This repository has been archived by the owner on Sep 18, 2024. It is now read-only.
Fix parameter file bug #667
Merged
Conversation
This file contains 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
Pull latest code
pull latest code
pull latest code
pull latest code
pull latest code
Pull latest code
pull latest code
pull code
pull code
pull code
pull code
pull code
Fix nniManager unit test (#515)
pull code
pull code
pull code
pull code
pull code
* Support distributed job for frameworkcontroller (#612) support distributed job for frameworkcontroller * Multiphase doc (#519) * multiPhase doc * updates * updates * Add time parser for 'nnictl update duration' (#632) Current nnictl update duration only support seconds unit, add a parser for this command to support {s, m, h, d} * fix experiment state bug (#629) * update top README.md (#622) * Update README.md * update (#634) * Integration tests refactoring (#625) * Integration test refactoring (#21) (#616) * Integration test refactoring (#21) * Refactoring integration tests * test metrics * update azure pipeline * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * updates * update trigger * Integration test refactoring (#618) * updates * updates * update pipeline (#619) * update pipeline * updates * updates * updates * updates * updates * test pipeline (#623) * test pipeline * updates * updates * updates * Update integration test (#624) * Update integration test * updates * updates * updates * updates * updates * updates
This reverts commit 62fc165.
Revert "Pull code"
pull code
Detect tuner failing (#635)
pull code
pull code
Refactoring nnimanager log (#652)
goooxu
approved these changes
Jan 28, 2019
xuehui1991
approved these changes
Jan 28, 2019
yds05
reviewed
Jan 28, 2019
| @@ -69,7 +69,7 @@ def get_next_parameter(): | |||
| params_filepath = os.path.join(_sysdir, params_file_name) | |||
| if not os.path.isfile(params_filepath): | |||
| request_next_parameter() | |||
| while not os.path.isfile(params_filepath): | |||
| while not (os.path.isfile(params_filepath) and os.path.getsize(params_filepath) > 0): | |||
There was a problem hiding this comment.
looks it still exist race condition issue here, because even if the size of params_file is larger than 0, it doesn't mean the param config file is intact.
suggest to use a solution totally resolving this issue, like create a .tmp file for each new config file, and rename this .tmp file after completing writing.
There was a problem hiding this comment.
Thanks, I will firstly merge this to unblock integration test, then adopt your solution to solve it completely.
yds05
approved these changes
Jan 28, 2019
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
To fix this multiphase SDK bug: #639
When training service is creating parameter.cfg file, sometimes the file is created but content not flushed to disk yet, during this short time period, SDK detects that the file is created, and try to load the content before the content is flushed by training service.
Solution:
SDK checks parameter.cfg file size > 0