This incident will now move to resolved. The culprit seems to have been an issue in a Kubernetes service of a cloud provider. We will take due time to tripple check all details and confirm with a full RCA on Monday.

A fix has been implemented, and the service is up again. We will intensify monitoring for the next 20 minutes and resolve the incident with an RCA afterwards.

We are investigating connection issues between our voice services and HALO, resulting in a service disruption for our Voice Agents. Next update in 20 minutes, or as events warrant.