Errors in document translations for customers using DeepL API in the US datacenter
none- Started
- May 2, 9:41 AM EDT
- Resolved
- May 2, 9:00 AM EDT
- Duration
- —
DeepL's updates (2)
- resolved · May 2, 9:00 AM EDT
Between 14:00 UTC and 14:20 UTC customers using document translations via the API hosted in the US Datacenter would have received translation failures and/or errors.
- postmortem · May 5, 4:26 AM EDT
We experienced an issue with our document processing services due to the temporary unavailability of certain system resources. This was caused by the eviction of some nodes, a common occurrence in our setup. Although the recovery took slightly longer than expected, it was resolved quickly without intervention. To prevent similar issues in the future, we implemented measures to ensure the availability of critical components and moved our infrastructure to more stable nodes. These changes will significantly reduce interruptions.
Updates are DeepL's own words from its status page (status: postmortem); we never add an inferred cause. The vendor's incident page ↗