CUCM 14(SU2) Host Not Found errors

[salvage210]

Hi all,

I have a lab environment with a cluster of one Pub and two Subs. A couple of 8841 phones and 7975G phones.

The 8841 phones and one 7975G phone can access the Corporate directory and User Preferences such as wallpapers and ringtones. All of the other phones present the Host not Found error and will not display the Corporate Directory or download wallpapers. All phones are registered to the same Subscriber.

• All phones show the URL for the corporate directory and services when I check their configuration
• The URL under "Enterprise Parameters" matches what the phones are showing me
• DNS seems to be working because I can browse to the Call Manager using the URL
• I tried replacing the URL with the IP address with no change.
• I refreshed the phones' security certificates
• I've restarted TVS, TFTP and Tomcat

I'd appreciate your thoughts.

[salvage210]

I should clarify:

• In the lab, I upgraded the cluster from 11.5.1 to 14.0.1(SU2) on TLS v1.1. All phones worked normally.
• I then was directed to implement TLS v1.2. I made a few test calls successfully but did not check ALL features of ALL phones.
• A couple of weeks later, we observed that the Corporate Directory stopped working. We don't use the lab call manager frequently.
• I opened this discussion after reviewing many old threads on the Host Not Found issue.

I mistakenly believed that perhaps a co-worker had done something to disrupt the network or call manager. I now understand that it's the TLS v1.2 that seems to have broken these features.  I'm regressing the cluster back to TLS v1.1.

I've also advised my management that it's really time to upgrade from these 79XX series phones which are end of support as of June 2023.

[Roger Kallberg · Accepted solution]

The default setting for TLS is this.

admin:show tls min-version
Configured TLS minimum version: 1.0

This is without a shred of doubt where you started off before you touched the version for TLS and set it to 1.2. Please revert it back to this and then test.

[salvage210]

Hey Roger,

This was the fix. I would have sworn that the cluster was starting at TLS v1.1 so that was what I regressed back to, but obviously I was suffering from a faulty memory. Reverting back to TLS v1.0 cleared the problem.

As an aside, in all of the old "host not found" threads that I found, none of them mentioned TLS as being the problem.  Thank you so much for your patience.