NEAX routing with URS

[Jeffrey Kelson]

To Genesys Support Community:

I was reading Genesys TServer for Neax manual and ran across the section dealing with setting up for URS.

Genesys suggests creating two routing points.

The first routing points, which should be set as type two in association has two two splits: split 1 gives control to URS which force routes the call to the second routing point and split 2 gives routing control to PBX in case Genesys is down.

The second routing point, which is set as type 1 is where the actual URS strategy should run.

They mention that the following CCV should be created and attached to PPN:

queue to split 50
delay 5
queue to split 1

I am not a PBX person, so, I do not understand:
1. where is CCV running? Is it running on routing point 1?
2. what is split 50? How do you define it in PBX and what does it do?


Here is what I have noticed:

if I create the following CCV and attach it to RP1, while setting RP association as type 1, everything works without the need for a redundant RP:

delay 5
queue to split 1

Genesys is worried that if there is a big load on URS then some of the calls may fall to default and this is why it is making us to have a second DN. But is it necessary?

For example, what is wrong with starting a treatment as a first thing that strategy does when call enters it? Would it not guarantee that there will be some kind of response in case URS is running within the five seconds? And still if it is not enough, wouldn't raising delay to 8 or 10 be sufficient?

In most cases, having Genesys is already considered to be a nonstandard operation, thus a 5 or 8 second delay should not matter that much, right?

Why does Genesys recommend it only for Neax and not for other PBXs as well?

[Klim]

When you route a call on Neax, there is always a chance that UR or TServer or something else will crash, leaving call physically stuck in a middle of nowhere with no way of getting it back. This is why you need this delay 5 etc thing behind.

Why do you need two routing points instead of one to accomplish it is very simple CCV on NEAX will continue executing the script after 5 seconds are up regardless of whether or not call is controlled by URS or not. This is why, you need two RP:

RP1: if URS works, it does not matter what you do, just make sure that you move the call away from this point as soon as possible, because if you linger for more than five seconds, the call will be sent to default split.

RP2: this is where your real strategy is ran. What happens to the calls what are queued on this RP when CTI goes down? Good question!
They get stuck!

So, having RP1>RP2 solution is not perfect: it will be able to deflect all the incomming calls to a separate split if CTI is down, but, nevertheless, all the existing calls queued on RP2 would still be stuck.

Me wonders if there isn't a more elegant solution out there.