https://www.rfc-editor.org/rfc/rfc8197.html )
>> 
>>                                                                 Response Code Number:  607
>> 
>>  
>> 
>>                                                 - RFC 3329:  Security Mechanism Agreement for the Session Initiation Protocol (SIP)
>> 
>>                                                                 - Response Code Number:  494  From https://datatracker.ietf.org/doc/html/rfc3329
>> 
>>  
>> 
>> Regards,
>> 
>> Shrawan
>> 
>>  
>> 
>> From: R.Jesske@telekom.de <mailto:R.Jesske@telekom.de> <R.Jesske@telekom.de <mailto:R.Jesske@telekom.de>> 
>> Sent: Wednesday, May 31, 2023 9:13 PM
>> To: Shrawan Khatri <skhatri@qti.qualcomm.com <mailto:skhatri@qti.qualcomm.com>>; br@brianrosen.net <mailto:br@brianrosen.net>
>> Cc: sipcore-chairs@ietf.org <mailto:sipcore-chairs@ietf.org>; sipcore@ietf.org <mailto:sipcore@ietf.org>; Lena Chaponniere <lguellec@qti.qualcomm.com <mailto:lguellec@qti.qualcomm.com>>
>> Subject: AW: WG adoption Request
>> 
>>  
>> 
>> WARNING: This email originated from outside of Qualcomm. Please be wary of any links or attachments, and do not enable macros.
>> 
>> Hi shrawan,
>> 
>> looking into the meeting notes I see that this was one of the comments that the draft needs to be more mature.
>> 
>> But there were also comments going along that the Cause Values/Reason Header existing are proper enough.
>> 
>>  
>> 
>> I have some problems with adopting a draft for transfer failures while we have the history info header which may also contain such information where and why the call was broken.
>> 
>>  
>> 
>> Did you have discussed such alternative in 3GPP.
>> 
>>  
>> 
>> Thank you and Best Regards
>> 
>>  
>> 
>> Roland
>> 
>>  
>> 
>> Von: sipcore <sipcore-bounces@ietf.org <mailto:sipcore-bounces@ietf.org>> Im Auftrag von Shrawan Khatri
>> Gesendet: Mittwoch, 31. Mai 2023 21:31
>> An: Brian Rosen <br@brianrosen.net <mailto:br@brianrosen.net>>
>> Cc: sipcore-chairs@ietf.org <mailto:sipcore-chairs@ietf.org>; sipcore@ietf.org <mailto:sipcore@ietf.org>; Lena Chaponniere <lguellec@qti.qualcomm.com <mailto:lguellec@qti.qualcomm.com>>
>> Betreff: Re: [sipcore] WG adoption Request
>> 
>>  
>> 
>> Hi Brian,
>> 
>> A related proposal was submitted in 3GPP CT WG1 (see C1-221242 <https://www.3gpp.org/ftp/tsg_ct/WG1_mm-cc-sm_ex-CN1/TSGC1_134e/Docs/C1-221242.zip>) but the feedback was that the group preferred the draft to be adopted by the SIPCore WG before being referenced in 3GPP specifications.
>> 
>>  
>> 
>> Regards,
>> 
>> Shrawan
>> 
>>  
>> 
>>  
>> 
>> From: Brian Rosen <br@brianrosen.net <mailto:br@brianrosen.net>> 
>> Sent: Wednesday, May 31, 2023 11:24 AM
>> To: Shrawan Khatri <skhatri@qti.qualcomm.com <mailto:skhatri@qti.qualcomm.com>>
>> Cc: sipcore-chairs@ietf.org <mailto:sipcore-chairs@ietf.org>; sipcore@ietf.org <mailto:sipcore@ietf.org>; Lena Chaponniere <lguellec@qti.qualcomm.com <mailto:lguellec@qti.qualcomm.com>>
>> Subject: Re: WG adoption Request
>> 
>>  
>> 
>> WARNING: This email originated from outside of Qualcomm. Please be wary of any links or attachments, and do not enable macros.
