Heroku 장애 기록

1. 2026년 8월 25일 PM 3:03 UTCinvestigating

   Heroku 엔지니어는 조사입니다.

2. 2026년 8월 25일 PM 3:40 UTCupdate

   EU 지역의 Common Runtime 응용 프로그램에 대한 최근 보안 업데이트는 앱 기본 도메인 TLS 종료 ( .herokuapp.com)에 사용되는 TLS 인증서에 대한 엄격한 정책을 시행 할 수 있습니다. TLS Handhake 실패를 경험하고 앱에 연결할 수 없습니다. 엔지니어는 문제점을 확인하고 우리는 중급에 일하고 있습니다.

3. 2026년 8월 25일 PM 3:55 UTCresolved

   잘못된 해결.

4. 2026년 8월 25일 PM 4:07 UTCupdate

   엔지니어는 문제점을 재화했습니다.

공식 인시던트 업데이트를 자동 번역했습니다.

1. 2026년 8월 18일 PM 9:37 UTCretroactive

   Heroku 엔지니어는 문제를 복고적으로 조사하고 있습니다.

2. 2026년 8월 18일 PM 9:37 UTCresolved

   문제가 해결되었습니다.

공식 인시던트 업데이트를 자동 번역했습니다.

1. 2026년 8월 10일 PM 3:19 UTCresolved

   Incident resolved.

2. 2026년 8월 10일 PM 3:26 UTCinvestigating

   Heroku engineers are investigating.

3. 2026년 8월 10일 PM 3:46 UTCupdate

   We are aware of an issue that affected the ability to provision new Heroku Data add-ons. To remediate, we completed a rollback of a recent change and validated that all services have been restored. We apologize for how you and your business may have been affected by this incident. We will thoroughly investigate the incident, confirming the technical trigger, the underlying cause, and preventive action to avoid a repeat in the future.

1. 2026년 8월 6일 PM 4:17 UTCretroactive

   Heroku engineers are investigating an issue retroactively.

2. 2026년 8월 6일 PM 4:17 UTCresolved

   The issue is now resolved.

3. 2026년 8월 6일 PM 4:19 UTCupdate

   Between 11:01 UTC on August 6, 2026 and 13:10 UTC on August 6, 2026, we identified a performance issue affecting Heroku release phase deploys. During this time, some customers experienced delayed completion of release phase commands during deploys, with some remaining in an `release phase executing` state longer than expected. We have confirmed that the issue is resolved. We apologize for how you and your business may have been affected by this incident. We will thoroughly investigate the incident, confirming the technical trigger, the underlying cause, and preventive action to avoid a repeat in the future.

1. 2026년 7월 6일 PM 1:33 UTCinvestigating

   Heroku engineers are investigating.

2. 2026년 7월 6일 PM 1:54 UTCupdate

   Engineers have identified an issue with an upstream service provider that impacts customers with apps in Cedar Private Spaces. Customers may experience failures with starting and scaling dynos of Cedar Private Spaces apps in Virginia. Engineers are working with the upstream provider to expedite a resolution.

3. 2026년 7월 6일 PM 2:12 UTCupdate

   Engineers have verified some service recovery and are monitoring for full impact resolution.

4. 2026년 7월 6일 PM 2:30 UTCresolved

   Incident resolved.

1. 2026년 6월 19일 PM 5:31 UTCresolved

   Incident resolved.

2. 2026년 6월 19일 PM 6:47 UTCinvestigating

   Heroku engineers are investigating.

3. 2026년 6월 19일 PM 7:09 UTCupdate

   We noted an issue that impacted Heroku Platform API services in all regions. Customers may have experienced broken pipeline API calls to ‘pipelines/:id/latest-builds’, and issues viewing the pipeline dashboard. We confirmed this incident was due to a recently deployed change, and have reverted this change to resolve the issue. We will continue to monitor the environment to ensure stability and prevent a recurrence. We will also thoroughly investigate the incident, confirming the technical trigger, the underlying cause, and any preventive action to avoid a repeat in the future.

