Getting Error on command bash riva_init.sh

AI & Data Science Deep Learning (Training & Inference) Riva
1

Please provide the following information when requesting support.

Hardware - GPU (RTX-8000)
Operating System: Ubuntu 22.04
Riva Version: riva_quickstart_v2.9.0
TLT Version (if relevant): Not sure

How to reproduce the issue ? :

My aim: I want to create an application that does the speech recognition. Speech to text and text to speech.

What am I trying: Get a local Docker using Quick Start scripts to set up a local workstation and deploy the Riva services using Docker.

Following: Quick Start Guide — NVIDIA Riva

Already installed: NVIDIA Driver, CUDA, Docker and downloaded the scripts for 2.9.0

Getting error on: bash riva_init.sh (error shown below)

Logging into NGC docker registry if necessary…

Pulling required docker images if necessary…

Note: This may take some time, depending on the speed of your Internet connection.

Pulling Riva Speech Server images.
Image nvcr.io/nvidia/riva/riva-speech:2.9.0 exists. Skipping.
Image nvcr.io/nvidia/riva/riva-speech:2.9.0-servicemaker exists. Skipping.

Downloading models (RMIRs) from NGC…

Note: this may take some time, depending on the speed of your Internet connection.

To skip this process and use existing RMIRs set the location and corresponding flag in config.sh.

docker: Error response from daemon: could not select device driver “” with capabilities: [[gpu]].

Error in downloading models.

FYI: I am totally new to Riva. I am from windows background and new to Linux OS too. So if you need more info on anything please give the exact script to extract the info from the prompt.

2

Hi @kavithap

Thanks for your interest in Riva

We will help you setup/guide the Riva

Quick check, can you run the following docker command in your machine and share the output (either as screenshot as image and uplaod or as text)

sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

if some error happens in running the above command, Please do the following installation

https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/archive/1.8.1/install-guide.html#setting-up-nvidia-container-toolkit

After installing nvidia-docker2 and restart, i hopefully believe the Riva will start working,

if you face issues in the process, please let me know, will be happy to help

Thanks

3

image
image1366×145 8.9 KB

4

I tried to install nvidia-docker2, but that seems to have some conflict. Please the text copied from the prompt below.

root@riva-vm-01:/home/aptask# distribution=$(. /etc/os-release;echo $ID$VERSION_ID) \

&& curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
&& curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
Warning: apt-key is deprecated. Manage keyring files in trusted.gpg.d instead (see apt-key(8)).
OK
deb https://nvidia.github.io/libnvidia-container/stable/ubuntu18.04/$(ARCH) /
#deb https://nvidia.github.io/libnvidia-container/experimental/ubuntu18.04/$(ARCH) /
deb https://nvidia.github.io/nvidia-container-runtime/stable/ubuntu18.04/$(ARCH) /
#deb https://nvidia.github.io/nvidia-container-runtime/experimental/ubuntu18.04/$(ARCH) /
deb https://nvidia.github.io/nvidia-docker/ubuntu18.04/$(ARCH) /
root@riva-vm-01:/home/aptask# sudo apt-get update
E: Conflicting values set for option Signed-By regarding source https://nvidia.github.io/libnvidia-container/stable/ubuntu18.04/amd64/ /: /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg !=
E: The list of sources could not be read.
root@riva-vm-01:/home/aptask# sudo apt-get install -y nvidia-docker2
E: Conflicting values set for option Signed-By regarding source https://nvidia.github.io/libnvidia-container/stable/ubuntu18.04/amd64/ /: /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg !=
E: The list of sources could not be read.
root@riva-vm-01:/home/aptask#

5

Apologies, Some older tags that used to work have been depreciated,

Please try the following command (as your OS is Ubuntu 22.02, giving this)

sudo docker run --rm --gpus all nvidia/cuda:11.8.0-base-ubuntu22.04 nvidia-smi

and share me the output

Thanks

6

root@riva-vm-01:/home/aptask# sudo docker run --rm --gpus all nvidia/cuda:11.8.0-base-ubuntu22.04 nvidia-smi
Unable to find image ‘nvidia/cuda:11.8.0-base-ubuntu22.04’ locally
11.8.0-base-ubuntu22.04: Pulling from nvidia/cuda
677076032cca: Already exists
bc572704fd22: Pull complete
82ca2dd0fe9d: Pull complete
335006729f70: Pull complete
1b9f8e302abf: Pull complete
Digest: sha256:3a85383782012c87cd78f9b47c33b6105b2b6e17feadf475d5e835234f87a7be
Status: Downloaded newer image for nvidia/cuda:11.8.0-base-ubuntu22.04
docker: Error response from daemon: could not select device driver “” with capabilities: [[gpu]].
root@riva-vm-01:/home/aptask#

7

Thanks, I guess the nvidia-docker2 is missing, Please install via instructions provided in below link

https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/archive/1.8.1/install-guide.html#setting-up-nvidia-container-toolkit

After Completing all steps of installing nvidia-docker2, please try riva setup again and let me know if it works

Thanks

8

Ok I shall do all the steps from starting again.

9

Commandline.txt (10.6 KB)

Please find the txt file that reflects my issues. I am not sure what is it and how to proceed. Is it possible to share screen or remote connect?

10

I am trying to set up the server to run the speech recognition code.
For that I have downloaded NVIDIA Riva Skills 2.9.0.
I have installed the prerequisites mentioned, There were too many issues I corrected it one by one with the help of forum and searching online. I could run the riva local docker also once, But now I am seeing different set of issues and not able to run bash riva_start.sh again. It comes like “Waiting for Riva server to load all models…retrying in 10 seconds
Health ready check failed.
Check Riva logs with: docker logs riva-speech”. Not sure what to do now.
What I need is just a small application exactly what is mentioned in Getting Started with NVIDIA Riva Speech Recognition - YouTube
The code is also readily available. But I am struggling with system settings and related errors and looks like compatibility issue or not sure what.
I am planning to scrap the current VM and take a fresh one.
I need help on setting the NVIDIA Riva Skills 2.9.0. on that.
Hardware - GPU (RTX-8000)
Operating System: Ubuntu 22.04 (if it is not compatible can choose a different one too)
What all do I need to install on that?, to get a simple ASR project run on that?

11

HI @kavithap

Apologies on the delay

You are encountering some problem with the initial setup only and not Riva related,

Nonetheless, The Problem you mentioned in your txt file can be resolved using below link reference
https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/troubleshooting.html#conflicting-values-set-for-option-signed-by-error-when-running-apt-update

Please try this and let me know

Thanks