Since approximately 06:35 UTC on March 15, 2026, our production service on US-East has had zero outbound connectivity. The container runs fine and serves inbound HTTP requests, but ALL outbound connections fail with ENETUNREACH and ETIMEDOUT.

Affected services (all failing simultaneously):

- MongoDB Atlas — MongoServerSelectionError: Server selection timed out after 20000 ms

- PostHog proxy — AggregateError [ETIMEDOUT] on every request to us.i.posthog.com

- All other outbound HTTPS connections on port 443

The app was operating normally through 06:29 UTC (cron jobs, database queries, Redis jobs all completing successfully).

First errors appeared at 06:35:38 UTC with no code changes or deployments.

Errors include both IPv4 ETIMEDOUT and IPv6 ENETUNREACH to multiple unrelated external IPs, suggesting this is a platform-level egress issue rather than a destination-specific problem.

Redeploying resolved the issue (new replica got working networking), but the original replica never recovered on its own.

Project: 251e0394-2884-4914-a75e-bed972d19609

Service: 63f25f0d-18f5-4cbf-bcff-cdd7258ab7ec

Affected replica: 75816190-bc77-48f5-8dc0-27e6b9afbe94

Region: US-East

This appears related to the EU-West egress outage reported by another user starting at ~06:45 UTC, suggesting a platform-wide issue across multiple regions?

Can we confirm if there are issues?