Unable to start riva: Triton server died before reaching ready state. Terminating Riva startup

Please provide the following information when requesting support.

Hardware - NVIDIA GeForce RTX 3070 Notebook GPU (8 GB)
Hardware - 12th Gen Intel(R) Core™ i7-12700H
Operating System - Ubuntu 20.04.5 LTS
Riva Version - 2.7.0
Cuda Version - 11.6

After following the quick start guide and run
$ bash riva_init.sh

I can’t start Riva

Starting Riva Speech Services. This may take several minutes depending on the number of models deployed.
74acd89a9b363eba1eabd3adde96301487caaab856b70474ad5602767dab9df6
Waiting for Riva server to load all models…retrying in 10 seconds
Waiting for Riva server to load all models…retrying in 10 seconds
Waiting for Riva server to load all models…retrying in 10 seconds
.
.
.
Waiting for Riva server to load all models…retrying in 10 seconds
Waiting for Riva server to load all models…retrying in 10 seconds
Health ready check failed.

I’ve seen that it can’t start because it runs out of memory, so I modified the config.sh file to deploy only the ASR models, I attached the file.
config.sh (9.7 KB)

I attach the log bash riva_init.sh | tee riva_init_log.txt
riva_init_log.txt (253.3 KB)

And this is docker logs riva-speech
riva-speech.log (109.6 KB)

Hi @nharo

Thanks for your interest in Riva

Riva is supported on any NVIDIA Volta or later GPU (NVIDIA Turing and NVIDIA Ampere GPU architecture) for development purposes. Care must be taken to not exceed the memory available when selecting models to deploy. 16+ GB VRAM is recommended.

https://docs.nvidia.com/deeplearning/riva/user-guide/docs/support-matrix.html#server-hardware

Thanks