The Command returned non-zero code: 137

[psonpal]

Hi, I am using sm-docker to build my custom environment image in aws sagemaker studio. However while solving environment it gets Killed with error code 137. I tried to do research and found out, that can be because of memory limit and we can specify memory in docker command. I am not sure how to do that with sm-docker command. Is there anything already there like passing an argument for that which I am missing? Thanks.

[rmontroy]

It's not documented, but I found the relevant CLI option in the code and it worked for me:

sagemaker-studio-image-build-cli/sagemaker_studio_image_build/cli.py

Lines 95 to 101 in 87c2505

	build_parser.add_argument(
	"--compute-type",
	help="The CodeBuild compute type (default: BUILD_GENERAL1_SMALL)",
	choices=["BUILD_GENERAL1_SMALL", "BUILD_GENERAL1_MEDIUM",
	"BUILD_GENERAL1_LARGE", "BUILD_GENERAL1_2XLARGE"],
	default="BUILD_GENERAL1_SMALL"
	)