-
Notifications
You must be signed in to change notification settings - Fork 4.5k
Restricting hosts where Python PreCommit runs due to environmental issues. #5149
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
Conversation
|
Host choice is due to having seen passing tests earlier today: |
6f59664 to
6430976
Compare
|
Other latest Python PreCommits are running in |
|
I'm also adding a check for existence of a directory before removal in gen_protos. |
6430976 to
4146b7d
Compare
aaltay
left a comment
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.
Approving this to enable tests. I do not think I understand why nodes become bad or recover later. Is it possible that our tests are leaving things in an invalid state?
sdks/python/gen_protos.py
Outdated
| finally: | ||
| sys.stderr.flush() | ||
| shutil.rmtree(build_path) | ||
| if os.path.exists(build_path): |
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.
I thought we decided to use
shutil.rmtree(build_path, ignore_errors=True instead of a conditional check (https://docs.python.org/2/library/shutil.html#shutil.rmtree)
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.
Ah I'm sorry. Good memory. We had not merged that fix, but I'll make the change now.
4146b7d to
acd6399
Compare
|
I've checked with Jason. I asked why Python tests pass/fail on some machines:
It doesn't seem that old Maven-based precommit had any sort of restrictions on hosts, so I'm not sure what's going on in this case: https://github.com/apache/beam/blob/a25157bc6a9203362f6dd23acc8b7e1f30daad8e/.test-infra/jenkins/job_beam_PreCommit_Python_MavenInstall.groovy Somehow it must be that the Gradle build and the Maven build had different assumptions? |
|
Run Seed Job |
|
Retest this please |
|
The maven-based precommit didn't use virtualenv (it modified and used the users install of Python). This should be more hermetic, assuming a recent virtualenv and pip. |
No description provided.