On Thursday 11 June 2009 22:57:14 Brett Nemeroff wrote: > 2009/6/11 Raúl Alexis Betancor Santana <r...@dimension-virtual.com> > > > On Thursday 11 June 2009 22:27:23 Brett Nemeroff wrote: > > > Does anyone know if there is an actual QSIG implementation of "REFER"? > > > > That > > > > > may make this whole issue moot. > > > > Could you describe what scenario are you trying to implement ? > > A call comes in to opensips, which sends the call to a TDM gateway. This > terminates to a ACD. And agent utilizes some mechanism to deflect the call. > It's necessary that the deflection occur on the far end of the sip > connection (so there is a sip to tdm translation in there, of which I don't > know if it can support a deflection). Possible options for the deflection > is either something that is supported natively in a qsig to sip mapping, or > an out of band solution. I was expecting this to be out of band such that a > script would poke at the fifo to perform such deflection. > > I know.. it's a weird setup.
So, you have ... SIP <-> TDM <-> SIP, yes is a weird and stupid setup ... You have 3 possible solutions: 1) SIP <-> B2BUA <-> TDM <-> SIP, Call-Deflection by DTMF controlled at B2BUA side 2) SIP <-> TDM <-> SIP ^ ^ |- SIP_faked_UAC -| So, for example, you could use some sipp script that "insert" a faked REFER from the "origin", but controlled by the far end SIP agent. 3) SIP <-> TDM <-> SIP, trying to use call-deflection features that the TDM gateway support (AFAIK, only usable with SS7 links or QSIG links but not with PRI, ANALOG or BRI links) -- Raúl Alexis Betancor Santana Dimensión Virtual _______________________________________________ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users