Feature: Add built-in network test #874
Conversation
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the WalkthroughA new internal self-test function is added to verify executor functionality immediately after initialization. The function submits a task, polls for completion with timeout, and raises an error with guidance if the task does not complete within the specified interval. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (2)
src/executorlib/executor/single.py (2)
459-463: Consider a more diagnostic error message.The error message prescriptively suggests
hostname_localhost=Trueas the solution, but timeout could occur for other reasons (network issues, resource contention, worker failures). Consider providing a more diagnostic message that describes what failed and suggestshostname_localhost=Trueas one possible solution rather than the definitive fix.🔎 Suggested improvement
if not f.done(): exe.shutdown(wait=False, cancel_futures=False) raise TimeoutError( - 'Plase try "hostname_localhost=True" in the initialization of the SingleNodeExecutor(hostname_localhost=True).' + 'Executor self-test timed out after 1 second. This may indicate network connectivity issues. ' + 'Try setting hostname_localhost=True in SingleNodeExecutor initialization if running on macOS or in HPC environments.' )
453-464: Consider optional refinements.Two minor refinements to consider:
The function returns
Truebut this value is never used (line 192 doesn't capture it). Consider removing the return statement or documenting why it returns a value.The static analysis tool suggests avoiding long exception messages outside the exception class (TRY003). While this is a style preference, you could consider defining a custom exception class if this pattern is repeated elsewhere.
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
src/executorlib/executor/single.py
🧰 Additional context used
🧬 Code graph analysis (1)
src/executorlib/executor/single.py (2)
src/executorlib/executor/base.py (2)
submit(71-108)shutdown(110-126)src/executorlib/task_scheduler/file/shared.py (1)
done(40-48)
🪛 Ruff (0.14.10)
src/executorlib/executor/single.py
461-463: Avoid specifying long messages outside the exception class
(TRY003)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (20)
- GitHub Check: unittest_openmpi (ubuntu-latest, 3.13)
- GitHub Check: unittest_flux_openmpi
- GitHub Check: unittest_openmpi (macos-latest, 3.13)
- GitHub Check: unittest_openmpi (ubuntu-latest, 3.12)
- GitHub Check: unittest_mpich (ubuntu-latest, 3.13)
- GitHub Check: unittest_openmpi (ubuntu-latest, 3.11)
- GitHub Check: unittest_openmpi (ubuntu-24.04-arm, 3.13)
- GitHub Check: unittest_mpich (macos-latest, 3.13)
- GitHub Check: unittest_openmpi (ubuntu-22.04-arm, 3.13)
- GitHub Check: unittest_mpich (ubuntu-22.04-arm, 3.13)
- GitHub Check: unittest_mpich (ubuntu-24.04-arm, 3.13)
- GitHub Check: unittest_mpich (ubuntu-latest, 3.11)
- GitHub Check: unittest_slurm_mpich
- GitHub Check: unittest_flux_mpich
- GitHub Check: notebooks
- GitHub Check: notebooks_integration
- GitHub Check: benchmark (ubuntu-latest, 3.13, .ci_support/environment-mpich.yml)
- GitHub Check: unittest_win
- GitHub Check: unittest_old
- GitHub Check: benchmark (ubuntu-latest, 3.13, .ci_support/environment-openmpi.yml)
🔇 Additional comments (2)
src/executorlib/executor/single.py (2)
1-1: LGTM!The import of
sleepis necessary for the polling mechanism in the self-test function.
456-458: Verify the timeout duration is sufficient.The self-test uses a 1-second timeout (10 iterations × 0.1s). Ensure this duration is sufficient for various environments, including slower systems, busy clusters, or CI/CD environments where resources may be constrained.
| log_obj_size=log_obj_size, | ||
| ) | ||
| ) | ||
| _executor_self_test(exe=self._task_scheduler) |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
# First, find the file and check its size
find . -name "single.py" -path "*/executor/*" | head -5Repository: pyiron/executorlib
Length of output: 98
🏁 Script executed:
# Examine the file around line 192 and the TestClusterExecutor area
sed -n '185,300p' src/executorlib/executor/single.pyRepository: pyiron/executorlib
Length of output: 8212
🏁 Script executed:
# Continue reading to see TestClusterExecutor.__init__ implementation body
sed -n '289,370p' src/executorlib/executor/single.pyRepository: pyiron/executorlib
Length of output: 4316
🏁 Script executed:
# Search for _executor_self_test definition
rg -n "def _executor_self_test" src/executorlib/Repository: pyiron/executorlib
Length of output: 130
🏁 Script executed:
# View the _executor_self_test function
sed -n '453,500p' src/executorlib/executor/single.pyRepository: pyiron/executorlib
Length of output: 474
🏁 Script executed:
# Check the structure of both __init__ methods to see if they both expose _task_scheduler
sed -n '100,200p' src/executorlib/executor/single.pyRepository: pyiron/executorlib
Length of output: 6781
Add self-test validation to TestClusterExecutor for consistency.
TestClusterExecutor initialization lacks the _executor_self_test() call present in SingleNodeExecutor, despite both having similar executor setup patterns. Consider adding _executor_self_test(exe=self._task_scheduler) after the super().__init__() call to validate executor initialization and catch connection issues early, consistent with SingleNodeExecutor's approach.
🤖 Prompt for AI Agents
In src/executorlib/executor/single.py around line 192, TestClusterExecutor's
initializer is missing the self-test call used in SingleNodeExecutor; add a call
to _executor_self_test(exe=self._task_scheduler) immediately after the
super().__init__(...) return so the TaskScheduler/connection is validated during
initialization, mirroring SingleNodeExecutor and allowing early failure
detection.
| def _executor_self_test(exe): | ||
| f = exe.submit(sum, [1, 1]) | ||
| counter = 0 | ||
| while not f.done() and counter < 10: | ||
| sleep(0.1) | ||
| counter += 1 | ||
| if not f.done(): | ||
| exe.shutdown(wait=False, cancel_futures=False) | ||
| raise TimeoutError( | ||
| 'Plase try "hostname_localhost=True" in the initialization of the SingleNodeExecutor(hostname_localhost=True).' | ||
| ) | ||
| return True |
There was a problem hiding this comment.
Fix typo in error message.
Line 462 contains a typo: "Plase" should be "Please".
🔎 Proposed fix
raise TimeoutError(
- 'Plase try "hostname_localhost=True" in the initialization of the SingleNodeExecutor(hostname_localhost=True).'
+ 'Please try "hostname_localhost=True" in the initialization of the SingleNodeExecutor(hostname_localhost=True).'
)🧰 Tools
🪛 Ruff (0.14.10)
461-463: Avoid specifying long messages outside the exception class
(TRY003)
🤖 Prompt for AI Agents
In src/executorlib/executor/single.py around lines 453 to 464, the error message
raised on timeout contains a typo ("Plase"). Update the string to correct the
spelling to "Please" so the message reads: 'Please try "hostname_localhost=True"
in the initialization of the SingleNodeExecutor(hostname_localhost=True).'
Ensure only the typo is fixed and preserve punctuation and quoting exactly as in
the original message.
|
Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.