>> 
>> Answering my own question:
>> 
>>  
>> 
>> 3GPP Dependencies on IETF Drafts <https://whatthespec.net/3gpp/ietfdependencies.php>
>> whatthespec.net <https://whatthespec.net/3gpp/ietfdependencies.php>	
>>  <https://whatthespec.net/3gpp/ietfdependencies.php>
>>  
>> 
>> Doesn't list it.
>> 
>>  
>> 
>> Could some of the other 3GPP sip folks please comment on whether is likely to be added to the dependency list?
>> 
>>  
>> 
>> Brian
>> 
>>  
>> 
>>  
>> 
>> On May 31, 2023, at 1:44 PM, Brian Rosen <br@brianrosen.net <mailto:br@brianrosen.net>> wrote:
>> 
>>  
>> 
>> Is this draft on the IETF-3GPP tracking list?
>> 
>> 
>> On May 31, 2023, at 1:37 PM, Shrawan Khatri <skhatri@qti.qualcomm.com <mailto:skhatri@qti.qualcomm.com>> wrote:
>> 
>> Dear SIPCore chairs and SIPCore members,
>> 
>> We submitted v04 of our I-D on a new SIP Response Code (497) for Call Transfer Failure (available at: https://datatracker.ietf.org/doc/draft-khatri-sipcore-call-transfer-fail-response/). This version has addressed all the comments received to date and we would therefore like to ask for WG adoption.
>> 
>> 
>> The purpose of the draft
>> Signaling plane and Media plane of an IMS calls can be transferred between devices using IMS Signaling. There are various reasons why an on-going call cannot be transferred between the devices. Some of these reasons are policy driven, for instance: the call to be transferred is in the circuit switched (CS) domain and the operator's policy does not allow transfer of a CS call, the call is an emergency call and the operator's policy does not allow transfer of an emergency call, the call is a mobile-terminated call and the operator's policy does not allow transfer of a mobile-terminated call, or the call is a video call and the operator's policy does not allow transfer of a video call. The user agent (UA) initiating the call transfer procedure will be notified of any failure through a SIP response code. However the existing SIP response codes are not suitable to adequately convey the information regarding why the call transfer request is not accepted by the network. This is because handling of these existing response codes has already been implemented by various devices, with an associated device behavior defined for a specific purpose not related to call transfer. For instance, upon receiving some of these response codes, such as 403 (Forbidden), the device may initiate IMS re-registration procedure, which is not needed in case of Call Pull/Call Push failure and will result in unnecessary SIP signaling.
>> A method is defined in this draft such that a call transfer failure SIP response code is defined along with an optional warning code in a Warning header field to convey the exact reason why the call could not be transferred, so that the UA can determine the subsequent steps (e.g. call termination) and optionally provide an indication of the reason for the failure to the user.
>> 
>> Intended target audience
>> The following is the intended audience of this RFC:
>> *             IMS Core Network Planners that support IMS call transfer across multiple devices.
>> *             IMS System Designer and Developers of IMS Core Network and User Agent that support call transfer using IMS Signaling.
>> Benefit
>> *             In the absence of this new mechanism,
>> 
>> -          The IMS Core network has to rely on existing SIP response codes, for which a specific device behavior has already been defined. For instance, upon receiving some of these response codes, such as 403 (Forbidden), the device may initiate IMS re-registration procedure, which is not needed in case of call transfer failure and will result in unnecessary SIP signaling
>> 
>>  
>> 
>>  
>> 
>> _______________________________________________
>> sipcore mailing list
>> sipcore@ietf.org <mailto:sipcore@ietf.org>
>> https://www.ietf.org/mailman/listinfo/sipcore
> <image001.png><image001.png>_______________________________________________
> sipcore mailing list
> sipcore@ietf.org
> https://www.ietf.org/mailman/listinfo/sipcore