[jdev] Presence probe question

[Vinod Panicker]

On 2/10/06, Michal Vaner (Vorner) <michal.vaner at kdemail.net> wrote:
> Dne pátek 10 únor 2006 13:23 Vinod Panicker napsal(a):
> > Hi,
> >
> > According to RFC 3921, section 5.1.3, point 4 -
> >
> > "Else, if the contact has at least one available resource, the
> > server MUST reply to the presence probe by sending to the user
> > the full XML of the last presence stanza with no 'to' attribute
> > received by the server from each of the contact's available
> > resources (again, subject to privacy lists in force for each
> > session)."
> >
> > Taking this into consideration, lets look at the following scenario -
> >
> > - contact and user are in each other's rosters, with a subscription type of
> > both - contact sets presence as busy
> > - contact sends a directed presence to user, type away
> > - user's server sends a presence probe to contact
> > - contact's server returns the last presence stanza received with no
> > 'to' attribute - busy
> >
> > I think that the contact's server should send the last directed
> > presence received for the user, if applicable.
> >
> > WDYAT?
> >
> > Regards,
> > Vinod.
>
> Hello,
> if I got it right, the probe should be sent by the server only in the ocassion
> that the user just connected and the server does not yet know the foreign
> user's state. If the foreign user sent a direct presence, the user was not
> connected ad it got dropped. Therefore I think the situation you describe
> will not happen.But I can be easilly mistaken.

Typically yes, but not necessarily.  A server might send a presence
probe on behalf of the user to find out if the contact is still
around.

Even in the case you described, a Directed presence is valid for a
"session".  So replies to presence probes should be of the directed
presence, and not the global presence.

Regards,
Vinod.