search

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
687 stars 45 forks source link

fix boot when allow_http=False #49

Closed catwell closed 6 months ago

catwell commented 6 months ago

Currently the server cannot boot with allow_http=False, because the log level checker fails like this with grpc:

Screenshot from 2023-12-21 14-40-18

This PR uses the correct field in the protobuf for the log_verbose_level parameter (uint32 instead of string).

piotrm-nvidia commented 6 months ago

Thank you for your pull request. It seems that you have fixed a bug in the log_verbose_level parameter handling. I appreciate your contribution to PyTriton.

However, before I can merge your pull request, I need you to do a one thing.

Please sign the Contributor License Agreement (CLA) and send it to triton-cla@nvidia.com. The CLA is mandatory for all contributors to Triton Inference Server and its related projects. You can download the CLA here: https://github.com/triton-inference-server/server/blob/main/Triton-CCLA-v1.pdf

I apologize for the inconvenience, but these are the standard procedures for accepting pull requests. I hope you understand and comply with them.

I also have to inform you that my team is on holiday until 27th December due to Christmas. Therefore, we will not be able to review your pull request until then. I hope you don't mind waiting a bit longer.

Thank you for your patience and cooperation. I wish you a merry Christmas and a happy new year.

catwell commented 6 months ago

CLA sent. No problem with waiting, happy holidays!

nv-kmcgill53 commented 6 months ago

CLA sent. No problem with waiting, happy holidays!

The CLA has been accepted. We may merge at our discretion.