1. 2026년 6월 9일 PM 5:01 UTCretroactive

   We noted a feature disruption that impacted the Heroku Platform API that impacted multiple Heroku services, starting at 15:32 UTC. Customers may have observed issues with authentication, application deployments and scaling, Continuous Integration and builds, and accepting Heroku team invitations. We identified the technical trigger as a recent change, and we performed a rollback to resolve the issue as of 16:07 UTC. We will continue to monitor the environment to ensure stability and prevent a recurrence. We will also thoroughly investigate the incident, confirming the technical trigger, the underlying cause, and any preventive action to avoid a repeat in the future.

2. 2026년 6월 9일 PM 5:01 UTCresolved

   The issue is now resolved.

1. 2026년 5월 22일 PM 2:49 UTCretroactive

   Heroku engineers are investigating an issue retroactively.

2. 2026년 5월 22일 PM 2:49 UTCresolved

   The issue is now resolved.

1. 2026년 5월 8일 AM 3:55 UTCinvestigating

   Heroku engineers are investigating.

2. 2026년 5월 8일 AM 4:14 UTCupdate

   Beginning at 12:25 UTC on May 08, 2026, a partial outage in an upstream service provider began impacting a subset of customers in the US region. Affected customers may be experiencing intermittent connectivity, elevated DB latency, or degraded performance with third-party add-on providers. Affected customers may wish trigger database failovers as a mitigation.

3. 2026년 5월 8일 AM 4:33 UTCupdate

   Engineers are continuing to investigate the ongoing impact of this upstream provider outage.

4. 2026년 5월 8일 AM 5:18 UTCupdate

   Engineers are working to restore impacted resources. While the partial outage from the upstream service provider is ongoing, some customers are observing signs of recovery. Upstream resources remain impacted, and our team continues to work toward full resolution.

5. 2026년 5월 8일 AM 5:56 UTCupdate

   Engineers are working to restore impacted resources. While the partial outage from the upstream service provider is ongoing, the majority of customers are observing recovery. Upstream resources remain impacted, and our team continues to work toward full resolution. We will post another update in approximately 2 hours.

6. 2026년 5월 8일 AM 8:18 UTCupdate

   Engineers continue working on restoring impacted resources. While the partial outage from the upstream service provider is ongoing, we are seeing little customer impact. Upstream resources remain impacted, and our team continues to work toward full resolution.

7. 2026년 5월 8일 AM 10:22 UTCupdate

   Engineers continue working on restoring impacted resources. While the partial outage from the upstream service provider is ongoing, we are seeing little customer impact. Upstream resources remain impacted, and our team continues to work toward full resolution. We will post an update in approximately 2 hours, or if anything changes sooner.

8. 2026년 5월 8일 AM 10:49 UTCresolved

   Incident resolved.

9. 2026년 5월 8일 PM 12:15 UTCupdate

   The issue is now resolved.

10. 2026년 5월 8일 PM 3:44 UTCupdate

    We’ve completed the migration to a new availability zone, and the Heroku-hosted web applications and customer databases are now fully operational. We apologize for how this incident affected you and your business. We'll undertake a full investigation of the incident, establishing the technical trigger, the underlying cause, and preventive action to avoid a repeat in the future.

11. 2026년 5월 8일 PM 3:59 UTCupdate

    Engineers continue working on restoring impacted resources. While the partial outage from the upstream service provider is ongoing, we are seeing little customer impact. Upstream resources remain impacted, and our team continues to work toward full resolution. We will post an update in approximately 2 hours, or if anything changes sooner.

12. 2026년 5월 8일 PM 5:24 UTCupdate

    We've confirmed a remediation path, and are applying the fix to the small subset of customers still experiencing a read-only state. We'll provide an update in 60 minutes or sooner if additional information becomes available.

13. 2026년 5월 8일 PM 6:53 UTCupdate

    We have successfully applied the fix to all remaining affected customers. Heroku databases and web applications are now fully operational for all customers. We apologize for how this incident affected you and your business. We will undertake a full investigation of the incident, establishing the technical trigger, the underlying cause, and preventive action to avoid a repeat in the future.

