Resolved
This issue is now resolved.
Monitoring
The engineers observe that metrics are within nominal ranges and keep monitoring them.
Update
Our third-party vendor is continuing to work on a number of different mitigation and resolution actions. They've identified the root cause of this issue to be an impairment of several network devices. While there are some early signs of recovery, there remains no ETA for full recovery yet.
Update
Multiple engineering teams from our third-party vendor are still actively working on parallel mitigation paths for this issue. While we are seeing incremental recovery, it is not consistent and there's still no ETA for full recovery.
Update
Our third-party vendor has partially mitigated the incident. Heroku continues to work closely with them for a full path to mitigation. There's no ETA for recovery at this time.
Update
Engineers have detected that our upstream provider is experiencing elevated error rates in the US region. This may impact dyno creation/formation, connect dashboard, database backups, maintenances for a subset of apps hosted in both the US and EU regions.
Issue
Engineers are looking into possible availability issues in the US and EU regions.
Customers will see failed resource creation, backups and maintenances on a subset of databases.