> On Fri, 2009-08-21 at 09:58 -0400, Robert Joly wrote:
> > I was investigating a recent failure in the sipXopenfire 
> CallWatcher 
> > plug-in.  That plug-in is effectively a client to the RLS 
> server and 
> > it has stopped working reliably in recent builds.  Looking into the 
> > issue, I traced the problem to a change made to the information 
> > content of the dialog event message bodies that the RLS sends.  I'm 
> > not sure what is the correct behavior, so I'm turning to the list.
> >  
> > In RLS from build 15701 I see that the 'entity' information 
> element of 
> > the dialog event body created by RLS is of the form:
> > entity="sip:2...@sipdomain"
> >  
> > Whereas the 'entity' IE created by RLS version 16208 is of the form:
> > entity="sip:[email protected];X-sipX-nonat"
> >  
> > In other words, the earlier version seemed to use the AOR 
> whereas the 
> > more recent version seems to use the contact.
> >  
> > Can someone enlighten as to which one is correct (if not
> > both) and whether this change was intentional?
> 
> Looking at the snapshot you provided, I see that the 
> subscription is to:
> 
> sip:[email protected]
> 
> That is, it is asking for full-format event lists.  In those 
> lists, the <dialog-info> bodies are for individual contacts 
> of the AORs listed in the resource list, and as far as I can 
> tell, "sip:[email protected];X-sipX-nonat" is a contact for 
> "sip:[email protected]".
> 
> I suspect that you previously subscribed to 
> sip:[email protected], the 
> consolidated list format, which synthesizes <dialog-info> 
> elements with entity attributes that match the members of the 
> resource list.
> 
> In any case, you shouldn't be correlating the <dialog-info> 
> bodies with the members of the resource list by examining the 
> entity attributes, you should be parsing the RLMI and 
> following the cid attributes of the <instance> elements.

Got it.  Thanks for the advice - I will do that.  I'll also probably
move back to the consolidated presence info as I do not need the
granularity offered by the full version.  Thanks for looking into that!!

bob
_______________________________________________
sipx-dev mailing list [email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev
sipXecs IP PBX -- http://www.sipfoundry.org/

Reply via email to