Pulumi 장애 기록

1. 2026년 8월 17일 오후 2:05 UTCinvestigating

   GitHub는 적극적이고 중요한 사건을 보고했습니다. Pulumi Console의 관련 기능은 일시적으로 사용할 수 없습니다.

2. 2026년 8월 17일 오후 2:30 UTCinvestigating

   https://www.githubstatus.com/incidents/zkxwbgr0cnmx

3. 2026년 8월 17일 오후 5:06 UTCresolved

   이 사건은 해결되었습니다.

공식 인시던트 업데이트를 자동 번역했습니다.

1. 2026년 7월 24일 오전 2:56 UTCidentified

   우리는 일부 팀 액세스 토큰이 Pulumi Cloud API에 인증 할 수없는 문제가 발생하고 실패한 배포 또는 API 오류로 발생할 수 있습니다. 우리는 원인과 수정이 배포됩니다. 수정이 생길 때마다 업데이트가 제공됩니다.

2. 2026년 7월 24일 오전 3:08 UTCidentified

   이 문제를 해결하기 위해 계속 노력하고 있습니다.

3. 2026년 7월 24일 오전 4:20 UTCmonitoring

   수정이 구현되어 결과를 모니터링하고 있습니다.

4. 2026년 7월 24일 오전 4:24 UTCresolved

   이 사건은 해결되었습니다.

공식 인시던트 업데이트를 자동 번역했습니다.

1. 2026년 7월 8일 오후 7:09 UTCinvestigating

   We are aware of an issue affecting login access to Pulumi Cloud and are investigating.

2. 2026년 7월 8일 오후 7:36 UTCmonitoring

   We have taken mitigation steps and are seeing signs of recovery. We are continuing to monitor the situation.

3. 2026년 7월 8일 오후 7:37 UTCmonitoring

   We are continuing to monitor for any further issues.

4. 2026년 7월 8일 오후 7:45 UTCresolved

   This incident has been resolved.

1. 2026년 6월 22일 오후 6:06 UTCinvestigating

   We are aware of degraded performance on Pulumi APIs and are actively investigating the issue.

2. 2026년 6월 22일 오후 6:30 UTCinvestigating

   We are continuing to investigate this issue.

3. 2026년 6월 22일 오후 7:14 UTCidentified

   We have identified a likely cause and are working on steps to address the issue.

4. 2026년 6월 22일 오후 7:44 UTCmonitoring

   A fix has been rolled out and we are monitoring results

5. 2026년 6월 22일 오후 7:45 UTCmonitoring

   We are continuing to monitor for any further issues.

6. 2026년 6월 22일 오후 7:52 UTCresolved

   This incident has been resolved.

1. 2026년 6월 11일 오후 10:20 UTCinvestigating

   We are investigating an issue affecting customers who sign in to Pulumi Cloud with a Bitbucket identity or use Bitbucket-connected accounts. Beginning at approximately 06:00 UTC on June 11, requests to refresh Bitbucket authentication tokens began failing. Affected users may experience failed sign-ins with Bitbucket, as well as intermittent errors in the Pulumi Console and API after their current Bitbucket session token expires (these tokens expire every two hours). Integrations that rely on Bitbucket credentials may also be affected. Our investigation indicates the failures originate from Bitbucket's OAuth token service rather than Pulumi infrastructure, and we have engaged Atlassian support. Customers using GitHub, GitLab, email, or SSO identities are unaffected. Workaround: Customers whose Pulumi account has an additional sign-in method (GitHub, GitLab, email/password, or SSO) can continue to sign in using that method. We will post updates as the investigation progresses.

