[Bug]: Fix AI operation execution boundaries
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Newbie friendliness
- 48/100
- Issue type
- Bug
- Clarity
- Clearly specified
- Activity status
- Active
- Tech stack
- aws, python
- Domain
- backend-api-design, testing
Research direction
Start with the listed AI implementation files, especially _shared.py, retry.py, transports.py, llm.py, and decider.py, and read their corresponding AI unit tests. Work through the eight required changes and add the specified branch-level tests, then add an end-to-end replay test under tests/e2e. Done means all acceptance criteria pass, including tests, type checks, formatting, and lint.
Written by the indexing model from the issue text.
Description
Expected Behavior
The experimental AI operations should enforce their public contract at each execution boundary:
- The extras package requires a core SDK version that supports custom step subtypes.
- An
AiTimeoutErrorretries in a later invocation, not during the ten-second reserve left for checkpointing. - An interrupted HTTP response is a retryable transport failure.
- The request thread preserves the step's context variables.
- An unavailable invocation deadline counts as no deadline.
llm().askcheckpoints only a response known to be complete.- A decider input is serialized once before the durable step starts.
- The documentation states the minimum usable function timeout.
Actual Behavior
Python SDK #928: add experimental decider and LLM operations merged these operations in the AI operations merge commit 8fbb2459. The following cases remain:
- The extras package allows core SDK 2.x, but every AI operation constructs
StepConfig(sub_type=...). Core 2.x has nosub_typeparameter, so a call fails withTypeErrorbefore it creates an operation. default_ai_retry_strategyschedules the first timeout retry after two seconds. A sibling operation can keep the invocation alive, allowing retries to start during the ten-second reserve and fail without calling the model.- The HTTP transport catches
URLErrorandOSError, but nothttp.client.HTTPException.IncompleteReadfrom either response-body read therefore becomes a non-retryableStepError. - The deadline path starts a bare daemon thread. Context variables set by the step, including an active OpenTelemetry context, are absent inside boto3, urllib, and custom transport calls.
- Deadline property access and invocation are not protected. A throwing getter fails the step, and non-finite numeric values reach the wait calculation.
llm().askaccepts a response with nostopReasonand checkpoints any text in it, although the response is not known to be complete.- Each decider call serializes its input for preflight validation, then serializes it again inside the step. A supported container subclass with observable iteration runs twice and can produce a different request on the second serialization.
- The retry documentation promises a new invocation, but it does not state that a function timeout of ten seconds or less leaves no request budget.
Steps to Reproduce
These results reproduce on main at 8fbb2459 with Python 3.14:
- Constructing
StepConfig(sub_type="LlmAsk")with core SDK 2.0.0 raisesTypeError: StepConfig.__init__() got an unexpected keyword argument 'sub_type'while the extras dependency accepts that version. - In a local-runner
parallel, give the AI call about 10.05 seconds remaining and keep a sibling active for seven seconds. Attempts 2 and 3 restart in the first invocation, during the reserve, without another model call. Attempt 4 runs in the next invocation. - Make an HTTP response's
read()raisehttp.client.IncompleteRead. The exception escapes_http_transport, anddefault_ai_retry_strategyreturns no retry. - Set a
ContextVarbeforecall_before_deadline. It has the set value without a deadline and its default value inside the deadline thread. - Return Bedrock output text without
stopReason._conversereturns the text instead of raisingLlmStoppedError. - Give
decider().checkalistsubclass whose__iter__records calls. It is iterated twice.
All 145 current AI tests pass, so they do not cover these outcomes.
SDK Version
main at 8fbb2459; introduced by Python SDK #928: add experimental decider and LLM operations. No release tag contains this commit as of 2026-11-13.
Python Version
3.14. The defects are not specific to 3.14.
Is this a regression?
No. These behaviors were introduced with the experimental AI operations.
Last Working Version
N/A
Required Changes and Tests
- Require
aws-durable-execution-sdk-python>=3.0.0directly. Do not add a core 2.x fallback. - Give
AiTimeoutErrora minimum retry delay greater thanINVOCATION_RESERVE_SECONDS, and test it with a decreasing deadline and a running parallel sibling. - Convert
http.client.HTTPExceptionfailures from both response-body reads intoDeciderError; preserve an HTTP error status when headers were received. Add truncated 200 and truncated error-response tests. - Run the request thread under
contextvars.copy_context().run, with a test that reads a context variable inside a transport. - Treat an exception, non-number, or non-finite deadline reading as no deadline, with a test for each branch.
