We have deployed the fix and confirmed resolution

This incident is now resolved. Between 14:40 UTC and 18:53 UTC on 12 October 2026, some users may have experienced intermittent HTTP 503 errors affecting package downloads.

We have applied a fix, and after monitoring, we confirmed normal operation for download requests.

This incident is now resolved. Between approximately 12:00 UTC on 3 October and 07:30 UTC on 5 October, customers in Europe (eu-west-1) experienced intermittent failures when pushing NuGet packages.

We have applied changes to the affected services and confirmed that package pushes have returned to normal.

Pulling and downloading packages was unaffected throughout, as were all other package formats.

This incident is now resolved.

We have confirmed that the API has returned to normal operations.

We have resolved the issue with pulling Docker images

We have now release a fix to resolve this issue. Our OIDC service is operating normally

The root cause was identified, and the issue is now resolved.

The root cause was identified and the issue is now resolved.

OSV advisory backlog processing has completed, and returned to normal processing at 07:00 PM UTC. All systems are now operating normally.

This incident is now resolved. Between 19:16 and 20:11 UTC, customers experienced request latency and timeouts affecting package downloads and API requests in the US East region. We have confirmed that these services have now returned to normal operations.

We have confirmed that all services are operating as normal.

We have confirmed that our services have returned to normal operations.

OSV advisory processing resumed at 23:50 UTC on August 25, and advisory freshness returned to normal at 04:35 UTC on August 26. All systems are now operating normally.

The issue has now been resolved, and package caching to the repositories has resumed.
Packages pulled between 17:00 and 18:45 UTC will be missing from the repositories. These missing packages will be cached during subsequent pulls.

Note: Package caching to the repositories was impacted. Package pulls remained unaffected throughout the incident. Webhooks subscribed to the Package Downloaded event were also impacted during this period, and missed events will not be replayed.

Scope has been determined to be non-production only, so the incident is marked as non-impacting.

No further impact to API or Web applications outside of the 1 minute window

We have confirmed that our work has returned services to normal operations.

We've reverted a change and confirmed resolution.

This did not impact Enterprise (custom) domains or Enterprise Customers with custom domains.

Packages have successfully resynced and package syncing is operating as expected

We have confirmed that our change has returned stability to the service