Skip to content

Troubleshooting

Find your symptom in the table, jump to the fix. Every runbook page follows the same shape: Symptom → Confirm → Fix → Prevent.

If you see…Go to
GPU installed but inference runs on CPU; log shows CUDA failure 500: named symbol not foundGPU not used — CUDA error 500
Inference container exits/restarts when GPU is enforced and the host GPU is brokenGPU not used — CUDA error 500
Host GPU works, but provider shows a lower tier than expected (e.g. CUDA instead of TensorRT, or CPU)Running on CPU when GPU expected
Log line Failed to load library libonnxruntime_providers_tensorrt.soRunning on CPU when GPU expected
Enabling a datasource bounces back to Disabled with model_not_paired or adapter_connect_failedDatasource down or faulted
A streaming OPC-UA / MQTT / CSV source stops; row shows a fault indicator / red bannerDatasource down or faulted
Datasource is Enabled but no predictions appearDatasource down or faulted
Settings → About highlights the frontend / backend version in amberFrontend / backend version mismatch
Version reads 0.0.0-dev… or ends in -dirtyFrontend / backend version mismatch
Disk filling up; database file far larger than the data it holds; deleting rows doesn’t shrink itDisk fills / database grows unbounded
A container shows unhealthy in docker ps but the app serves fineContainer marked unhealthy but service works