Some 7941 phones can't find the corporate directory since upgrading to CUCM 184.108.40.20600-8. Phone load is SCCP41.9-0-3S. Forum is loaded with similar problems. I've tried some of the solutions suggested none of which have corrected the problem. It seems that the phones in question are being seen as 'secure' phones because they are picking up the Secure Phone URL's from Enterprise params. I can't see why this is happening all phones are set to use the Standard SCCP non-secure profile. The only difference I can find between 7941 that were working and ones that weren't is the URL pointers. I tried changing the secure Enterprise params to the same as the non secure Enterprise params. The phones picked up the new URL but still couldn't find the corporate directory (host not found). Guess I could try deleting the phone and recreating it but that would leave me not knowing what caused it if it's resolved by this action. Other than that I'm running short on ideas.
I've attached the latest phone log from one of the lost 7941. Seems it's not finding the proper .CTL file?
I hope you are doing great,
Would you mind confirming me the following information:
Is the TVS service running on the node to which the phone is registered to?
Are you getting a “Node not trusted” alarm?
Is the phone having a valid ITL file?
The TVS service it is a network service on CUCM 8
Cisco Trust Verification Service
To verify that you need to go to:
In the ITL file you will find the
TVS with the FQDN of your cucm.
and also the TFTP server with the FQDN.
if you do not find this, delete the ITL file and restart the IP phone.
From my experience the problem is usually the TVS entry in the TFTP file the phone gets. On CUCM 8.x the services provisioning method is set to internal not external so newer phones like the 7941 won't look at the service URL so having an IP there vs a hostname doesn't do anything for those phones. You can check that you have a hostname in the TVS entry by downloading a phone's config using "tftp x.x.x.x get SEP0123456789AB.cnf.xml" view it in note pad and search for "TVS", you'll likely see a hostname, which is why you're getting the host not found if your phone isn't configured for DNS or cannot resolve that name. The fix is to change under System > Server, the hostname to an IP address there, save it, then restart th Trust Verification Service (TVS), possibly TFTP (I don't remember if this is necessary but won't hurt). Then if you download the phone's configuration file again it should have an IP address instead of hostname in the TVS section and your phone should be able to access your directory.
Gentlemen thank you for your responses.
At the time I originally posted the problem it was only a handful of phones that were experiencing this issue. Since that time it became 2 fold; It is all relative bear with me and read on. I was getting new reports of some phones also having an issue with the message button (voice mail) resulting in a 'no service configured' response when pressing it. Initially one of the Admins reported it when they created a new phone. All of this of course was occuring since the upgrade. It all quickly escalated once one of the sites had a power outage (over 200 7941) causing all phones to reload. At that point we had resolved the issue of the corporate directory getting the message 'host not found' but the messages key 'no service config'd remained. Several things were done that may have resolved the 'host not found' problem; The Enterprise Services param was changed from 'interna' to ''external'; The TSV was also restarted and in some instances factory defaulting the phone causing them to reload their images.
I had never changed or set these params prior to the upgrade nor the Enterprise URL service params which pointers were set to IP not DNS. The Messages URL of the same params was blank keep this in mind. (I'm assuming they were set by default somehow) . Confused? Sorry read on...
Now we had these 200 phones with no functioning message key. They could dial the voice pilot directly to retrieve their voice mails as a bandaid while we scrambled to figure out what was going on. I didn't wait long before calling TAC again. TAC now reset the Enterprise Service Provisioning param to 'both' this gave the message button it's normal function back.
I definitely need to do more research on all of this functionality but as a result I have gained a better understanding of it all.
Remember the Enterprise URL param for messages was blank at the start of all this and still is; Consider these notes in the case from Engineer I worked with on the second ticket I opened. He was inquiring to his team of this functionality. Remember there were 2 tickets the first for the 'host not found' error on the corporate directory the second on the message button getting 'no service configured'; The first ticket we reset the Services provision from 'internal to 'external' the second ticket it was set to 'both' but never back to 'internal' as the Engineer eludes to:
My Customer has the "Services Provisioning" parameter set on his phone
as External URL. This result that the Messeges button to responds with
"no services configured" when pressed.
When the "Services Provisioning" parameter is set to Internal or Both,
the Messeges button calls the VM Pilot as expected.
My Question is: Is this working as designed?
The response he got back:
think it probably works as designed, please check the help content in CUCM of ��Service
If you do not configure the external url of ��Message��, and configure ��Service
Provisioning�� as External, it will show ��no services configured��.
Please check it
The Enterprise messages was blank and still is!!!
Hope I'm not confusing anyone with all this; I just thought it may help someone sort all of this out in the future. The response that were posted here were close to the mark I thank you for them.
With Services Provisioning set to Internal, the phone will receive its subscribed phone services from the phone's configuration file and store these (and their corresponding service URLs) in flash. This allows the phone to access the service URLs directly on a web server without first querying the Cisco CallManager IP Phone Service. With Services Provisioning set to Internal, the Corporate and Personal Directories default services also have an extra level of redundancy built into the phones. When these services are selected, the phone will attempt to send an HTTP message with the proper URL string to the Unified CM with which it is currently registered. Therefore, the Unified CM Group configuration of the phone's device pool provides redundancy for these services.
With Services Provisioning set to External URL when a user presses the Services button, an HTTP GET message is sent from the IP Phone to the Unified CM getservicesmenu.jsp script by default (step 1). You can specify a different script by changing the Phone URL enterprise parameter. The getservicesmenu.jsp script returns the list of phone service URL locations to which the individual user has subscribed (step 2). The HTTP response returns this list to the IP Phone (step 3). Any further phone service menu options chosen by the user continue the HTTP messaging between the user and the web server containing the selected phone service application (step 4).
This is from the SRND http://www.cisco.com/en/US/docs/voice_ip_comm/cucm/srnd/8x/cmapps.html#wp1184498
Let me know if this answers your question.
PS: Rate the post if it helps
It certainly helps having that information but I think I should have clarified my issue. It was brought to my attention today that the corporate directory wasnt working. The first thing I did was check the directory on my phone (9971) and it came up fine. Next I checked on my 6941 and it came up fine. Then I checked on multiple 7941's and they are getting "host not found". The only thing we changed in the week was adding third part certificates to our CUCM cluster for tomcat. The directory and service URLs in Enterprise Services are all correct.
I guess I am searching for why this just started happening out of the blue and what else there is to check? Seems like the concept is simple enough...
Can you collect the console logs from a non-working 7941.
Just to be sure, you are not using DNS right?. Can you also collect a wireshark packet capture from the 7941. SPAN or SPAN to PC ==> YES on CUCM.
Follow the procedure from this post
PS:Rate the post if it helps.
Going back to my earlier post, since the tomcat certificates have been changes the Trust Verification Service (TVS) in 8.x will come into play as with any phone service in CUCM 8.x using HTTPS to validate the certificates. The certificates are not stored on the phone so when the phone makes an HTTPS request, and it's presented with a certificate the phone queries TVS to see if it can trust the certificate based on what call manager trusts. The call manager will reply with an accept or deny for the certificate, if it's an accept the phone will connect using HTTPS. The host not found many times is the TVS entry in the config file. If you download the config file for the non working phone and look for the TVS entry you'll either see a hostname or IP address. Host not found usually indicates that the TVS entry is a hostname. This is populated from System > Server.