Hi All
 
We want to use the external call control feature of the call manager. We started simply and wrote a simple application that only sends responses to keep-alive messages that recieved from CUCM (8.6.2.21011-1).
 

 
Our appliaction is an ASP.Net web site, on Page_load event, we configured the response header and sent the response. Also we added a small log that writes the incoming request header fields when the page is called. But, it seems to be there's no call to web application. We are also pretty sure that the translation pattern works coreectly, our test call hit the translation pattern. But at least, CUCM must send keepalive messages to application where there's no sign of messages.
 
Note that, the server that hosts tha application does not have an IP address with a zero in any of the octets.
 
Any ideas?
 
Thanks
Ergin

Hi Michael

Thanks for the quick reply. No, I did not specified port ont he URL. Is it a must?
 
Edit: We opened External Call Routing trace, in the command line interface we see
 
 
|HttpHandler - sendRoutingRej profile registered app is undefined
!!ERROR!! -ConnectionManager- wait_AuDisconnectRequest :CI NOT FOUND IN TABLE,CI(32364269,0),discType=1,IFCreated(0,0),PID(0-0,0-0),IFHandling(0,0),MCNode(0,0)|1,100,56,1.82981^IPADDRESS^MACADDRESS
 
Thanks

Michael

Thank you very much for the information. Are there any constraints about the port number?

I used

http://applicationName:80/default.aspx

but still no movement..
 
Edit : I can call the web site from my browser with this link and I have no problem. But no keepalive messages from CUCM right now.
CM detailed logs changed, after adding port number, I started to get log:
 
15:57:43.772 |HttpHandler - sendRoutingRej profile registered app is undefined|1,100,56,1.83625^IPADDRESS^MACADDRESS
15:57:43.772 |HttpHandler - wait_RoutingReq no connection available/in service or handler not in active, send back routing rej |1,100,56,1.83625^IPADDRESS^MACADDRESS
 
 
Thanks

I'd suggest that you change the URL in the profile to use an IP address instead, to check that the problem isn't in name resolution.

Hi All

Thanks for your responses. Now we have a URL with http://applicationName:80 and we saw that while resolving the url from dns server, call manager sends queries with AAAA type which means the request type is IPV6 but the problem is, IPV6 is disabled on call manager. Because of that, our dns servers returns a response with failure because it does not support IPV6. We are unable to understand why call manager behaves like this.

Also, now we are unable to call this application via IP address, but tomorrow we are going to request a new server and deploy the application on it, in order to call the application via IP address.

Any ideas about this IPV6 issue? Did you faced with that problem before?


@Michael:
Yes our traces are open and we are watching them with command line interface. While making a call, it says that "profile registered app is undefined". I think this is because, there's not a clear HEAD messaging between web application and CUCM.


Thanks

If there's no handshake using HEAD, Unified CM will never send a route request. Going to have to resolve the other issue first.

I don't have access to my lab at the moment, and all my URIs are currently set up as IP addresses. Will try DNS when I have a chance.

I switched my URI to use a domain name instead of IP address and it works fine.

http://userver2.mytest.org:80/block1

(not a real domain name, but served by a normal BIND server)

Two things:

1. Did you enable DNS client on your Unified CM server when you installed it? That's your one shot at enabling DNS on a cluster.

2. I try to keep my URIs pretty simple to alleviate issues. Let the web server take care of the work of mapping a simple URL to the complexity of your web server. The acutal path to my appliciation is /var/www/wsgi-scripts/sampleBlockApp.wsgi, but I let Apache alias it to /block1 instead. This works much more reliably...

Mark