Historique des pannes de Heroku

1. 25 août 2026 à 15:03 UTCinvestigating

   Les ingénieurs Heroku enquêtent.

2. 25 août 2026 à 15:40 UTCupdate

   Une récente mise à jour de sécurité pour les applications communes Runtime dans la région de l'UE peut avoir involontairement appliqué une politique plus stricte pour les certificats TLS utilisés pour la terminaison du domaine TLS par défaut de l'application ( .herokuapp.com). Cela a amené certains clients à éprouver des défaillances de poignée de main TLS et à ne pas pouvoir se connecter aux applications. Les ingénieurs ont identifié le problème et nous travaillons à l'assainissement.

3. 25 août 2026 à 15:55 UTCresolved

   L'incident est réglé.

4. 25 août 2026 à 16:07 UTCupdate

   Les ingénieurs ont corrigé le problème.

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 18 août 2026 à 21:37 UTCretroactive

   Les ingénieurs Heroku enquêtent rétroactivement sur un problème.

2. 18 août 2026 à 21:37 UTCresolved

   La question est maintenant réglée.

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 10 août 2026 à 15:19 UTCresolved

   Incident resolved.

2. 10 août 2026 à 15:26 UTCinvestigating

   Heroku engineers are investigating.

3. 10 août 2026 à 15:46 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. 6 août 2026 à 16:17 UTCretroactive

   Les ingénieurs Heroku enquêtent rétroactivement sur un problème.

2. 6 août 2026 à 16:17 UTCresolved

   La question est maintenant réglée.

3. 6 août 2026 à 16:19 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.

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 6 juillet 2026 à 13:33 UTCinvestigating

   Heroku engineers are investigating.

2. 6 juillet 2026 à 13:54 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. 6 juillet 2026 à 14:12 UTCupdate

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

4. 6 juillet 2026 à 14:30 UTCresolved

   Incident resolved.

1. 19 juin 2026 à 17:31 UTCresolved

   Incident resolved.

2. 19 juin 2026 à 18:47 UTCinvestigating

   Heroku engineers are investigating.

3. 19 juin 2026 à 19:09 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. 9 juin 2026 à 17:01 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. 9 juin 2026 à 17:01 UTCresolved

   The issue is now resolved.

1. 22 mai 2026 à 14:49 UTCretroactive

   Heroku engineers are investigating an issue retroactively.

2. 22 mai 2026 à 14:49 UTCresolved

   The issue is now resolved.

1. 8 mai 2026 à 03:55 UTCinvestigating

   Heroku engineers are investigating.

2. 8 mai 2026 à 04:14 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. 8 mai 2026 à 04:33 UTCupdate

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

4. 8 mai 2026 à 05:18 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. 8 mai 2026 à 05:56 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. 8 mai 2026 à 08:18 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. 8 mai 2026 à 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. 8 mai 2026 à 10:49 UTCresolved

   Incident resolved.

9. 8 mai 2026 à 12:15 UTCupdate

   The issue is now resolved.

10. 8 mai 2026 à 15:44 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. 8 mai 2026 à 15:59 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. 8 mai 2026 à 17:24 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. 8 mai 2026 à 18:53 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. 5 mai 2026 à 14:57 UTCinvestigating

   Heroku engineers are investigating.

2. 5 mai 2026 à 15:48 UTCresolved

   Incident resolved.

3. 5 mai 2026 à 16:32 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. 5 mai 2026 à 18:03 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. 29 avril 2026 à 15:48 UTCretroactive

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

2. 29 avril 2026 à 15:48 UTCresolved

   The issue is now resolved.

1. 12 mars 2026 à 20:38 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. 12 mars 2026 à 20:38 UTCresolved

   The issue is now resolved.

1. 23 février 2026 à 22:28 UTCretroactive

   Heroku engineers are investigating an issue retroactively.

2. 23 février 2026 à 22:28 UTCresolved

   The issue is now resolved.

3. 23 février 2026 à 23:30 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. 18 février 2026 à 01:49 UTCinvestigating

   Heroku engineers are investigating.

2. 18 février 2026 à 02:09 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. 18 février 2026 à 02:43 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. 18 février 2026 à 02:50 UTCresolved

   Incident resolved.

5. 18 février 2026 à 02:54 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. 18 février 2026 à 03:03 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.