VoIP.info Troubleshooting

SIPTRUNK is registered but inbound calls or audio fail

Trace the selected authentication mode, registered Contact, provider attempt, PBX route and direct-media RTP without treating portal registration as end-to-end proof.

Reviewed 2026-10-19Current first-party SIPTRUNK device, registration, inbound-call, IP-authentication, failover, registration-status, interconnection and direct-media documentation was reviewed October 19, 2026. Credential registration and static-IP authentication remain different service paths; portal registration, Contact address, observed public source, inbound route, signaling gateways, direct-media RTP sources and failover treatments remain separate evidence. No SIPTRUNK account, trunk, credential, registration, number, route, gateway, DNS answer, PBX, firewall, packet, media stream, failover event, price, availability, compatibility or live call was inspected or testedAdvancedDocumentation based · not lab tested

Freeze one failed inbound call and one bounded control

Record timestamp and timezone, ordered trunk, authentication mode, DID, provider gateway, portal registration, Contact and observed source, provider attempt, edge receipt, PBX response, route destination, answer, SDP, RTP endpoints and user-visible audio. Classify the earliest gap: no current destination, wrong Contact, no attempt, blocked signaling, unmatched route, rejected call, no RTP or one-way RTP.

Redact credentials, numbers, addresses and call content. A working outbound call or green registration indicator is useful control evidence, not proof of inbound delivery or media.

Confirm the selected authentication and destination model

If the trunk uses credential registration, preserve the 401 challenge, authenticated REGISTER, 200 response, accepted expiry and current portal row. If it uses IP authentication, verify the stored public IP and port instead of diagnosing the intentional absence of REGISTER.

Do not enable both paths or change credentials, NAT and inbound routes together. Correct the selected service boundary first and retain a rollback.

Compare Contact with the public source observed by SIPTRUNK

For the registration path, compare the Contact IP and port with the alias or public source the provider reports. SIPTRUNK documents that inbound traffic is attempted toward Contact. A private, stale or remapped Contact can coexist with a successful registration and still send an inbound INVITE to the wrong place.

Confirm DNS, selected gateway, NAT ownership, port mapping and registration refresh. Do not guess a public address from a different node or time interval.

Trace the provider attempt into the loaded inbound route

Join the portal or support attempt to an edge capture by timestamp. Preserve source gateway, transport, Request-URI, To and From. SIPTRUNK documents an eleven-digit To user by default for inbound North American calls, while From normally uses ten digits. Compare the actual To with endpoint identification, context and the first loaded Asterisk or FreePBX match.

If nothing reaches the edge, remain at provider routing, Contact or network policy. If the PBX responds, retain the first final SIP result before changing destinations or number normalization.

Trace direct-media RTP independently of SIP

After answer, read the actual SDP addresses and ports and identify the first RTP packet in each direction. SIPTRUNK documents that RTP can arrive from addresses other than its signaling gateways. A firewall limited to gateway addresses can therefore pass SIP while dropping direct media.

Change only the demonstrated media rule within the approved security design. Do not expose an arbitrary UDP range merely because the provider article mentions varied sources; preserve segmentation, active-call validation, state, rate controls and the PBX’s configured RTP range.

Correct the earliest mismatch and test recovery separately

Correct only the first proven divergence: wrong service mode, expired binding, stale Contact, port translation, DNS or gateway choice, DID assignment, blocked SIP, endpoint identification, context, number match, SDP address or RTP policy. Repeat the same call and confirm answer plus both audio directions.

Then test only the authorized Secondary Trunk, PSTN Backup, PSTN Forward or voicemail treatment, recording trigger and restoration separately. Do not infer active-call survival, recovery time or future availability. No SIPTRUNK account, registration, PBX, firewall, packet, media stream, failover event or live call was inspected for this diagnostic.

Sources & applicability

Primary references for the technical details above. Operational examples and planning checklists are VoIP.info editorial guidance.

Examples require adaptation to your topology. No live PBX or hardware testing is claimed.