Avaya SIP CM 10.2 and IX messaging: SIP phone can't reach VM internally

[hebegb41]

Hello Sports fans. Today we have an Avaya SIP CM 10.2 riding on a Juniper L2 network. The Entire Avaya layout is SIP with a few analog phones for conf rooms etc. The entire Avaya infrastructure is VM on an ESXi host platform. Our issue is an odd one that reeks of certificate issues but we can't seem to find any that are bad or outdated/expired.
Here is the troubleshooting we have done. This site uses J100 series phones, just FYI. A week ago all was well.....this week.....this....

1. Calls from outside (via SIP trunks and ASBCE) to SIP phones are normal
2. Calls from outside to SIP phones that cover to VM get VM and hear prompts and can leave messages
3. Users can call in from outside (i.e. via cell) and access their mailboxes and retrieve messages
4. Calls direct to VM pilot from internal SIP phones appear to get dead air but call stays up until they hang up.
5. Calls from SIP phone – SIP phone that cover to VM get an initial tone indicating a cover to VM … then same dead air until hang up.
6. Calls from an analog phone direct to VM pilot connect and hear all prompts and can leave / retrieve messages (i.e. normal ops).
7. SIP traces on phones and between SBC's and or CM to IX messaging all look normal. No real indication of anything wonky.
8. We are TLS over port 5061 from the SM SIP entities/links to the IX messaging system. This is what prompted us to look at certs as possible issue.
9. We have rebooted and or restarted all systems involved and denied/allowed new service etc still no fix.
   
   So grab your cup of Joe and tell me what is up with this issue and how stupid I am.....

 

[Sean655]

Have you done a traceSM or Wireshark on IX yet? This could easily tell you what is happening. I recently found an issue with SRTP by seeing a flood of reINVITE messages during a call and I found a certificate issue on IX by running Wireshark on the IXM server.

 

[dialtonebone]

Start a traceSM, filtered to your station. Then call VM and analyze the results, or share them with us.

 

[hebegb41]

Have you done a traceSM or Wireshark on IX yet? This could easily tell you what is happening. I recently found an issue with SRTP by seeing a flood of reINVITE messages during a call and I found a certificate issue on IX by running Wireshark on the IXM server.

yes on traceSM no on wireshark on IX. We are doing that now

 

[hebegb41]

Start a traceSM, filtered to your station. Then call VM and analyze the results, or share them with us.

yep, done. Nothing of note there. all call processing looks normal

 

[Shaun E]

Are there any firewalls in the path that might be blocking RTP traffic?

Signalling will flow via Session Manager but RTP will be direct between IX and your endpoints (ie: the J100 series phones)

If IX is on the same network segment as your media gateways and SBC then that'll probably explain why analog and external calls work. The same applies for outside calls that cover to IX - the RTP audio will flow directly between the SBC and IX.

If it was a TLS issue with signalling then I wouldn't expect the call to connect in the first place but a successful connect and dead air typically means RTP.