I've experienced this problem, too. I've gotten MOH to work with parking two ways. One, using # transfer inside asterisk. Two, using a true SIP blind transfer. A SIP consultative transfer (such as the one my Polycom IP 500 does) causes the problem you describe.
I eventually enabled the # asterisk transfer to take the client out of it all together.
Using sip blind transfer makes no difference in my tests. I'll have to test # transfer, but after looking at the code for app_valetparking, I don't believe it will work regardless, unless I'm reading everything in ast_masq_valetpark_call backwords, it seems as though the musicclass variable member is not copied from the original channel to the masq/clone, thus MOH doesn't work. I'm putting together a test environment right now. I'll keep everyone posted on my progress.
As for normal * parking working with MOH, I cannot be sure if I tested sip blind to * parkandannounce (using callback setting) or not, but I think I did. I've tested so many configurations right now that I honestly cannot say if I have or not.
I'm also looking into exposing some chan_sip.c internals to the dial plan so you can _see_ (from the dialplan) the difference between a 'transfer' and an ordinary call. From looking at the code to chan_sip.c this looks very possible, again, I'll keep everyone posted. Is there a reason why some of the internal variables of chan_sip.c (REFFED-BY, and REFER-TO are key ones, as well as the determination between blind and attended transfers) are not already shared with the dialplan?
I realize this is becoming a -dev topic, and I'll move it there for further discussion about code (non dialplan) changes.
Thanks, Chris
_______________________________________________ Asterisk-Users mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
