CUCM and CUC 10.5 SIP Trunk Configuration

[sdavis1980]

Hi everyone, I'm troubleshooting an issue with an environment that has 2 CUCM, 2 CUC and 5 Gateways for different locations.  We had an issue where the publisher server went down and when making calls to 1 specific site, it was RNA. All other sites appeared to be working. All calls for each site go to a SCH in Unity and all sites but 1 were ringing to the correct SCH.  Device Pool has both servers, identical for each site. Regarding the SIP trunk from CUCM to CUC, are there any issues with having 1 SIP trunk with the destination address for both Unity servers in the 1 trunk or do I need a dedicated trunk for each Unity server along with route list?

thanks in advance.

[Terry Cheema]

Can you explain little more which pub went down CUCM or CUC - if you can explain your scenario bit more.

Regards to your query the recommended way to setup integration is to use route list/route groups with two SIP trunks.

Refer here:

http://www.cisco.com/c/en/us/td/docs/voice_ip_comm/connection/10x/integration/guide/cucm_sip/cucintcucmsip/cucintcucmsip060.html#pgfId-1132738

Make sure you have added all the CUCM servers in CUC as well and also on CUCM you can tick the checkbox run on all active nodes.

-Terry

Please rate all helpful posts

[sdavis1980]

Hi Terry, thanks for the input.  The CUCM publisher and CUC publisher is what went down (server that runs the VMs went down at this site), however all phones are registered to the CUCM subscriber at an alternate site. I have the SIP trunk check box checked to run on all active nodes and both CUCM ip addresses are listed in Unity.

 You think having 1 SIP trunk that points to destination for both CUC servers may cause this type of issue?  I haven’t had time to take down the publisher to replicate and diagnose, however I will be doing that this weekend. Just trying to prepare for all potential solutions in the mean time.

thanks,

Steve

[Terry Cheema]

So the issue is for one site, if you call them call fwd no answer never goes to voicemail and instead keeps on ringing?

Technically it should work with two destinations I think there may be something else as well...

But you can still change the config to recommended method and try again. Lets see what others have to say on this.

-Terry

Please rate all helpful posts

[Ayodeji Okanlawon]

Hi Terry I toally agree with you, using two different trunks is much better.

There are a few things to consider in this scenario

1. Outbound transport protocol used on the sip trunk. TCP or UDP

2. If its UDP, then you need to look at your SIP retry INVITE and timers. By default your retry INVITE is set to 6, so failover takes forever ( frommemomry I think CUCM retry timer is set to 500ms)

Ofcourse it its TCP, then its different game entirely as CUCM will not initiate an INVITE to that trunk unless a TCP connection has been established.

[Kasiraman S]

Its good to consider enabling SIP OPTIONS PING under SIP Profile if not already, which helps not to route the call to the node which is considered to be down.

Use OPTIONS Ping to dynamically track the state of each destination IP address on each SIP trunk and the collective state of the trunk as a whole. If a destination address is unreachable, Unified CM will not extend calls to this device. When all destinations are unreachable, the SIP trunk is considered to be out-of-service.

Thanks,

Kasi