Sign-in, the app and the API are fully restored and have been stable since 19:18 UTC.

We are now gradually resuming workflow and AI agent processing. Queued runs will catch up over the next while, so you may see some delay before they complete. We'll resolve this incident once background processing is fully back to normal.

A fix has been applied and database performance is back to normal. Sign-in and the app should be working again.

Workflow runs and AI agent runs are temporarily paused while we safely restore background processing; queued work will resume shortly and nothing has been lost. Some AI agent runs that started during the incident may have failed and can be re-run.

We're monitoring and will post another update once background processing is fully restored.

Since 18:05 UTC, users have been unable to sign in to Cargo, and parts of the app and API have been slow or unresponsive.

We have identified the cause: a sudden surge in AI agent activity triggered an inefficient database query, which overloaded our primary database. Sign-in depends on that database, so logins timed out. Our authentication provider is not affected.

We are applying a database fix now and have a permanent code fix ready to deploy. We'll post another update once sign-in is restored.