Previous incidents

Issue has been fully resolved and confirmed. Remediation work is underway to ensure this doesn't happen again.

Tesla's Fleet API recovered at 15:40 UTC and has remained stable since. All energy site commands and data are operating normally. Total impact window was approximately 20 minutes; vehicle data, commands, and streaming were unaffected throughout.

Tesla confirmed that this issue was caused by a change they made, which specifically triggered on unsigned configurations with no expiration. The original root cause should now be fixed, however Teslemetry is going to stay on the signed configuration moving forward.

After the previous incident, our Tesla OAuth Refresh service had hung, causing users to gradually lose their Tesla authentication. The service was manually restarted and monitored to ensure service fully restored. We are making changes to ensure this refresh service is more resilient to issues like this in the future. We apologies for the interruption and appreciate everything who reached out to report issues.

The Teslemetry backend suffered an out of memory crash. We are still trying to understand how this happened, but the service has been restored.