Hi,
we have a little issue with some phones, which is very annoying.
We are using following hardware/software:
- FreePBX Distro (Version: 13.0.54; Core Version: 13.0.42; Asterix Version: 11.20.0)
- snom 320 (Firmware: snom320-8.7.5.35-SIP-f.bin)
- snom 370 (Firmware: snom370-8.7.5.35-SIP-f.bin)
We have one external datacenter with IPfire and FreePBX and multiple office locations with DrayTek and FRITZ!Box as router.
Between all offices and the datacenter we don’t have any IPsec/LAN-to-LAN connection. All phones uses the PBX domain as registrar, which points to an external IP address mapped on the IPfire. IPfire has NAT/Forwarding rules for following ports to the FreePBX server:
- 5060/UDP
- 10 000/UDP - 20 000/UDP
I’ve already tried the IP instead of the domain for the registrar, but this doesn’t improved or fixed it.
Network information:
- Datacenter: 10.40.1.0/24
- Office A: 10.1.1.0/24
- Office B: 10.20.1.0/24
- No LAN-to-LAN / IPsec connection anywhere
Anyway, at all locations we have some phones with that problem, that they are sometimes not reachable until I click on “Re-Register” in the webinterface or just reboot the phone. Not all phones are affected by this problem.
In the log of one snom phone, I could find following very often:
Feb 9 11:38:13 [ERROR ] PHN: RTCP: invalid port RC4 for stream
Feb 9 11:38:13 [ERROR ] PHN: RTCP: invalid port RC4 for stream
Feb 9 11:38:13 [ERROR ] PHN: RTCP: invalid port RC4 for stream
Feb 9 11:38:13 [ERROR ] PHN: RTCP: invalid port RC4 for stream
Feb 9 11:38:18 [ERROR ] PHN: RTCP: invalid port RC4 for stream
Feb 9 11:38:18 [ERROR ] PHN: RTCP: invalid port RC4 for stream
This was also found in the logs:
Feb 9 12:04:51 [WARN ] SIP: process_registrar_packet: 401 needs 128 bit nonce
Feb 9 12:04:51 [NOTICE] SIP: process auth: Match challenge for user=024, realm=asterisk
Feb 9 12:04:52 [ERROR ] PHN: Wrong Dst values: .
Feb 9 12:08:03 [NOTICE] PHN: TPL: Socket 60 idle/connect timeout
Feb 9 12:08:08 [NOTICE] PHN: TPL: Socket 61 idle/connect timeout
Feb 9 12:10:07 [CRITIC] PHN: Warning: Ignore invalid parameter: REREGISTER
Feb 9 12:10:07 [WARN ] SIP: process_registrar_packet: 401 needs 128 bit nonce
Feb 9 12:10:07 [NOTICE] SIP: process auth: Match challenge for user=024, realm=asterisk
Feb 9 12:10:07 [WARN ] SIP: process_registrar_packet: 401 needs 128 bit nonce
Feb 9 12:10:07 [NOTICE] SIP: process auth: Match challenge for user=024, realm=asterisk
Feb 9 12:10:08.001 [NOTICE] SIP: Registration Metrics failed
Feb 9 12:10:08.001 [NOTICE] PHN: TPL: Socket 62 idle/connect timeout
I’ve already searched for “RTCP: invalid port RC4 for stream”, “Wrong Dst values” and “process_registrar_packet: 401 needs 128 bit nonce”, but I couldn’t found anything (helpful) related to this.
Due I’m not able to call these failed pones internal by dialing the extension number, I think it isn’t a trunk issue, why I won’t post these settings yet. If you need them, I can post them, if needed.
Are there any special configurations for NAT needed? Do I need to change something on IPfire? What is/are the causes for this issue?
If you need any further information, don’t hesitate to ask me for.
I hope, somebody can help me out. Thanks in advance!
Best Regards,
Sebastian