cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
507
Views
0
Helpful
4
Replies

CUBE DNS SRV Record for H323 Trunks to CUCM

DaveBarilla
Level 1
Level 1

Hi,

Is it possible to use a DNS SRV records in CUBE for H323 Trunk redundancy?

I have seen it posted in other discussions that DNS SRV are used for VCS H323 trunks and endpoint registration. The service format for call signalling is as follows:

_h323cs._tcp.company.com. srv 50 0 1720 trunk1.company.com.

_h323cs._tcp.company.com. srv 60 0 1720 trunk2.company.com.

So my theory is that I should be able to configure the same service record on CUBE using IP Host commands and then using the dns srv as the session target in the dial-peer. Much like one would do with a SIP trunk.

Config example:

ip host trunk1.company.com A.B.C.1

ip host trunk2.company.com A.B.C.2

Ip host _h323cs._tcp.cucm.company.com. srv 50 0 1720 trunk1.company.com.

Ip host _h323cs._tcp.cucm.company.com. srv 60 0 1720 trunk2.company.com.

dial-peer voice 100 voip

 tone ringback alert-no-PI

 destination-pattern .........

 session target dns:cucm.company.com

 voice-class codec 1

 voice-class h323 1

 dtmf-relay cisco-rtp h245-alphanumeric

Can anyone confirm weather this will work? Is it supported? 

Thanks,

Dave

4 Replies 4

Rajan
VIP Alumni
VIP Alumni

Hi Dave,

I have configured the same way while testing Expressway in lab. Configured DNS SRV records in the Voice Gateway and pointed its IP as the DNS server on the required PC. It was working.

You need to make sure that if any other device which needs to resolve this will have this CUBE IP used as its DNS server.

HTH

Rajan

Pls rate all useful posts

Hi Rajan,

Thank you for the quick response. I'm wondering will this work for trunks originating from the CUBE? My setup is PSTN PRIs in, H323 Trunk out to CUCM

PSTN-----PRI--->CUBE---H323--->CUCM

Thanks,

Dave

Hi Dave,

It should ideally work as long as you enable DNS on the CUBE.

Thanks 

Rajan

Chris Deren
Hall of Fame
Hall of Fame

What version is your IOS as if it's relatively newer you can just use "voice class server-group" feature  preferred over SRV.