2. 2026년 6월 11일 오후 10:56 UTCidentified

   Atlassian has acknowledged an incident affecting Bitbucket OAuth login, which is the cause of the Bitbucket sign-in and authentication failures affecting Pulumi Cloud. Their incident is being tracked at: https://bitbucket.status.atlassian.com/incidents/h51sr145gsmx The issue is external to Pulumi, and no action is required from Pulumi customers. Until Atlassian resolves their incident, customers signing in with a Bitbucket identity may continue to experience failed sign-ins and intermittent Console and API errors after their current session token expires. Customers using GitHub, GitLab, email, or SSO identities remain unaffected. Workaround: Customers whose Pulumi account has an additional sign-in method (GitHub, GitLab, email/password, or SSO) can continue to sign in using that method.

3. 2026년 6월 12일 오전 12:02 UTCmonitoring

   Atlassian has deployed a fix for their incident affecting Bitbucket OAuth login, and we have confirmed that Bitbucket authentication requests from Pulumi Cloud are succeeding again. Error rates have returned to normal levels and sign-in with Bitbucket is working.

4. 2026년 6월 12일 오전 12:17 UTCresolved

   The incident has now been resolved and the service is operating normally for all customers.

1. 2026년 5월 12일 오후 10:50 UTCinvestigating

   Investigating elevated API error rates and response times, particularly with ESC endpoints

2. 2026년 5월 12일 오후 11:03 UTCmonitoring

   A fix has been deployed and error rates and response times have recovered. We are continuing to monitor.

3. 2026년 5월 12일 오후 11:14 UTCresolved

   The incident has been resolved.

1. 2026년 5월 6일 오후 7:05 UTCresolved

   Some Pulumi Cloud organizations experienced elevated errors when decrypting ESC environments. A subset of environment revisions were affected, and those revisions failed to decrypt 100% of the time during the window. Overall, 3.32% of decrypt-attempt requests across Pulumi Cloud failed due to this issue. At the start of the window, we rolled out a change that prevented secret ciphertext copied from an ESC environment from being decrypted in other environments. Customers who had previously copied ciphertext between environments were affected. Users saw a message: “decryption failed. If this secret's ciphertext was copied from a different environment, copy and paste the plaintext instead.” This change was intended as a security improvement, but the customer impact was higher than expected, so the change was rolled back. The ESC team has conducted an internal postmortem to understand the misjudged customer impact, improve our responsiveness to scenarios like this, and created a plan to launch this security improvement without customer disruption.

1. 2026년 4월 15일 오전 9:43 UTCinvestigating

   Users authenticating with Bitbucket are currently unable to sign in to Pulumi Cloud or use the Pulumi Cloud API. This affects all operations that rely on Bitbucket authentication, including CLI usage and API calls. Other authentication methods are not affected. We have identified the cause of the issue and are actively working on a fix. We will provide an update once the fix has been deployed.

2. 2026년 4월 15일 오전 10:26 UTCidentified

   We have started deploying a fix for this issue. We will provide an update once the deployment is complete.

3. 2026년 4월 15일 오후 12:26 UTCmonitoring

   The fix has been deployed and we are seeing recovery. We are continuing to monitor to confirm the issue is fully resolved.

4. 2026년 4월 15일 오후 2:10 UTCresolved

   This incident has been resolved.

1. 2026년 4월 10일 오후 5:58 UTCidentified

   We are aware of reduced response times across Pulumi Cloud APIs. We have identified the cause and are working on a fix

2. 2026년 4월 10일 오후 6:10 UTCresolved

   This incident has been resolved.

1. 2026년 3월 6일 오전 11:29 UTCinvestigating

   We are aware of issues with machine tokens having issues opening environments as part of running previews. We are currently investigating

2. 2026년 3월 6일 오전 11:30 UTCinvestigating

   We are continuing to investigate this issue.

3. 2026년 3월 6일 오전 11:41 UTCidentified

   We are working on a rollback that should resolve this incident. As this rollback is deploying, you may see error rates decrease for a time until the deploy is complete.

4. 2026년 3월 6일 오전 11:45 UTCmonitoring

   Rollback has deployed. We are continuing to monitor

5. 2026년 3월 6일 오전 11:52 UTCresolved

   This incident has been resolved.

