Skip to content
Phone Guides

Asterisk troubleshooting: the first checks

Start with one failed call and identify the first stage that behaves differently from the intended path. Repeatedly changing authentication, NAT and routing settings together makes the evidence harder to interpret.

Source-reviewed, 18 October 2026. Prepared from the cited documentation. Commands and examples have not been independently tested by Phone Guides. Scope: Asterisk 22/PJSIP; command availability depends on loaded modules.

Write a small incident record

Record the time and timezone, originating account, destination, expected outcome, observed symptom and recent change. Keep sensitive numbers and subscriber details private. State whether the problem affects every call, one direction, a single endpoint or a provider route.

Stage Initial question Useful evidence
Process Is the intended Asterisk instance running? Service state, version and startup errors
Registration Does the intended AOR have a contact? Endpoint/auth/AOR settings and registration exchange
Route selection Which context and extension received the call? Loaded dialplan and call execution log
Outgoing setup Did Asterisk attempt the expected endpoint or trunk? Destination channel and SIP response sequence
Answer/media Did both parties answer, and can each hear the other? SDP addresses, codecs and RTP path
Termination Which participant ended the call, and did the other clear? SIP termination and channel cleanup

This separates a missing route from a missing contact, and an answered call from a working media path. The upstream PJSIP troubleshooting guide provides detailed symptom checks.

Inspect the running state

At the Asterisk CLI, start with commands relevant to the failed stage:


core show version
module show like pjsip
pjsip show endpoints
pjsip show contacts
pjsip show registrations
dialplan show pg-internal
core show channels

Substitute your actual context for pg-internal. Local phones registering to Asterisk are inspected through contacts; pjsip show registrations addresses Asterisk's outbound registrations. Do not declare local phone registration broken because the latter list is empty.

Inspect a single object with pjsip show endpoint 6001 or pjsip show aor 6001 when investigating our example accounts. Compare loaded state to the file/database that owns it. A typo in an include, an unloaded module or a reload rejection can leave the old object active.

Use the CLI help guide to find the commands supported by your installation. Avoid destructive commands while collecting the first evidence.

Collect a bounded trace only when needed

For a controlled reproduction, the PJSIP logger can show SIP traffic:


pjsip set logger on

Reproduce one call, preserve the relevant evidence securely, then stop the extra logging:


pjsip set logger off

Signalling logs can contain identities, numbers, addresses and authentication material. Do not post an unredacted trace publicly. Keep the time window short and follow your site's access/retention rules. The official logging overview describes logging destinations and runtime controls.

Interpret the sequence

One authentication challenge is not proof of a failed account. Look for the authenticated retry and final outcome. One SIP status without the request, direction and call stage can also mislead: a response from the phone is a different observation from the same status returned by the provider.

If the intended route executes and both parties answer, investigate audio separately using NAT and RTP. If the call never reaches the intended route, inspect contexts and patterns first. Use the SIP response code lookup to interpret a code in context.

Change one justified setting, preserve its previous value, and repeat the same call. Record the outcome even if it is unchanged. If the change does not help, return to the known configuration before trying a different explanation.

Return to Asterisk documentation.