1. 2026년 5월 5일 PM 2:57 UTCinvestigating

   Heroku engineers are investigating.

2. 2026년 5월 5일 PM 3:48 UTCresolved

   Incident resolved.

3. 2026년 5월 5일 PM 4:32 UTCupdate

   Review app creation is now fully operational. Some customers were unable to Create review Apps using the Heroku Pipelines feature, where apps became stuck in a "Creating review app" state with no build logs generated. Running apps weren't affected during this time. We increased worker capacity, restarted the affected backend workers, and confirmed Review app creation was functioning normally. We apologize for how this incident affected you and your business. We will undertake a full investigation of the incident, establishing the technical trigger, the underlying cause, and preventive action to avoid a repeat in the future.

4. 2026년 5월 5일 PM 6:03 UTCissue

   Upon further investigation, we've determined that the start time of impact is different than initially understood. We've revised the start time of the Trust post to more accurately reflect the time customers may have begun to experience impact. We apologize for any confusion caused by this.

1. 2026년 4월 29일 PM 3:48 UTCretroactive

   Heroku engineers identified an issue with the Heroku Dashboard. During this time, users were unable to log in. Functionality has been restored.

2. 2026년 4월 29일 PM 3:48 UTCresolved

   The issue is now resolved.

1. 2026년 3월 12일 PM 8:38 UTCretroactive

   Between 18:45 UTC and 20:00 UTC, on March 12, we noted a feature disruption impacting the Heroku Dashboard. During this time, a subset of customers may have experienced issues accessing the dashboard. Routine operations usually managed via the Dashboard would have had to be managed by alternate means. Application runtimes weren't impacted as part of this incident. We identified a recent code change as the trigger and have since deployed a fix to resolve the issue. We apologize for any impact this incident may have had on your business. We are conducting a thorough investigation to confirm the technical trigger and underlying cause, and we will implement preventive actions to avoid a recurrence.

2. 2026년 3월 12일 PM 8:38 UTCresolved

   The issue is now resolved.

1. 2026년 2월 23일 PM 10:28 UTCretroactive

   Heroku engineers are investigating an issue retroactively.

2. 2026년 2월 23일 PM 10:28 UTCresolved

   The issue is now resolved.

3. 2026년 2월 23일 PM 11:30 UTCfollow-up

   Customers using Heroku in the Common Runtime may have experienced intermittent dyno disruptions earlier today between 17:50 UTC and 20:24 UTC. The issue has been resolved after we disabled a related service and rolled back a recent change. We’re validating the update in lower environments and will continue monitoring for stability.

1. 2026년 2월 18일 AM 1:49 UTCinvestigating

   Heroku engineers are investigating.

2. 2026년 2월 18일 AM 2:09 UTCupdate

   We are continuing to investigate an issue affecting applications on the Heroku platform in the US region. We’ll provide an update in 30 minutes or sooner if additional information becomes available.

3. 2026년 2월 18일 AM 2:43 UTCupdate

   The Technology team is investigating the issue impacting applications on the Heroku platform in the US region. Applications began experiencing database connectivity errors with H99, H34, H27, and H21 error codes. As a result, users can't access certain applications or may experience disruption when attempting to connect to database services. A recent infrastructure change has been rolled back to restore connectivity. We'll provide an update in 30 minutes or sooner if additional information becomes available.

4. 2026년 2월 18일 AM 2:50 UTCresolved

   Incident resolved.

5. 2026년 2월 18일 AM 2:54 UTCupdate

   Applications are beginning to recover after the recent infrastructure change was rolled back. We continue monitoring the services and are seeing positive results. We'll provide an update in 30 minutes or sooner if additional information becomes available.

6. 2026년 2월 18일 AM 3:03 UTCissue

   We’ve confirmed that the solution implemented to mitigate the issue causing the service disruption was successful. Engineers have observed that the service is stable and have resolved the incident. Customers should no longer see any impact. We will fully investigate the incident, confirming the technical trigger, the underlying cause, and preventive action to avoid a repeat in the future. We apologize for how you and your business may have been affected by this incident.