1. 2026년 2월 3일 오후 11:26 UTCinvestigating

   We have received reports of errors when creating new Pulumi Neo tasks. We are investigating the issue.

2. 2026년 2월 3일 오후 11:33 UTCresolved

   We've identified and rolled back the change that caused the issue. The incident is now resolved.

1. 2025년 11월 18일 오후 1:44 UTCidentified

   Cloudflare is having an outage which is impacting certain regions ability to resolve artifacts via get.pulumi.com

2. 2025년 11월 18일 오후 3:45 UTCresolved

   Cloudflare has restored the impacted services.

1. 2025년 10월 20일 오후 3:05 UTCidentified

   These are affected by the global AWS outage

2. 2025년 10월 20일 오후 3:43 UTCidentified

   Customers in the eastern part of the United States may experience intermittent errors loading documentation and registry pages

3. 2025년 10월 20일 오후 5:41 UTCidentified

   We are continuing to work on a fix for this issue.

4. 2025년 10월 21일 오전 3:26 UTCresolved

   All impacted areas have recovered

1. 2025년 9월 8일 오후 5:27 UTCinvestigating

   We are currently experiencing issues with displaying some previews in Pulumi Cloud. Our team is actively investigating the problem and working toward a resolution. We will provide updates as soon as more information is available.

2. 2025년 9월 8일 오후 6:00 UTCidentified

   We’ve identified the root cause of the preview display issues in Pulumi Cloud and are actively rolling out a fix. Thanks for your patience — we’ll share another update once the rollout is complete.

3. 2025년 9월 8일 오후 7:21 UTCresolved

   The issue affecting previews in Pulumi Cloud has been resolved. Previews are functioning normally again. Please note that affected previews during the incident window (between 11pm UTC Sept 7, 2025 and 7pm UTC Sept 8, 2025) were not recoverable and will not be available in your history.

1. 2025년 7월 28일 오후 4:42 UTCinvestigating

   We are currently investigating a major issue with a release this morning. We are in the process of rolling this back.

2. 2025년 7월 28일 오후 4:43 UTCmonitoring

   We are noticing a significant drop in error rates and are continuing to monitor.

3. 2025년 7월 28일 오후 4:50 UTCresolved

   This incident has been resolved.

1. 2025년 6월 12일 오후 2:42 UTCidentified

   The issue has been identified and a fix is being implemented.

2. 2025년 6월 12일 오후 2:42 UTCmonitoring

   A fix has been implemented and we are monitoring the results.

3. 2025년 6월 12일 오후 2:44 UTCresolved

   This incident has been resolved.

1. 2025년 6월 11일 오후 1:56 UTCidentified

   As part of a programmed maintenance, old stacks might not load logs or outputs. We are actively working on resolving this issue

2. 2025년 6월 11일 오후 3:30 UTCmonitoring

   A fix has been implemented and we are monitoring the results.

3. 2025년 6월 11일 오후 3:42 UTCresolved

   This incident has been resolved.

1. 2025년 5월 1일 오후 7:03 UTCidentified

   The issue has been identified and a fix is being implemented.

2. 2025년 5월 1일 오후 7:30 UTCidentified

   A fix has been implemented and is being deployed. We are monitoring rollout, ETA of approx 30 minutes.

3. 2025년 5월 1일 오후 8:08 UTCresolved

   This incident has been resolved.

1. 2025년 4월 29일 오후 6:41 UTCidentified

   The issue has been identified and a fix is being implemented.

2. 2025년 4월 29일 오후 7:06 UTCmonitoring

   A fix has been implemented and we are monitoring the results.

3. 2025년 4월 29일 오후 7:09 UTCresolved

   This incident has been resolved.

1. 2025년 3월 28일 오전 9:45 UTCinvestigating

   We are currently tracking an issue where logs for updates aren't being saved and displayed

2. 2025년 3월 28일 오전 10:50 UTCidentified

   A fix is being prepared.

3. 2025년 3월 28일 오전 11:29 UTCresolved

   This incident has been resolved.