Freeze one failed transaction and the ordered mode
Record the timestamp and timezone, Australian account and trunk identifiers, service order, point of interconnect, operation mode, PBX release and driver, loaded target, registration result, direction, dialed and normalized number, caller identity, response code, answer state, SDP, first RTP packet in each direction and user outcome. Redact secrets and personal numbers.
Classify the first divergence as wrong mode, no request, no response, authentication failure, wrong SIP URI, rejected INVITE, number not routed, caller identity rejected, channel refusal, no alert, no answer, no media, one-way media or later degradation. A completed REGISTER ends only the registration branch.
Prove the interconnect and current target
For registration, compare the loaded issued server, username owner, Contact, expiry and complete response sequence. For URI diversion, compare the configured remote number, server and port with DNS and the observed incoming transaction. For Teams gateway, preserve the Direct Routing and trunk boundaries separately.
Do not migrate between modes while diagnosing one call. Confirm whether the service uses public internet or an ordered private IP path and identify the customer, access-provider and VoIPline demarcations before changing ACL, NAT or routing.
Trace numbers, caller identity and channels independently
For outbound failure, retain normalization, Request-URI, From and asserted identity, selected route, provider response and concurrent-call state. For inbound failure, identify the active or ported number, provider route, delivered Request-URI, PBX rule and destination. The service schedule restricts presented caller identity to valid allocated numbers controlled by the customer or end user.
A number preconfigured for a future port, a completed port, an active registration and an available channel are different states. Correct only the earliest demonstrated identity, authorization, routing or capacity mismatch and repeat the same permitted call.
Compare SDP and both packet directions
Preserve the offer and answer, selected codec and packetization, DTMF method, advertised media addresses, NAT rewrites, first and last packet in each direction, loss, jitter and decoded result. Compare the negotiated call with the service schedule’s documented codec boundary; do not enable every codec or open every port to conceal one mismatch.
No RTP, packets in one direction, packets delivered to the wrong interface and packets decoded but not heard are different outcomes. A provider network claim or status page does not identify the fault in this call.
Retest narrowly and remove temporary exposure
Repeat one authorized inbound and outbound call, confirm intended identity and number route, answer, two-way audio, DTMF and clean release, then compare one negative control. Confirm the active channel count, remove temporary captures and broad rules, and restore monitoring and escalation.
Do not weaken authentication, publish credentials, bypass caller-identity requirements or use an emergency number merely to clear an alarm. No account, interconnect, registration, number, port, firewall, packet, media stream, provider incident or live call was inspected or tested for this diagnostic.
Sources & applicability
Primary references for the technical details above. Operational examples and planning checklists are VoIP.info editorial guidance.
- VoIPline Telecom SIP trunking ↗
- VoIPline Telecom pricing and rates ↗
- VoIPline Telecom legal documents ↗
- VoIPline Voice and SMS (IP) service schedule ↗
- VoIPline SIP trunks critical information summary ↗
- VoIPline Telecom network ↗
- VoIPline Telecom number porting ↗
- VoIPline Telecom service status ↗
Examples require adaptation to your topology. No live PBX or hardware testing is claimed.