Our client is a US-based online financial advisor and personal wealth-management company Headquartered in California, with offices in Colorado. Its primary offerings include providing free and paid wealth management products.
The company has CISCO UC servers at two different locations. One location was hosting primary servers and the other location was hosting the secondary servers. The reported issue was experienced in their primary location.
Customers were not able to reach voicemail when calls were not answered. End users were unable to reach their own voicemails to check messages.
During testing, we found that when calls were unanswered, instead of reaching the voicemail the calls were disconnected immediately. Similarly, when a user tried to test and check their voicemail, the call would get disconnected. We verified the database replication, and checked the voicemail ports’ status and found no issues.
We performed a few tests, including downloading the traces from the call manager and unity connection servers, and found that there was an issue with the name-to-IP translation, therefore it was a DNS issue. We verified the SIP trunk connection between CUCM and CUC and found it was configured using a hostname.
The SIP trunk between CUCM and CUC servers was re-configured using the IP address instead of the host name. After re-configuring the SIP trunk, a reset of the trunk was performed, after which the voicemail functionality started working.
The provided solution – changing the domain name to IP address in the SIP trunk between CUCM and CUC servers – will never affect the connectivity between both the servers, even if the domain name is changed in the future.