Describe the bug
Getting "OSError: [Errno 30] Read-only file system: '/opt/ml/models/code'" when using multi-model mode with docker 2.1 cpu
To reproduce
1) Set the below in dockerfile.
ENV SAGEMAKER_MULTI_MODEL=True
ENV SAGEMAKER_MULTI_MODEL_UNIVERSAL_BUCKET=s3 bucket name
ENV SAGEMAKER_MULTI_MODEL_UNIVERSAL_PREFIX=s3 prefix, which holds the inference code.
2) Build and Publish the docker to ECR.
3) Create model, endpoint config and endpoint via sagemaker runtime of boto3. Above error's get thrown while, creating the endpoint.
4) Invoke Endpoint.
Expected behavior
Should create the directory opt/ml/models/code and download the code from S3 to the opt/ml/models/code, so that custom inference can work.
Screenshots or logs
System information
A description of your system. Please provide:
Toolkit version: 2.1
Framework version: 2.1
Python version: 3.6
CPU or GPU: CPU
Custom Docker image (Y/N): Y (Just the changes to dockerfile about setting those above listed ENV variables)
Additional context
NA
Describe the bug Getting "OSError: [Errno 30] Read-only file system: '/opt/ml/models/code'" when using multi-model mode with docker 2.1 cpu
To reproduce 1) Set the below in dockerfile. ENV SAGEMAKER_MULTI_MODEL=True ENV SAGEMAKER_MULTI_MODEL_UNIVERSAL_BUCKET=s3 bucket name ENV SAGEMAKER_MULTI_MODEL_UNIVERSAL_PREFIX=s3 prefix, which holds the inference code. 2) Build and Publish the docker to ECR. 3) Create model, endpoint config and endpoint via sagemaker runtime of boto3. Above error's get thrown while, creating the endpoint. 4) Invoke Endpoint.
Expected behavior Should create the directory opt/ml/models/code and download the code from S3 to the opt/ml/models/code, so that custom inference can work.
Screenshots or logs
System information A description of your system. Please provide: