Failover and Retries

Configure failover and retry behavior on Telnyx SIP Trunking. Route calls to backup destinations on SIP failure codes for high availability.

Telnyx SIP connections automatically retry failed call attempts through different routes and IP addresses.

Signaling IP addresses

Telnyx uses two geographically redundant signaling IPs per region:

Region	Primary (IP1)	Secondary (IP2)
US	192.76.120.10	64.16.250.10
EU	5.172.39.10	5.172.39.25
Canada	193.108.220.10	193.108.220.25
Australia	103.135.104.10	103.135.104.25

Failover behavior

Single route

1. SIP INVITE sent from IP1
2. On failure, retry from IP2

Multiple routes

1. Attempt all routes via IP1 in configured order
2. On failure, retry all routes via IP2

Route order depends on configured preference (Sequential or Round Robin).

Credential authentication

Calls route through the registered KSS instance with three levels of internal failover.

Call forward on failure

When enabled, calls that fail on all SIP routes forward to PSTN (up to 10 termination carriers).

Response codes

Triggers failover:

Code	Meaning
408	Request Timeout
480	Temporarily Unavailable
503	Service Unavailable
504	Server Timeout
Transport error	Network/TCP failure

Does NOT trigger failover (call considered connected):

Code	Meaning
180	Ringing
200	OK (answered)
404	Not Found
486	Busy Here
603	Decline

DNS configuration

SRV records (recommended):

_sip._udp.example.com. 3600 IN SRV 10 10 5060 sip.telnyx.com.

Regional domains: sip.telnyx.com (US), sip-eu.telnyx.com (EU)

SRV records automatically resolve to both IP1 and IP2.

A records (alternative):

Configure separate A records for each signaling IP and add both as routes.

Configuration

PATCH /v2/ip_connections/:

{
  "default_routing_method": "sequential",
  "call_forwarding": {
    "forwarding_type": "on_failure"
  }
}