Q. How many CTI ports are required when for the makecall function and monitoring a station? is the CTI port persistent connection for each monitoring device?
A. To make a call 2 end points are needed. So in this case 2 CTIPorts or 1 CTIPort and a IPphone. Can you explain a bit on 'monitoring a station' (what you want to accomplish, what feature etc)?
Q. I use CTI for click to call without media. CTI just instructs the CUCM to call the called user IP Phone and the calling party(another Ip phone or PSTN number). I also use CTI to monitor the user's IPhone for on hook and off hook events. Does each CTI makecall in this case require 1 CTI connection and does the connection release once the makecall is completed or when the user ends the call. When requesting CUCM to monitor the phone events, does each request require 1 CTI connection and does the CTI connection remain connected.
A. So you need to do 2 things:
1. Make a call from users IP phone to another IP phone or PSTN number.
2. Find out if there is a active call on users IP phone.
You don't need to another CTIPort to do this. Cisco JTAPI interface Call.connect() API allows application to make a call from IP phone. Call activity on the phone are reported to Call observer. Application just needs to add call observers on users IP phone.
makeCall is a sample Cisco JTAPI application that uses connect() API to make the call. But it expects the called IP phone to be in the users control list. You should be able to make code changes to by pass that and use makeCall to make a call from users ip phone to any number.
Run JTrace sample app to see the call events reported. CallActiveEv indicates the start of a call and CallInvalidEv indicates the end of the call.
Q. but I still need to understand the CTI port usage to determine if we have enough CTI port capacity in our cluster.
A. First, go to service parameters and navigate to CtiManager.
The amount refers to the count of the Cti connections opened.
You can change the amount by two different setting options: "Maximum Devices Per Provider" field or "Maximum Devices Per Node" field.
Next, click on the fields blue link (as I mentioned), there you could find what is the maximum usage u can using regardlng to Cisco rules.
You can see the current Cti connections amount by RTMT and Cti.
If using JTAPI to make call between two IP Phones, how many CTI ports are needed and do these ports release once the call established.
If using JTAPI to monitor ONE IP Phone for events such as on hook and off hook, how many CTI ports are needed and do these po?ts remain active in CUCM?
Think CTI port as TCP transaction.
if any action performed on the phone, in the background the Cti Client (Phone/PC) makes new TCP request to the CTI Service (CTIManager in CUCM for example).
Each TCP flow belong to same source takes one CTI Port.
When the call is established the Cti port is going down till the next action.
But if we talk about Cti monitoring, the Cti port doesn't close till the application will stop it.
So in case you will make a call, you catch one temporary Cti port till side B will answer the call.
But if side B have Cti monitoring, you have one Cti port permanent.