maybe - it's hard to tell, and I did bring up that other option as well. It's just that when a 1-1 call upgrades to a multiparty call, there should a be a lot of new stuff going on at signaling level to convert that 1-1 call to multiparty - and a chance is that it's a combination that leads to this bug - a call gets upgraded to a multiparty AND a number being added is already part of the call...