Author Message
MichaelNorman
Joined: Jun 3, 2015
Messages: 448
Offline
Attempting to use the following code from a web service method to obtain UUI for the call:

Call call = CallFactory.getCall(UCID);
String contextId = call.getContextId();


Fails with:

Could not execute GETCONTEXTID:com.avaya.collaboration.call.async.AsyncCallSipImpl incompatible with com.avaya.collaboration.call.Call

JoelEzell
Joined: Nov 15, 2013
Messages: 780
Offline
Hi Michael. With this sort of scenario in a multi-node cluster, it's likely that the HTTP request could land on a different server than has the call. Our expert in this area thinks that we may have fixed a problem in this area in our 3.3 or 3.4 release.

I checked with our product manager and he said that an upgrade to Breeze 3.4 is in the works at Comcast. Do you happen to know if that's the system you're working on?

We'd really like to get you on a more recent release, as it takes development cycles from our team to diagnose issues on old releases that may have been fixed in recent releases.
MichaelNorman
Joined: Jun 3, 2015
Messages: 448
Offline
Ok you're probably right on this. I adjusted my code to only attempt to parse when the call is on the local node. This is just more reason why we need to get the load balancer working from the other thread to be sure we hit the right node correctly every time.

To answer your question it's a complicated question. We have 4 divisions and each has their own instances. Eventually this program is likely to be running on all of them, and if getting to 3.4 is a requirement, we are definitely going to need some concrete reasons as to why, but we can handle that when we get there.

National (which I work with) does not have Breeze just yet, but will be 3.4 when we install in a month or so. West Division has 3.2 and I'm not sure on their timetable to upgrade. North has 3.3, and has no plans to upgrade as of now. Central I am not sure what they are running, but attempting to find out.
JoelEzell
Joined: Nov 15, 2013
Messages: 780
Offline
OK, thanks Michael. There's a chance that the load balancer difficulties might go away with a more recent version as well. We'll need to carefully consider how much debugging effort to invest into 3.2. I'll pass this information back to our management chain and see how they'd like to handle it.
Go to:   
Mobile view