BYOC Trunk Failover Breaking ServiceNow Webhook Context

Stuck on a context persistence issue during SIP re-INVITEs on our BYOC trunk.

We have a robust integration where Genesys Cloud Architect flows trigger Data Actions to create tickets in ServiceNow via REST API. Under normal conditions, the cust_ref_id and conversation metadata pass cleanly through the webhook payload. However, when the primary SIP trunk fails over to the backup provider, the session seems to reset. The subsequent re-INVITE invalidates the Scripting API v1 variables, causing the Data Action to send a malformed or empty payload to the ServiceNow endpoint. This results in a 400 Bad Request because the required fields are missing. The environment is running Genesys Cloud 2024-08, and the ServiceNow instance is on the Washington DC release. We have verified that the OAuth tokens remain valid, so this is strictly a data loss issue during the failover event. Is there a mechanism to persist these variables across the SIP re-INVITE boundary, or should we be implementing a separate webhook listener to cache the data before the trunk switches? We need a reliable way to ensure the ticket creation logic survives the network transition without manual intervention.