Add monarch.actor.unhandled_fault_hook workaround for proc_mesh.stop bug #73
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This is a better workaround than commenting out proc_mesh.stop() calls. The unhandled_fault_hook prevents proc_mesh.stop from raising exit code 1 which was causing all other tests to fail. Files updated: - test_store.py - test_tensor_slice.py - test_state_dict.py - test_large_tensors.py - test_resharding_basic.py - test_keys.py
meta-cla
bot
added
the
CLA Signed
With the unhandled_fault_hook workaround in place, we can safely uncomment all _proc_mesh.stop() calls that were previously commented out. Files updated: - test_store.py (4 locations) - test_tensor_slice.py (4 locations) - test_state_dict.py (2 locations) - test_large_tensors.py (1 location) - test_resharding_basic.py (2 locations)
Contributor
Author
|
Turns out this workaround still doesn't work. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
This PR adds a better workaround for the monarch bug where
proc_mesh.stop()raises exit code 1, causing all other tests to fail.Changes
monarch.actor.unhandled_fault_hook = lambda failure: Noneto all test files that use monarch actorsFiles Updated
tests/test_store.pytests/test_tensor_slice.pytests/test_state_dict.pytests/test_large_tensors.pytests/test_resharding_basic.pytests/test_keys.pyTesting
This workaround allows
proc_mesh.stop()calls to remain in place while preventing the exit code 1 error.Notes
This is a temporary workaround until the underlying monarch bug is fixed.