Set up SIP failover across DIDHub PoPs

High-availability VoIP needs a primary and secondary SIP trunk. DIDHub publishes 23 geographically separated PoPs across 8 regions (NOAM, EURO, MENA, INDIA, APAC, ANZAC, LATAM, AFRICA) plus 200+ Global Edge anycast endpoints with consistent credentials. This tutorial wires up DNS SRV-based failover and active probing.

Option A: DNS SRV (simplest)

# DIDHub already publishes:
_sip._udp.sip.didhub.io.   3600 IN SRV 10 60 5060 us-east.didhub.io.
_sip._udp.sip.didhub.io.   3600 IN SRV 10 40 5060 us-west.didhub.io.
_sip._udp.sip.didhub.io.   3600 IN SRV 20 100 5060 eu-west.didhub.io.

Most modern PBXs (Asterisk pjsip 18+, FreeSWITCH 1.10+, Kamailio 5+) honor SRV records natively. Point your trunk at sip.didhub.io and failover happens automatically.

Option B: explicit primary + secondary trunks

Use this if your platform does not honor SRV (older 3CX, some hardware SBCs).

Configure outbound route → trunk sequence: primary, secondary, tertiary. On 503 / timeout, the next trunk is tried.

Option C: Kamailio dispatcher (high-volume)

See the Kamailio dispatcher tutorial or the OpenSIPS dispatcher tutorial — both give weighted load-balancing plus active SIP OPTIONS probing.

Active probing settings

# Asterisk pjsip AOR
qualify_frequency=30      # OPTIONS every 30s
qualify_timeout=5         # mark down if no response in 5s

# 3CX
# Trunk -> Options tab -> Re-register every 60s

# Kamailio dispatcher
modparam("dispatcher", "ds_ping_interval", 30)
modparam("dispatcher", "ds_probing_threshold", 3)

Real failover test

1. Place a 30-second test call.
2. Mid-call, block us-east.didhub.io at your firewall (simulating the PoP going down).
3. Place a second call — should connect via us-west.didhub.io within 5-10 seconds.
4. Unblock and re-test.