Pulumi Outage History

Pulumi Cloud experienced authentication failures for users signing in with Bitbucket identities, causing failed sign-ins and intermittent errors in the Console and API after session tokens expired every two hours. The issue was caused by an incident with Atlassian's Bitbucket OAuth service, not Pulumi's infrastructure, and affected Bitbucket-connected accounts and integrations while users with GitHub, GitLab, email, or SSO identities remained unaffected. The incident was resolved after approximately 11 hours when Atlassian deployed a fix for their OAuth service.

Pulumi experienced elevated API error rates and increased response times, particularly affecting ESC endpoints, for 2.2 hours. The incident impacted the API service, causing degraded performance for users attempting to access these endpoints. A fix was deployed that restored normal error rates and response times, resolving the issue.

Pulumi Cloud experienced a 4.5-hour outage where users authenticating with Bitbucket were unable to sign in or use the API, affecting both the console and CLI operations. Other authentication methods remained functional during the incident. The issue was identified, a fix was deployed, and service recovery was confirmed through monitoring.

Pulumi Cloud experienced reduced API response times for 14 minutes, affecting both the Console and API components. The issue was quickly identified and resolved, with normal service restored.

Machine tokens were unable to open referenced environments when running stack previews in Pulumi's API service. The issue affected users' ability to run stack previews using machine token authentication. The incident was resolved after 26 minutes through a rollback deployment.