Voicemail does not respond on CUC 8.6.2 and CUCM 8.6.2 environment

[layhlaing]

Hi,

We started experiencing issues in which voicemail just went into silence without responding anything recently when we completed migration of sites from Unified Communications 6.0 to 8.6. I did not see any related message in RTMT or console on any of CUC or CUCM at the time and had to reset the SIP trunk to resolve the issue. In some cases, it was required to restart CUC servers to resolve the issue.

Followings are our 8.6 environment details:

CUCM: 8.6.2.22900-9 (One Publisher and two Subscribers)

VMware Installation: 2 vCPU Intel(R) Xeon(R) CPU X5570 @ 2.93GHz, disk 1: 80Gbytes, 4GB RAM

CUC: 8.6.2ES44.22900-44 (One Publisher and one Subscriber)

VMware Installation: 2 vCPU Intel(R) Xeon(R) CPU X5570 @ 2.93GHz, disk 1: 200Gbytes, 6GB RAM

Integration between CUC and CUCM: SIP integration

There are 399 voice-mail ports in our CUC cluster and ports 1 - 200 are on the subscriber and they are assigned to answer calls, and the next ports 201 - 300 are on the publisher and they are assigned for Message Notification/MWI dial-out. The last ports 301 - 399 are on the subscribers and they are assigned for Message Notification/MWI dial-out.

Can someone please share your knowledge if you have experienced similar to this in your environment? Thanks in advance.

Regards,

Lay

[Manish Prasad]

One possible reason i can see is MTP resource get exhausted , as per your previous post you mentioned that MTP required is checked on the SIP trunk between CUCM and CUC and the preferred codec is g729.

Try look for media resource error on RTMT when you face this problem.

As there is no limit for SIP trunk concurrent calls , its depend on server capability and location setting. You can also check location bw and see if not hitting the maximum limit.

[layhlaing]

Thanks Manish again for this.

I did get few MediaReosurceListExhausted events from RTMT alerts but was not paying much attention to them. Location bandwidth settings have been checked and increased a bit for sites which were repeatedly getting these events. I will monitor to see if that reduces these alerts. Could you please advise how I can verify it is hitting the maximum limit?

I have also raised this with Cisco TAC and some cores were found on the CUC subscriber server. We seemed to have run into a bug according to Cisco TAC as in here https://tools.cisco.com/bugsearch/bug/CSCug61581/?reffering_site=dumpcr and both CUC servers have been upgraded to 8_6_2_24901_1. It has fixed cores and we are now monitoring the voicemail service for few days to see if it will reoccur.

Regards,