-
-
Notifications
You must be signed in to change notification settings - Fork 132
Fix test warnings #1061
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.
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
base: master
Are you sure you want to change the base?
Fix test warnings #1061
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #1061 +/- ##
==========================================
+ Coverage 72.70% 73.02% +0.32%
==========================================
Files 21 21
Lines 5111 5105 -6
==========================================
+ Hits 3716 3728 +12
+ Misses 1395 1377 -18
Flags with carried forward coverage won't be shown. Click here to find out more.
🚀 New features to boost your workflow:
|
- Add timeout (600s) and error handling to TPI.py client.gather() with fallback to serial computation - Increase SS.py client.gather() timeout from 300s to 600s for consistency - Update SS.py to use logging.warning instead of print for better consistency - Adjust LocalCluster timeouts in test_TPI.py fixture: - Increase memory_limit from 2GB to 3GB - Increase communication timeout from 300s to 900s - Increase heartbeat_interval from 10s to 30s - Increase death_timeout from 60s to 120s These changes address issues with long-running computationally intensive tasks where communication can break down due to overly aggressive timeouts. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
The previous commit incorrectly used client.gather(timeout=) which is not a valid parameter. This commit fixes the implementation to use the correct approach: - Use distributed.wait() with timeout to wait for futures to complete - Check if any futures did not complete (not_done) and raise TimeoutError - Only call client.gather() after confirming all futures are done - Maintain the same error handling and fallback to serial computation This properly implements timeout handling for long-running dask computations in both TPI.py and SS.py. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
|
@jdebacker. I have reviewed the code of this PR and approved it. Is it ready to merge? |
|
@rickecon Changing to WIP -- still resolving some local tests (dask timing out in some TPI full runs) and looking to improve performance. |
This PR seeks to address Issue #1047 and reduce the number of pytest warnings.