- Raise
LlmStoppedErrorwhenstopReasonis missing, with and without response content. - Capture the validated serialized decider input and pass that string into the step so serialization occurs once.
- State that the function timeout must leave more than the ten-second reserve.
- Add an end-to-end test under
tests/e2ethat exercises the public decider and LLM operations through replay with deterministic fake boundaries.
Files
packages/aws-durable-execution-sdk-python-extras/pyproject.tomlpackages/aws-durable-execution-sdk-python-extras/src/aws_durable_execution_sdk_python_extras/ai/_shared.pypackages/aws-durable-execution-sdk-python-extras/src/aws_durable_execution_sdk_python_extras/ai/retry.pypackages/aws-durable-execution-sdk-python-extras/src/aws_durable_execution_sdk_python_extras/ai/transports.pypackages/aws-durable-execution-sdk-python-extras/src/aws_durable_execution_sdk_python_extras/ai/llm.pypackages/aws-durable-execution-sdk-python-extras/src/aws_durable_execution_sdk_python_extras/ai/decider.pypackages/aws-durable-execution-sdk-python-extras/docs/ai.md- AI unit and end-to-end tests
Out of Scope
- Changes to the core SDK.
- New AI operations, model hosts, retry options, or fallback paths for core SDK 2.x.
- Changes to checkpoint or error wire formats.
Acceptance Criteria
- The extras package requires core SDK 3.0.0 or later.
- A timeout retry cannot start during the invocation reserve.
- Interrupted HTTP bodies produce the documented retry decision.
- Request work sees the step's context variables.
- Unavailable or unusable deadline readings do not fail an otherwise valid call.
- A Bedrock response without
stopReasonis never checkpointed as a successful answer. - Each decider input is serialized exactly once.
- The minimum function timeout is documented.
- Unit and end-to-end tests fail when any corrected branch is removed or inverted.
- Tests, type checks, formatting, and lint pass.
- Dominant language
- Python
- Stars
- 59
- Forks
- 32
- Avg merge
- 1d 5h
- Merged PRs (30d)
- 76
Getting set up
- No Dockerfile or Docker Compose file
- Has a pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from aws/aws-durable-execution-sdk-python
-
needs-triage
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
aws/aws-durable-execution-sdk-python#977 ·
Maintainers usually reply within 1 day
-
[AgentCore session] Remove the unused session timeoutPossibly taken @srivilliamsai claimed this today. Openneeds-triage
Difficulty 2/5 1-3 hours Newbie friendliness 77/100
aws/aws-durable-execution-sdk-python#970 · 1 comment ·
Maintainers usually reply within 1 day
-
enhancement needs-triage
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
aws/aws-durable-execution-sdk-python#952 ·
Maintainers usually reply within 1 day
-
[AgentCore session] W4: Document sessions in the worker READMEPossibly taken @ParidelPooya claimed this 3 days ago. Openenhancement needs-triage
Difficulty 1/5 1-3 hours Newbie friendliness 76/100
aws/aws-durable-execution-sdk-python#950 ·
Maintainers usually reply within 1 day
-
[AgentCore session] W3: Add the session ID and job count to worker logsPossibly taken @ParidelPooya claimed this 3 days ago. Openenhancement needs-triage
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
aws/aws-durable-execution-sdk-python#949 ·
Maintainers usually reply within 1 day
All issues in aws/aws-durable-execution-sdk-python
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 73/100
Maintainers usually reply within 1 day
-
Unused loot bag group names left in dreamcraft en_US.langPossibly taken A pull request linked to this issue is open or already merged. OpenStatus: Triage Suggestion
Difficulty 1/5 Under an hour Newbie friendliness 90/100
GTNewHorizons/GT-New-Horizons-Modpack#27789 ·
Maintainers usually reply within 1 day
-
bug frontend sev-5
Difficulty 2/5 1-3 hours Newbie friendliness 83/100
-
core good first issue
Difficulty 2/5 1-3 hours Newbie friendliness 65/100
-
b4b code health next
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
ESCOMP/CTSM#4330 · 1 comment ·
Maintainers usually reply within 1 day