Closed slackr31337 closed 1 year ago
CC @NateMeyer
After re-reading the beta-2 release notes I see that the Nvidia driver has to be version 530 or >. I'm currently running 515.65.01. I will upgrade the Nvidia driver and see if that resolves this issue.
Hope that driver helps, I'm not sure what else would be the issue. Not sure if that tensorrt log in the tmp folder is accessible if you still have a problem, that might give another clue.
Hello,
I was able to upgrade TrueNAS to TrueNAS-SCALE-23.10-RC.1 which includes NVIDIA driver 535.54.03. The docker container now generates the trt model successfully.
Also, the tensorrt log in the tmp folder was empty.
Describe the problem you are having
Starting docker image 0.13.0-beta2-tensorrt for the first time fails to generate model.
Cuda failure in file 'yolo_layer.cu' line 51: 35 /etc/s6-overlay/s6-rc.d/trt-model-prepare/run: line 82: 204 Aborted
Version
0.13.0-0858859
Frigate config file
docker-compose file or Docker CLI command
Relevant log output
Operating system
Debian
Install method
Docker CLI
Coral version
Other
Any other information that may be helpful
HPE DL380 Gen10 TrueNAS-SCALE-22.12.3.2 Intel(R) Xeon(R) Silver 4110 CPU @ 2.10GHz 16 cores (32 threads) 125.6GiB total available (ECC) Tesla P4 GPU
root@frigate-ix-chart-7bf9b5b4f5-b6z4x:/tmp# nvidia-smi Tue Sep 26 20:04:44 2023
+-----------------------------------------------------------------------------+ | NVIDIA-SMI 515.65.01 Driver Version: 515.65.01 CUDA Version: 11.7 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 Tesla P4 Off | 00000000:37:00.0 Off | Off | | N/A 52C P0 23W / 75W | 397MiB / 8192MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| +-----------------------------------------------------------------------------+