Repository navigation
TypeError: StoredFunction.__init__() got an unexpected keyword argument 'hmac_key' #5495
Description
PySDK Version
- PySDK V2 (2.x)
- PySDK V3 (3.x)
Describe the bug
Upon using the step decorator, I receive errors supposedly at function bootstrap time. The logs erroneously indicate the step was run successfully.
To reproduce
from sagemaker.mlops.local.local_pipeline_session import LocalPipelineSession
from sagemaker.mlops.workflow.function_step import step
from sagemaker.mlops.workflow.pipeline import Pipeline
@step(
image_uri="public.ecr.aws/sagemaker/sagemaker-distribution:3.7.0-cpu",
instance_type="ml.m5.large",
)
def preprocess() -> None:
print("Preprocessing step")
if __name__ == "__main__":
name = "test"
res_preprocess = preprocess()
session = LocalPipelineSession()
pipeline = Pipeline(
name=name,
steps=[preprocess()],
sagemaker_session=session,
)
session.create_pipeline(pipeline=pipeline, pipeline_description="foo")
session.start_pipeline_execution(PipelineName=name)
Expected behavior
A clear and concise description of what you expected to happen.
Screenshots or logs
Remote:
local:
pzu6ht9kua-algo-1-frkcn | /app/.venv/lib/python3.11/site-packages/sagemaker/train/remote_function/invoke_function.py:104: DeprecationWarning: sagemaker.train.remote_function.core.stored_function has been moved to sagemaker.core.remote_function.core.stored_function. Please update your imports. This shim will be removed in a future version.
pzu6ht9kua-algo-1-frkcn | from sagemaker.train.remote_function.core.stored_function import StoredFunction
pzu6ht9kua-algo-1-frkcn | 2026-01-19 13:55:14,079 sagemaker.remote_function ERROR Error encountered while invoking the remote function.
pzu6ht9kua-algo-1-frkcn | Traceback (most recent call last):
pzu6ht9kua-algo-1-frkcn | File "/app/.venv/lib/python3.11/site-packages/sagemaker/train/remote_function/invoke_function.py", line 144, in main
pzu6ht9kua-algo-1-frkcn | _execute_remote_function(
pzu6ht9kua-algo-1-frkcn | File "/app/.venv/lib/python3.11/site-packages/sagemaker/train/remote_function/invoke_function.py", line 106, in _execute_remote_function
pzu6ht9kua-algo-1-frkcn | stored_function = StoredFunction(
pzu6ht9kua-algo-1-frkcn | ^^^^^^^^^^^^^^^
pzu6ht9kua-algo-1-frkcn | TypeError: StoredFunction.__init__() got an unexpected keyword argument 'hmac_key'
pzu6ht9kua-algo-1-frkcn exited with code 0
Aborting on container exit...
Container pzu6ht9kua-algo-1-frkcn Stopping
Container pzu6ht9kua-algo-1-frkcn Stopped
[01/19/26 14:55:15] INFO ===== Job Complete ===== image.py:323
[01/19/26 14:55:16] INFO Pipeline step 'preprocess-1e842e1e-0a59-4828-8c6c-7bb549f0600d' SUCCEEDED. pipeline_entities.py:180
INFO Pipeline execution eb1dabc4-48b3-4e79-b6e5-229d92df098e SUCCEEDED
System information
A description of your system. Please provide:
- SageMaker Python SDK version: 3.3.1
- Framework name (eg. PyTorch) or algorithm (eg. KMeans): n/a
- Framework version: n/a
- Python version: 3.11
- CPU or GPU: cpu
- Custom Docker image (Y/N): N
Additional context
It appears l.110 in invoke_function.py is the source of the bug: https://github.com/aws/sagemaker-python-sdk/blob/master/sagemaker-train/src/sagemaker/train/remote_function/invoke_function.py#L110
Activity
nkmglobalp commentedon Jan 20, 2026
+1 I also faced same issue
fix: prescreen finds merged fixes that never cite the issue, and fixe…
AkshayShah03 commentedon Oct 10, 2026
This looks fixed on master. The traceback comes from
sagemaker/train/remote_function/invoke_function.py, which passedhmac_key=toStoredFunction, while the coreStoredFunctiontakessigning_key. #6294 renamed that argument tosigning_keyin the train copy ofinvoke_function(the core copy already used it), and the change first shipped in v3.23.0.Upgrading to sagemaker 3.23.0 or later (
pip install --upgrade sagemaker) should resolve it, and the issue can probably be closed. If the error still shows up on 3.23+, a traceback from that version would help.