Subject: Deploys fail at container creation in US West (build OK) — service stuck, broken egress

Service: SAFEVENDAS — project "beckend safevendas", service 89d72b67-8dce-4e9a-89f9-fb639ed16966, region US West, attached volume "safevendas-volume" (/data, 141MB, holds our production SQLite DB).

Since the "AWS Cloud Workloads Failures" incident (Jun 11), EVERY deployment fails at the "Create container" / start stage, even though the build completes and the image is pushed. The status page now shows "Fully Operational", but our service still cannot schedule a new container.

The currently-running container (deployed Jun 11 17:58 UTC, before the incident) still serves inbound requests, but its OUTBOUND network is broken — from inside the container, every external host times out (tested google.com, challenges.cloudflare.com, gov endpoints). This breaks our login (Cloudflare Turnstile) and all outbound integrations, and it won't self-heal — a fresh container is needed.

Failed deploys (all "build OK, container creation failed"):

• 7dac5874-2c1e-4275-a981-e8857a7efa2a (Jun 12 05:28 UTC)
• 6f178ea4-49a7-496c-b91a-4868d0f264e1 (Jun 12 01:12 UTC — reached DEPLOYING then failed)
• 8c003f93, 2607483b, 6518043c, 746b2c57 (Jun 11 evening)

We CANNOT remove the volume (production DB). Please check the scheduling/storage health for our service's host in US West, or migrate the service to a healthy host. Thank you — this is impacting production logins and fiscal document emission.