Is there a clean way to configure retry logic for predictive dialer campaigns when our 15 BYOC trunks in AP-Southeast-1 are experiencing intermittent SIP 408 Request Timeout responses from the carrier gateway?
We are running a high-volume outbound campaign using Genesys Cloud Predictive Dialing, targeting approximately 2,000 calls per hour across these specific trunks. The issue manifests as a sudden spike in failed call attempts, where the dialer reports the call as “No Answer” or “Network Error” immediately after the initial INVITE is sent. Upon inspecting the SIP trace logs via the Admin console and our own monitoring tools, we see the carrier returning a 408 timeout roughly 3-5 seconds after the INVITE, without any subsequent ringing or connect events.
The environment details are as follows:
- Region: AP-Southeast-1 (Singapore)
- Trunk Type: BYOC (Bring Your Own Carrier)
- Carrier: Local Tier-1 ISP with SIP trunking
- Genesys Cloud Version: Current stable release (patch level 2023-Q4)
- Outbound Strategy: Predictive, with a configured answer rate of 85% and a max concurrent call limit of 150 per trunk.
We have already verified that the SIP registration status for all 15 trunks is stable and online. There are no obvious network packet drops on our side, as confirmed by our local firewall logs. The carrier has stated that their gateway is functioning normally and is not rejecting calls based on rate limiting at this volume. However, they mentioned that they do not support immediate re-INVITE or rapid retry sequences, suggesting that the client side should handle the backoff.
In Genesys Cloud Architect, we have attempted to adjust the “Call Failure” handling, but the predictive dialer seems to treat the 408 as a hard failure, removing the contact from the queue immediately rather than retrying. We need to know if there is a specific configuration in the Outbound Campaign settings or the Architect flow that allows for a retry on SIP 408 errors before marking the contact as unreachable. Additionally, are there any recommended timeouts or retry intervals that balance carrier compliance with efficient dialing rates? We want to avoid being flagged for aggressive dialing while still maintaining a high contact rate.