Heroku Outage History

Heroku reported intermittent routing availability affecting some applications in the EU region, with users potentially experiencing routing failures or degraded connectivity. The incident began at 6:21 PM UTC and was resolved by 7:21 PM UTC. Heroku posted two status updates during the incident, noting that engineers were investigating the cause and impact. The Apps component transitioned from up to warn and back to up as services were restored, with EU-region routing returning to full availability after about an hour.

Heroku experienced an outage affecting runtimes in the US Virginia region, impacting Common, Private Spaces, and Fir runtimes. Users in that region may have seen errors or delays. The incident began at Aug 21, 2025 8:11 PM UTC and was resolved by Aug 21, 2025 9:06 PM UTC, lasting about 55 minutes. The provider posted two status updates during the incident while connectivity issues in an upstream provider were confirmed.

The impact was regional to the US Virginia area, with affected components listed as the Common, Private Spaces, and Fir runtimes; there is no indication of a broader, global impact.

Heroku conducted a scheduled maintenance window affecting the Apps component. The maintenance began at 9:16 AM UTC when Apps transitioned to maintenance and ended at 10:36 AM UTC, lasting about 1 hour and 20 minutes.

During the incident, the status page communicated the maintenance and its completion, with two status updates issued. Apps returned to normal operation after the window was completed.

Heroku conducted a scheduled maintenance window for the Apps platform on 2025-08-13. The maintenance began at 12:01 AM UTC and lasted about 1 hour, ending at 1:01 AM UTC. During the window the Apps component was listed as in maintenance, which may have caused interruptions or degraded operations for users deploying, scaling, or accessing their applications. The incident was documented in two status updates: one announcing the scheduled maintenance and another confirming that the maintenance was underway, followed by restoration of the Apps service as of 1:01 AM UTC.

Heroku experienced a prolonged outage that affected customer apps due to an issue with uneven routing distribution to dynos, causing increased latency and timeouts for many apps. The incident began on 10:00 PM UTC and persisted until 2:31 PM UTC when services were restored. During the incident, Heroku issued two status updates: the first at Aug 1, 2025 2:16 PM UTC describing the uneven routing issue and that engineers were actively working to resolve it, and the second at Aug 1, 2025 2:31 PM UTC confirming that Apps had returned to normal.

Beginning on {{date:2025-07-17}}, some Heroku customers reported uneven routing distribution to dynos, leading to increased latency and timeouts for affected apps. The issue impacted the Apps component and disrupted normal app operation for users relying on Heroku dynos.

The incident progressed to a full outage for the Apps component, with the status changing from up to down at 6:41 PM UTC and being resolved by 1:36 AM UTC, for a total duration of about 7 hours and 55 minutes. Heroku posted two official updates during the incident, describing ongoing engineering work to restore proper routing. The outage reflected a widespread impact on apps hosted on Heroku during the event.

Heroku Automated Certificate Management (ACM) experienced delays due to an outage with a third-party provider, impacting certificate issuance and renewal. The issue began at Jul 21, 2025 8:46 PM UTC and escalated to a full outage for the ACM Tools component by Jul 21, 2025 10:41 PM UTC, which could cause delays for new certificates.

Engineers monitored the progression as the incident unfolded. At Jul 22, 2025 10:51 AM UTC the ACM status moved from down to warn as stabilization progressed and the delayed certificates were issued. The Tools component returned to up at Jul 22, 2025 4:31 PM UTC. The incident lasted from Jul 21, 2025 8:46 PM UTC to Jul 22, 2025 4:31 PM UTC (~20 hours), across four official status updates.

Heroku reported a degraded state for data.heroku.com, including Dataclips and the Postgres Data Dashboard. Some customers experienced inconsistent dashboard loading, while databases were not affected. The incident began at Jul 16, 2025 6:16 PM UTC and was resolved by Jul 16, 2025 9:51 PM UTC, lasting about 3 hours and 35 minutes. User feedback described issues with Dataclips not loading.

An incident in Heroku's Common Runtime in the US region allowed some application processes to continue running after their dynos were scaled down or stopped. These processes were not visible via the usual Heroku CLI but could generate continued logs and related side effects. The incident began at 6:16 PM UTC and was resolved by 8:06 PM UTC, lasting approximately 1 hour and 50 minutes. The issue primarily affected the Apps component and could impact logging and resource behavior for affected apps.

Heroku's status page shows two updates during the incident, with the Apps component moving from a warning to an operational state as engineers investigated and resolved the root cause. The event was localized to the US region, and services returned to normal after the incident window.

Heroku experienced a partial outage centered on dyno formations and autoscaling. Users could see inconsistencies in the number of dynos running and occasional autoscale failures, impacting apps and related tooling. The outage lasted approximately 8 hours and 30 minutes.

The outage began at Jun 18, 2025 4:31 PM UTC and was resolved by Jun 19, 2025 1:01 AM UTC. The status page issued two updates during the incident: the first notified customers of an ongoing investigation into dyno formations and autoscaling, and the second confirmed restoration of normal operation for the Tools and Apps components.