Hello everyone…
I’m trying to run Issac Omniverse for Remote Workstation Deployment
docker run --runtime=nvidia --env NVIDIA_VISIBLE_DEVICES=0 -e “ACCEPT_EULA=Y” --rm -p 47995-48012:47995-48012/udp -p 47995-48012:47995-48012/tcp -p 49000-49007:49000-49007/tcp -p 49000-49007:49000-49007/udp nvcr.io/nvidia/isaac-sim:2020.1_ea
but I’m getting the next error related to ray tracing:
2020-06-30 16:54:17 [1,952ms] [Info] [carb.renderer.plugin] Skipping scene renderer ‘iray’ because it was not in /carb/renderer/enabled list = rtx
2020-06-30 16:54:17 [1,953ms] [Warning] [carb.renderer.plugin] Failed to create context for scene renderer ‘rtx’.
2020-06-30 16:54:17 [2,329ms] [Error] [omni.kit.plugin] Startup Error
2020-06-30 16:54:17 [2,329ms] [Error] [omni.kit.plugin] Unable to create renderer. Make sure you are running with an NVIDIA GPU supporting RTX Ray Tracing with a minimum of 6GB GPU memory, and a recent driver.
2020-06-30 16:54:17 [2,329ms] [Error] [omni.kit.plugin] carb::renderer::createContext failed.
/isaac-sim/_build/linux-x86_64/release/isaac-sim-headless.sh: line 4: 10 Segmentation fault (core dumped) “$SCRIPT_DIR/…/…/target-deps/kit_sdk_release/_build/linux-x86_64/release/omniverse-kit” --config-path “$SCRIPT_DIR//experiences/isaac-sim-headless.json” --no-window $@
According to the requirements, it is necessary NVIDIA GPU (RTX 2070 or higher) and
NVIDIA GPU Driver minimum version 440.59
I have access to a server with three GeForce RTX 2080 Ti and NVIDIA-SMI 440.100 (Driver Version: 440.100, CUDA Version: 10.2)
Please, I need help to find a solution…
Best regards
Hi. Welcome to the Forums.
It looks like you are using the old nvidia docker. Please take a look at the new NVIDIA Container Toolkit. You can do a fresh install or upgrade from nvidia-docker2. Please test if the install works by running:
docker run --gpus all nvidia/cuda:10.0-base nvidia-smi
You should then run with:
docker run --gpus all -e "ACCEPT_EULA=Y" --rm -p 47995-48012:47995-48012/udp -p 47995-48012:47995-48012/tcp -p 49000-49007:49000-49007/tcp -p 49000-49007:49000-49007/udp nvcr.io/nvidia/isaac-sim:2020.1_ea
Note:
–gpus all: to use all available gpus
–gpus '“device=0”: to use specific gpu device (device 0)
Hi, I have the same issue and I am using a K80:
NVIDIA-SMI 450.51.05 Driver Version: 450.51.05 CUDA Version: 11.0 |
|-------------------------------±---------------------±---------------------+
| 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 K80 Off | 00000001:00:00.0 Off | 0 |
| N/A 36C P0 70W / 149W | 0MiB / 11441MiB | 1% Default |
| | | N/A |
And NVIDIA-docker is already v2 and the latest docker version.
Is the card to small to run ISAAC Sim 2020?
Cheers,
Axel
Hi, I’m afraid Tesla K80 is not supported. The Tesla T4 is recommended or any other RTX cards like 2060.
Please use driver version 440 and CUDA 10 for Isaac Sim 2020.1.
Here’s our system requirements:
https://docs.omniverse.nvidia.com/robotics/text/Robotics_QuickStart.html#local-requirements
1 Like
Thanks!!! Somehow I missed it. It helps.