-
Notifications
You must be signed in to change notification settings - Fork 1.9k
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
Added working version of checking whether file is available for access #4938
Conversation
With this build in the draft testing PR #4893, I have tested my changes with AutoFitRegressionTest() and AutoFitBinaryTest() running 100 iterations. These tests that used to hang are not not hanging anymore. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
Recently AutoFitRegressionTest() and AutoFitBinaryTest() have been failing occasionally on our MachineLearning-Full CI builds with the error:
(Specific errors for AutoFitRegressionTest and AutoFitBinaryTest from CI builds)
This indicates that the .dataset file that is trying to be reached still has a lock on it, and is not yet ready for accessing. The current method of checking whether this file is ready to be accessed (checking
new FileInfo(dataFile).Length > 0
is not sufficient.The fix below remedies this by obtaining these dataset files from the local
test/data
path, and thereby eliminating the possibility of download locks. This fix also does the same for other datasets that are available locally that were being downloaded from a commit before, so that similar file lock issues for those datasets can be avoided as well.