Stub process 'REGIS_0' is not healthy

triton-inference-server / pytriton

PyTriton is a Flask/FastAPI-like interface that simplifies Triton's deployment in Python environments.

https://triton-inference-server.github.io/pytriton/

Apache License 2.0

719 stars 50 forks source link

Stub process 'REGIS_0' is not healthy #37

Closed lfxx closed 11 months ago

lfxx commented 11 months ago

Description

Run this server in docker,and will exit after a while with the error above

To reproduce

It happened by chance and cannot be reproduced 100%.

Environment

Python interpreter distribution and version: [conda 4.7.12 with Python 3.8 environment]
pip version: [23.1.2]
PyTriton version: [0.2.5]
Deployment details: [single gpu A10]

Additional context after first time,i solve this by adding "ipc=host" to dockercompose file.But afterwards,it happened by chance.Anyone come here to help me out!Thanks!

pziecina-nv commented 11 months ago

Hi lfxx, can you share more details on this issue:

how you're starting docker with script hosting pytriton (exact docker command or sample of docker compose yaml file)
obtained logs from failed job
does fail happens on model initialization? during inference? first inference? after some time?

lfxx commented 11 months ago

Hi lfxx, can you share more details on this issue:

how you're starting docker with script hosting pytriton (exact docker command or sample of docker compose yaml file)

obtained logs from failed job

does fail happens on model initialization? during inference? first inference? after some time?

Thanks for your rapid reply.Finally,i realize i add "ipc=host" to docker in a wrong way.I have resloved this issue by adding this para correctly.