Re: [sipcore] Reason as a parameter rather than an escaped header

Hi Hadriel,

for 1) I see this as a option which the service provider could choose. Not everybody will be mandated to put in a Reason on a internal event within the server which ends in a retargeting/diversion.
for 2) I think it could be a normal Reason value, based on a equivalent processes as it would be sent out by an other sever. E.G. If a timer expires a 408 could be included.

Best Regards

Roland

  _____

Von: sipcore-bounces@ietf.org [mailto:sipcore-bounces@ietf.org] Im Auftrag von Hadriel Kaplan
Gesendet: Freitag, 10. Dezember 2010 17:47
An: Mary Barnes
Cc: sipcore@ietf.org WG
Betreff: Re: [sipcore] Reason as a parameter rather than an escaped header

OK, so I think there's general agreement to add additional H-I header fields for internal retargeting.

I have two questions then:

1) the text below says: "For retargets as a result of timeouts or internal events, a Reason MAY be associated with the hi-targeted-to-uri that has been retargeted."
Why is this a MAY?  What's the alternative?  Does it mean they may do something else, like use a RFC 4458 style cause URI parameter?  Does it mean they may associate it with a different hi-targeted-to-uri? (I assume not, but the text isn't clear)

For example, is this what you really want to say: "For retargets as a result of timeouts or internal events, a Reason MUST be associated with the hi-targeted-to-uri that has been retargeted and encoded as an embedded Reason header field in the URI, unless the reason for the retargeting is unknown."

2) What form of reason-value protocol can be used for a Reason from an internal operation?  Can it be a "SIP" cause?  Ultimately if this info is used by a receiver of the H-I entries to trigger different behavior/features, it would be really nice not to have to create a bunch more values that the receiver would have to understand/support.

-hadriel

On Dec 6, 2010, at 10:38 AM, Mary Barnes wrote:

Hi Marianne et al,

I totally agree that there was some text removed from RFC 4244 that was intended to handle the internal retargeting case. I would suggest we add that back, updating the paragraph to be a little more concise as I suggested earlier in the thread and add a note with regards the definition of any new Reason headers - something like the following:

  If the response contains any Reason header fields, then
  the Reason header fields MUST be captured as Reasons
  associated with the hi-targeted-to-uri that has been
  retargeted.  If the SIP response does not include a Reason header field
  (see [RFC3326]), the SIP  Response Code that triggered the retargeting
  MUST be included as the Reason associated with the hi-targeted-to-uri
  that has been retargeted.

  For retargets as a result of timeouts or internal events, a Reason
  MAY be associated with the hi-targeted-to-uri that has been
  retargeted.  [MB: this is the original text from RFC 4244.]

  In the case that additional Reason headers are defined, per RFC 3326,
  the use of these Reason headers for the History-Info header field
  MUST follow the same rules as described above.

Thanks,
Mary.

On Thu, Nov 25, 2010 at 4:33 PM, <marianne.mohali@orange-ftgroup.com<mailto:marianne.mohali@orange-ftgroup.com>> wrote:

Hi,

I agree that draft-mohali-sipcore-reason-extension-application could live independently of 4244bis, except for the section "Reason in the History-Info header" that should still allow wat is proposed in draft-reason.

Note that RFC4244 is compatible with the draft-reason proposal: As work on 4244bis was in progress, we based the draft on the following text from RFC4244: "For retargets as a result of timeouts or internal events, a Reason MAY be associated with the hi-targeted-to-uri that has been retargeted."

Unfortunately, this sentence disappeared and only the last sentence about timeout suggests to insert a Reason for an internal process.

If there is no objection, we could put this text back in 4244bis to keep explicit the ability to insert the Reason header field in a H-I entry for *internal* reasons (with a MAY).

Regards,
Marianne

> -----Message d'origine-----
> De : sipcore-bounces@ietf.org<mailto:sipcore-bounces@ietf.org>
> [mailto:sipcore-bounces@ietf.org<mailto:sipcore-bounces@ietf.org>] De la part de Christer Holmberg
> Envoyé : jeudi 25 novembre 2010 07:48
> À : Paul Kyzivat
> Cc : Worley, Dale R (Dale); sipcore@ietf.org<mailto:sipcore@ietf.org>
> Objet : Re: [sipcore] Reason as a parameter rather than an
> escaped header

>
>
> Hi,
>
> >>I think we should ask ourselves: assuming we allowed to do what
> >>Marianne is proposing, would anything break?
> >>
> >>Does anyone really care whether a H-I entry was inserted based on a
> >>"real" or "virtual" response? Aren't people more interested in the
> >>actual reason value?
> >
> >I don't currently see a problem with permitting this (though I'm
> >interested to hear if somebody else sees an issue).
> >
> >But IMO the current text doesn't suggest to me that this is valid.
> >So if the desire is for it to be valid it would be good to have some
> >text that makes it so.
>
> I agree. We would need to add some text.
>
> Regards,
>
> Christer
>

> _______________________________________________
> sipcore mailing list
> sipcore@ietf.org<mailto:sipcore@ietf.org>
> https://www.ietf.org/mailman/listinfo/sipcore
>
_______________________________________________
sipcore mailing list
sipcore@ietf.org<mailto:sipcore@ietf.org>
https://www.ietf.org/mailman/listinfo/sipcore

<ATT00001..c>