Fix flaky FileIOTest.testMatchWatchForNewFiles test under CI filesystems#38047
Fix flaky FileIOTest.testMatchWatchForNewFiles test under CI filesystems#38047bvolpato wants to merge 2 commits intoapache:masterfrom
Conversation
Summary of ChangesHello, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request addresses a flaky test in FileIOTest by ensuring that file attributes are correctly maintained when files are replaced. By preserving original timestamps and increasing the wait time for file operations, the test becomes more resilient to the timing and filesystem constraints often encountered in CI environments. Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here. Footnotes
|
bvolpato
force-pushed
the
bvolpato/fix-fileio-flaky-test
branch
2 times, most recently
from
5b40803 to
4350fb8
Compare
addresses apache#19480 The `FileIOTest.testMatchWatchForNewFiles` test occasionally flakes in the CI environment because the `updOptions` configuration in `CopyFilesFn` does not preserve file attributes when overwriting existing files. In some CI filesystems, this causes the copied file to register a `lastModifiedMillis` timestamp of `0`. Thus, when `ExtractFilenameAndLastUpdateFn` parses this file, it throws a `RuntimeException` at `FileIO.java:800`, failing the pipeline run. This PR adds `StandardCopyOption.COPY_ATTRIBUTES` to preserve the file's original timestamps, avoiding the exception.
bvolpato
force-pushed
the
bvolpato/fix-fileio-flaky-test
branch
from
4350fb8 to
5e86d7e
Compare
|
[review-prs-verification] Verified after pushing
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request modifies FileIOTest.java to include COPY_ATTRIBUTES in file update options and introduces manual modification time updates using future timestamps to ensure distinct mtimes during test execution. Feedback suggests adopting a more deterministic method for timestamp generation to avoid fragility and recommends capturing the system time once when updating multiple files to ensure consistency.
| Files.setLastModifiedTime( | ||
| sourcePath.resolve("first"), FileTime.fromMillis(System.currentTimeMillis() + 2000)); |
There was a problem hiding this comment.
Using a future timestamp (System.currentTimeMillis() + 2000) is a workaround for filesystem resolution issues but can be fragile in environments that validate file timestamps. Consider if a more deterministic way to ensure increasing timestamps (e.g., using a base time and incrementing) would be more reliable.
| Files.setLastModifiedTime( | ||
| sourcePath.resolve("first"), FileTime.fromMillis(System.currentTimeMillis() + 4000)); | ||
| Files.setLastModifiedTime( | ||
| sourcePath.resolve("second"), FileTime.fromMillis(System.currentTimeMillis() + 4000)); |
There was a problem hiding this comment.
Calling System.currentTimeMillis() multiple times for related updates can lead to inconsistent timestamps if the clock ticks between calls. It's better to capture the timestamp once. Also, the use of future offsets (e.g., + 4000) remains a potential source of flakiness on some platforms.
| Files.setLastModifiedTime( | |
| sourcePath.resolve("first"), FileTime.fromMillis(System.currentTimeMillis() + 4000)); | |
| Files.setLastModifiedTime( | |
| sourcePath.resolve("second"), FileTime.fromMillis(System.currentTimeMillis() + 4000)); | |
| long now = System.currentTimeMillis(); | |
| Files.setLastModifiedTime( | |
| sourcePath.resolve("first"), FileTime.fromMillis(now + 4000)); | |
| Files.setLastModifiedTime( | |
| sourcePath.resolve("second"), FileTime.fromMillis(now + 4000)); |
|
Assigning reviewers: R: @kennknowles for label java. Note: If you would like to opt out of this review, comment Available commands:
The PR bot will only process comments in the main thread (not review comments). |
1 participant
Please add a meaningful description for your change here
addresses #19480
The
FileIOTest.testMatchWatchForNewFilestest occasionally flakes in the CI environment because theupdOptionsconfiguration inCopyFilesFndoes not preserve file attributes when overwriting existing files. In some CI filesystems, this causes the copied file to register alastModifiedMillistimestamp of0. Thus, whenExtractFilenameAndLastUpdateFnparses this file, it throws aRuntimeExceptionatFileIO.java:800, failing the pipeline run.This PR adds
StandardCopyOption.COPY_ATTRIBUTESto preserve the file's original timestamps, avoiding the exception.Thank you for your contribution! Follow this checklist to help us incorporate your contribution quickly and easily:
addresses #123), if applicable. This will automatically add a link to the pull request in the issue. If you would like the issue to automatically close on merging the pull request, commentfixes #<ISSUE NUMBER>instead.CHANGES.mdwith noteworthy changes.See the Contributor Guide for more tips on how to make review process smoother.