Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

SageMaker HTTPService automatically starts in r21.07/r21.06

See original GitHub issue

Description

Triton is starting SageMake HTTP Service by default.

I0730 15:43:10.303199 118 grpc_server.cc:4072] Started GRPCInferenceService at 0.0.0.0:8001
I0730 15:43:10.303403 118 http_server.cc:2795] Started HTTPService at 0.0.0.0:8000
I0730 15:43:10.350331 118 sagemaker_server.cc:134] Started Sagemaker HTTPService at 0.0.0.0:8080
I0730 15:43:10.399905 118 http_server.cc:162] Started Metrics Service at 0.0.0.0:8002

even the flag is https://github.com/triton-inference-server/server/blob/6733c3f35317f74f6b49ea2e8c5a1f29c847e94b/build/CMakeLists.txt#L49

Triton Information What version of Triton are you using? nvcr.io/nvidia/tritonserver:21.07-py3 Are you using the Triton container or did you build it yourself? Using Triton container

To Reproduce Run Triton Server with an ONNX model.

Expected behavior SageMaker HTTPService not to start.

Issue Analytics

State:
Created 2 years ago
Comments:9 (5 by maintainers)

Top GitHub Comments

1reaction

GuanLuocommented, Aug 4, 2021

@philschmid The fix is merged into main

0reactions

philschmidcommented, Aug 2, 2021

Can you confirm that the port is used by Triton even if SageMaker is disabled? I ran 21.07 container with SageMaker disabled and I didn’t see port 8080 is used by Triton:

I must have done something wrong. After setting up again and adding --allow-sagemaker maybe it was still running or something else. ✅

Could reference the issue when the fix is applied?