Langfuse: Degraded LLMaJ execution in EU
Affected components: LLM-as-a-judge
Timeline of updates
- investigating
Eval execution in EU region is taking longer than usual
- identified
evaluation throughput was overwhelmed by a slow tenant/model workload, with lock-expiry retries multiplying work; the concurrent worker rollout prolonged the incident.
- monitoring
our autoscaling responded slower than expected, but responded nonetheless. the backlog is draining
- resolved
Steady throughput for the last 30 minutes. Delay < 5min.
Archive proof
Rewindex commits each version of an incident it captures to a daily checkpoint, chains it to the day before, and has the checkpoint timestamped by FreeTSA and DigiCert.
How the archive is checked, and what it cannot prove.
This incident predates live tracking (October 17, 2026). It came from the vendor’s own published history, which is not covered by checkpoints.
Impact, status and times are the vendor’s own, as last archived. View on status.langfuse.com (opens in new tab).