Skip to content

Update runtime for new version of SageMaker shim #4377

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
jmsmkn merged 4 commits into from
Oct 24, 2025
Merged

Update runtime for new version of SageMaker shim #4377

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
2456537
Remove constant env vars
jmsmkn Oct 23, 2025
61f267c
Always check the signature
jmsmkn Oct 23, 2025
3a667fe
Do not wait as S3 is now strongly consistent
jmsmkn Oct 23, 2025
930b566
Fix test
jmsmkn Oct 23, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 0 additions & 3 deletions app/grandchallenge/components/backends/amazon_sagemaker_base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,5 @@
import logging
import re
import time
from abc import ABC, abstractmethod
from datetime import timedelta
from json import JSONDecodeError
Expand Down Expand Up @@ -719,8 +718,6 @@ def _handle_failed_job(self, *, event):
"Out of Memory. Please use a larger instance",
):
try:
# Allow time for the log file to be written
time.sleep(10)
users_process_exit_code = self._get_task_return_code()
except UncleanExit:
users_process_exit_code = None
Expand Down
9 changes: 2 additions & 7 deletions app/grandchallenge/components/backends/base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -411,8 +411,6 @@ def invocation_environment(self):
"LOG_LEVEL": "INFO",
"PYTHONUNBUFFERED": "1",
"no_proxy": "amazonaws.com",
"GRAND_CHALLENGE_COMPONENT_WRITABLE_DIRECTORIES": "/opt/ml/output/data:/opt/ml/model:/opt/ml/input/data/ground_truth:/opt/ml/checkpoints:/tmp",
"GRAND_CHALLENGE_COMPONENT_POST_CLEAN_DIRECTORIES": "/opt/ml/output/data:/opt/ml/model:/opt/ml/input/data/ground_truth",
"GRAND_CHALLENGE_COMPONENT_MAX_MEMORY_MB": str(
self._max_memory_mb
),
Expand Down Expand Up @@ -820,17 +818,14 @@ def _get_task_return_code(self):

body = response["Body"].read()

signature_hmac_sha256 = response["Metadata"].get(
"signature_hmac_sha256"
)
signature_hmac_sha256 = response["Metadata"]["signature_hmac_sha256"]
body_signature_hmac_sha256 = hmac.new(
key=self._signing_key, msg=body, digestmod=hashlib.sha256
).hexdigest()

if signature_hmac_sha256 and not secrets.compare_digest(
if not secrets.compare_digest(
body_signature_hmac_sha256, signature_hmac_sha256
):
# TODO The signature should always be present when all images use sagemaker shim >= 0.5.0
logger.error(
"The invocation response object has been tampered with"
)
Expand Down
34 changes: 17 additions & 17 deletions app/tests/components_tests/test_amazon_sagemaker_training_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,5 @@
import hashlib
import hmac
import io
import json
from datetime import datetime, timedelta, timezone
Expand Down Expand Up @@ -202,8 +204,6 @@ def test_invocation_json(settings):
"LOG_LEVEL": "INFO",
"PYTHONUNBUFFERED": "1",
"no_proxy": "amazonaws.com",
"GRAND_CHALLENGE_COMPONENT_WRITABLE_DIRECTORIES": "/opt/ml/output/data:/opt/ml/model:/opt/ml/input/data/ground_truth:/opt/ml/checkpoints:/tmp",
"GRAND_CHALLENGE_COMPONENT_POST_CLEAN_DIRECTORIES": "/opt/ml/output/data:/opt/ml/model:/opt/ml/input/data/ground_truth",
"GRAND_CHALLENGE_COMPONENT_MAX_MEMORY_MB": "7168",
"GRAND_CHALLENGE_COMPONENT_SIGNING_KEY_HEX": "746f74616c6c79736563726574",
},
Expand Down Expand Up @@ -556,23 +556,23 @@ def test_handle_completed_job():
time_limit=60,
requires_gpu_type=GPUTypeChoices.NO_GPU,
use_warm_pool=False,
signing_key=b"",
signing_key=b"itsasecret",
)

return_code = 0

with io.BytesIO() as f:
f.write(
json.dumps(
{"return_code": return_code, "pk": f"algorithms-job-{pk}"}
).encode("utf-8")
)
f.seek(0)
executor._s3_client.upload_fileobj(
Fileobj=f,
Bucket=settings.COMPONENTS_OUTPUT_BUCKET_NAME,
Key=executor._result_key,
)
content = json.dumps(
{"return_code": 0, "pk": f"algorithms-job-{pk}"}
).encode("utf-8")
signature = hmac.new(
key=b"itsasecret", msg=content, digestmod=hashlib.sha256
).hexdigest()
executor._s3_client.upload_fileobj(
Fileobj=io.BytesIO(content),
Bucket=settings.COMPONENTS_OUTPUT_BUCKET_NAME,
Key=executor._result_key,
ExtraArgs={
"Metadata": {"signature_hmac_sha256": signature},
},
)

assert executor._handle_completed_job() is None

Expand Down