Cisco Support Community
Showing results for 
Search instead for 
Did you mean: 
New Member

Dialing Sx20 from CUCM through SIP Trunk to VCS.

Hey Everyone,

  I am doing some configuration and testing to tie our CUCM and VCS infrastructures together using SIP trunking. I have a lab environment setup with a CUCM Pub/Sub, CUPS, test phone, VCS Control, 6000MXP, and Sx20. I have base configs on the CUCM and VCS to allow basic call handling and routing. I then added the configuration to both units identified in the


document on Cisco’s website. I can get a call established between the 6000MXP and the phone just fine. It works both dialing phone from codec and the other way around.

  The Sx20 is a different story. I can dial the phone from the Sx20 and that works fine. However, if I try to dial the Sx20 from the phone, the Sx20 rings for a split second and then immediately shows a missed call. The phone doesn’t even complete 1 ring before it disconnects. If I look in the VCS logs it shows that call as rejected with “Request Terminated” as the detail. I have everything set to SIP TCP and am not trying to use TLS. The software versions of everything are below.

CUCM 8.6.2

VCS: X7.2

Sx20: TC6.2.0

6000MXP: F9.3.1 NTSC

Any ideas on what I am missing to get this working?


Everyone's tags (4)
Cisco Employee

Re: Dialing Sx20 from CUCM through SIP Trunk to VCS.

What do you see in the search history on the vcs? It would be most helpful to take a diagnostic log from the vcs. Go to Maintenance > Diagnostics > Diagnostic Logging and set the network and interworking log level to Debug. Start the logging, then place the test call. When it fails, stop the log and download it. Now search through the logs for the inbound SIP invite from Cucm. From the invite, copy the call-Id and search through the logs using that. You should see in the logs where the call failed.

Is the call being interworked to h323 on the vcs when you call from the phone to the sx20? It almost sounds like the tcp connection is being torn down. Can you copy and past the search history from the vcs into here for us to see?

Chad Patterson
Sent from Cisco Technical Support iPad App

New Member

Dialing Sx20 from CUCM through SIP Trunk to VCS.

It is a SIP to SIP call, no h323. I did the debug log and found the area where the cancel SIP message is being sent out (shown below). It looks like the reason is identified as Q.850 reason=3 which after doing a little digging seems to be a "no route to destination" error. But if there is no route to destination how does the Sx20 see the call come in and report the missed call? Also the 6000MXP works fine and the only difference is it's SIP address is 891500.

2013-11-19T08:48:24-06:00 cisco tvcs: UTCTime="2013-11-19 14:48:24,787" Module="network.sip" Level="INFO": Src-ip="" Src-port="51534"   Detail="Receive Request Method=CANCEL, Request-URI=sip:891501@:5060, Call-ID=a2e33180-28b17a38-5339-99211cac@"
2013-11-19T08:48:24-06:00 cisco tvcs: UTCTime="2013-11-19 14:48:24,787" Module="network.sip" Level="DEBUG": Src-ip="" Src-port="51534"
|CANCEL sip:891501@:5060 SIP/2.0
Via: SIP/2.0/TCP :5060;branch=z9hG4bK53611c1d2903;received=
Call-ID: a2e33180-28b17a38-5339-99211cac@
CSeq: 101 CANCEL
From: "Steven QA Phone 2062" <>@>;tag=74587~1266a0f2-0c69-456a-b667-1d7a6a208bb9-46714251
To: <891501>>
Max-Forwards: 70
Date: Tue, 19 Nov 2013 14:48:24 GMT
Reason: Q.850 ;cause=3
Content-Length: 0


New Member

Dialing Sx20 from CUCM through SIP Trunk to VCS.

Hi Steven

I had a similar problem some months back, the details are a bit sketchy in my head, here are the notes I took at the time.

When the VCS receives SIP request from the CUCM the format is 111@ This is not how the end devices register on the VCS, a device would register as 111@domain.local.

So to allow the search rules to find a match we must “transform” the domain from an IP address to a URI format.

Also when a call is made from Unified CM to VCS, the callback address (buried in a SIP message) is presented as number@. If the VCS-registered endpoint returns the call, the VCS needs to be able to route it back to CUCM. To enable this, the domain portion of the address must have the IP address removed and the video domain added (so that the existing search rule can route the call to Unified CM). A transform is also required to convert the ip address to the fqdn of the CUCM.

Also make sure the vcs fqdn and the CUCM fqdn are fully resolvable in DNS.


Dialing Sx20 from CUCM through SIP Trunk to VCS.

Hi Steven,

it seems that you are running into the known issue "CSCty07061"

Symptom:Call fails when DNS look up for a contact header fails, even if Record-Route header is present. Unified CM does update the route header properly using the received received record-route header. 
Call fails because DNS lookup for FQDN in Contact is unsuccessful.  The root of the problem is that the UCM should be using the value in the Record-Route header instead of that in the Contact, since both are present in the 18x response.
Conditions:Both Contact and Record-Route header are received in a 18x response without Rel1XX enabled and the FQDN specified in the Contact header cannot be resolved.

Workaround:Enabled Rel1XX on the SIP Profile of the SIP Trunk (may work, may not work).

however this can be confirmed by looking at full log analysis. Please open a TAC case for proper analysis.



New Member

Dialing Sx20 from CUCM through SIP Trunk to VCS.

If I am running into that known issue, is there a specific reason it is only affecting the Sx20 and not the 6000MXP which I can dial just fine? It does not seem to be model specific in the bug report.

Dialing Sx20 from CUCM through SIP Trunk to VCS.

Hi Steven,

yes, it not model specific but it can happen when the contact header and record route header both are present in the provisional response messages to cucm.

as i said open a tac case for the analysis, cucm folks should be able to tell you workaround if you are hitting the above said bug.



CreatePlease to create content