Skip to content
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

TST: Add Github CI workflow to run core tests on a crippled FS #4245

Merged
merged 1 commit into from Mar 6, 2020

Conversation

mih
Copy link
Member

@mih mih commented Mar 6, 2020

This aims to disentangle detection of issues with Windows and issues
with adjusted branch operation. There is no expectation that a run
would initially succeed... even on Linux.

Appveyor and Travis runs will be canceled.

TODO

  • VFAT or NTFS?

This seems to be a great success. In 8min we get to see that a large chunk of tests don't work out:

Ran 117 tests in 481.873s
FAILED (SKIP=6, errors=43, failures=5)

The most popular error is

'git annex sync --no-commit --no-push --no-pull' failed with exitcode 1 under /crippledfs/datalad_temp_check_create_path_semantics1p3hq_fs [err: 'fatal: entry '<imagine subdataset path here>' object type (blob) doesn't match mode type (commit)

which I feel has been reported before.

This aims to disentangle detection of issues with Windows and issues
with adjusted branch operation. There is no expectation that a run
would initially succeed... even on Linux.
@kyleam
Copy link
Collaborator

kyleam commented Mar 6, 2020

This aims to disentangle detection of issues with Windows and issues
with adjusted branch operation

Thanks. I often try to figure this out locally (by either using Yarik's handy tools/eval_under_testloopfs script or by setting annex.crippledfilesystem=true temporarily in my global config), so I'd find this additional run helpful.

which I feel has been reported before.

An error with the same message was reported in gh-3824. It seems like that particular case was fixed on git-annex's side.

@kyleam
Copy link
Collaborator

kyleam commented Mar 6, 2020

An error with the same message was reported in gh-3824. It seems like that particular case was fixed on git-annex's side.

Or rather, the git-annex in neurodebian that is being pulled in for this run doesn't yet have that fix. The fix was first released with 7.20191024, while this run is using 7.20190819+git2-g908476a9b.

@mih
Copy link
Member Author

mih commented Mar 6, 2020

If only Github would offer a Debian unstable runner, we could be tracking git-annex more closely. I have started to enjoy the speed of the non-standalone git-annex.

Thx for the analysis @kyleam

@mih mih merged commit 011b46b into datalad:master Mar 6, 2020
1 of 16 checks passed
@mih mih deleted the tst-crippled branch Mar 6, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants