Degradation in some services

Update timeline

1. investigating Oct 20, 2025, 08:25 PM UTC

   We are seeing some 500 errors on our backend API service. Currently, our service appears to be alive in both the Europe and US-based backend regions, though more errors are happening on the US-backend.

2. investigating Oct 20, 2025, 10:18 PM UTC

   Our error rates for our core backend services have remained very low and within normal bounds for the past hour. We do still have some downstream dependencies that are unavailable, resulting on some of our offline processes queuing up information to send to these third parties once they are available. As a result, a few specific areas of our application may take time to recover to normal operations.

3. investigating Oct 20, 2025, 11:38 PM UTC

   We are seeing degraded performance on the EU backend, with requests timing out. We are still investigating the scope of impact.

4. investigating Oct 20, 2025, 11:49 PM UTC

   Requests to EU backend are no longer timing out after a restart of the multi-tenant Sombra clusters. Our core backend error rates for both EU & US regions continue to remain low and within normal bounds for the past hour.

5. resolved Oct 21, 2025, 05:05 AM UTC

   We've implemented recommended mitigations from our upstream vendors, which has allowed our systems to fully recover. Some Airgap.js telemetry information may encounter processing delays, as our systems catch up.