The Heroku API and dyno infrastructure are now fully online and stable.
We're continuing to recover the remaining affected databases and will be tracking the recovery processes via a new status incident.
The API is now fully functional. Our engineers are continuing to work to recover the remaining affected databases.
We are continuing to restore stability to the underlying affected infrastructure components. The API is in a read-only mode. Some databases are also affected and our engineers are restoring them now.
Git-push deploys have been re-enabled. The API remains in read-only mode.
The Heroku API has been transitioned from maintenance to read-only mode. App unidling has been restored. H99 error rates are falling rapidly, but some apps are still experiencing errors.
Our engineers are continuing efforts to move Heroku components onto working infrastructure. The API and deploys remain disabled.
Heroku API remains in maintenance mode while we attempt to recover from infrastructure issues.
Some running apps are experiencing H99 errors, and apps are not able to unidle currently. Deploys are also temporarily disabled.
We're continuing to replace routing infrastructure to improve performance for apps using A records.
We have detected a failure of an HTTP router servicing one of our apex domain IP addresses. Apps with apex domains using A records mapped to this IP may experience partial HTTP errors. Engineers are failing over to a backup router and expect to resolve the issue shortly.
We're experiencing an issue with our infrastructure provider. We've placed our API into maintenance mode while we fix the underlying issues.
Some applications may experience issues launching processes or unidling dynos.
The Heroku API is having availability issues. We're investigating.
Running apps are not affected.