Heroku's Common Runtime dropped router and application logs intermittently in US regions for 1h 36m on 24 September 2026. The fault sat in Heroku's own logging pipeline, not in anything you deployed.
Started
14:56 UTC
Duration
Lasted 1h 36m
Source
IsDown
Next time
Heroku is degraded. Is your product working?
An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.
50 monitors for your own websites. No card.
Affected components: Apps
What happened?
Heroku's status page lists one incident on this day, "Heroku Feature Degradation", starting at 14:56 UTC and running 1h 36m before resolution. The first substantive update says the actual onset was earlier, at 13:31 UTC, when some Common Runtime customers began seeing intermittent loss of application logs in the US region. By 15:51 UTC Heroku narrowed the scope, saying the known impact was intermittent router logs being dropped in a subset of US regions, and that app logs were unaffected at that point, a slight contradiction with the earlier update that is worth noting rather than smoothing over. At 16:10 UTC Heroku said engineers had fixed the issue and were monitoring, with all router logs being successfully delivered again. That is the last update on record, so the incident is treated as closed at that timestamp. Only one component, Apps, is listed as affected on Heroku's status page. Four updates were published in total, all shown here, and no further filing followed the fix. No user reports were logged in this window, so there is no independent account of what the dropped logs looked like from outside Heroku. Over the past 90 days Heroku has logged four incidents with a median duration of 1 hour 9 minutes, making this one somewhat longer than typical. Across all time Heroku has filed 377 incidents, averaging 4.9 a month, so a logging degradation like this one sits within a fairly high baseline rate of filings.
Learning
This incident was about logging infrastructure, not about whether your app served requests, which is exactly the kind of fault an uptime check against your own endpoint will not catch. Your probes test whether a request reaches your app and gets a response, and Heroku's own updates never say that application routing or request handling failed, so your dashboards likely stayed green the whole time. Meanwhile router and application logs were being dropped intermittently, which matters if you rely on those logs for debugging, auditing or alerting, because a gap in logs during an incident window can hide the very thing you need to investigate later. This is a layer below what your product-level monitoring is built to see: the platform's internal pipeline, not your deployed code. Watching your own app's availability tells you your half is fine, but it tells you nothing about whether the provider's logging, routing or build layers are degraded underneath you. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.
