Cheers everyone,
ive been testing and searching for a way to achive FIFO within multiple Queues that an agent is part of?
So Agent A ist Part of Queue 1 and 2
all agents of these Queues are in Calls
Caller X is waiting for 2 Minutes and Caller Y for 6
Agent A hangs up and gets Caller Y.
Right now in tests it was often Caller X. As i learned it had something to do with Threat Race.
But i need to have different Queues for later analysing different Queues and Displaying Queue Names for when the Agent gets the Call.
Any way to archive this?
Thx in advance
It’s possible, but it requires a major rewrite of app_queue.c! I did something like this internally nearly ten years ago, but I’m retired now, and, in any case, I only implemented it for our use cases, e.g no consideration was given to Asterisk Realtime Architecture. I can’t remember the details now. I seem to remember keeping a single master queue, and also some mechanism for reserving callers, when an agent came free.
The basic problem is that agents don’t pick callers, but rather callers pick agents, with some special provisions to ensure that agents don’t get stolen from higher priority queues.
Note that the GPL only requires source code to be provided if you have the corresponding binary, and there is a time limit on that.
Also note that I haven’t looked at app_queue for many years, so this is based on how it worked about ten years ago.
thank you very much for your reply.