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

Improve work directory cleanup #17

Closed
jmmv opened this issue Feb 22, 2014 · 3 comments
Labels
bug

Comments

@jmmv
Copy link
Owner

@jmmv jmmv commented Feb 22, 2014

From jmmv@google.com on March 30, 2011 11:55:59

In revision 648ed6360b2b7cda81a6079b00dc436d09c745b8 of ATF, there have been several bug fixes regarding the clean up of work directories. These bug fixes are related to dealing with race conditions that may arise, and Kyua needs to deal with them as well.

Basically, during cleanup, any errors raised should not stop the cleanup. Some of these errors can be ignored (like failing to scan the contents of a directory iff the directory itself can be removed) and some other errors can be retried a few times after a grace period. No matter what, unrecoverable errors during cleanup (i.e. errors that prevent us from cleaning the work directory altogether) should report the test case as broken, not crash the runtime engine. (Kyua most likely does not crash in these conditions, but an "expected" exception propagates too high up the call chain and that's not desired.)

Lastly, we also need to implement unmounting of directories within the work directory.

Original issue: http://code.google.com/p/kyua/issues/detail?id=17

@jmmv

This comment has been minimized.

Copy link
Owner Author

@jmmv jmmv commented Feb 22, 2014

From jmmv@google.com on June 02, 2011 14:22:29

This issue was updated by revision r134 .

This functionality will soon be used in the work directory cleanup code
to ensure stale mount points left by the test case are properly cleaned.

@jmmv

This comment has been minimized.

Copy link
Owner Author

@jmmv jmmv commented Feb 22, 2014

From jmmv@google.com on June 04, 2011 08:58:11

This issue was updated by revision r135 .

Testing for this functionality is quite tricky and unportable. While I
think the new code should do the right thing and cover the problems we
discovered before in atf-run, I'm not sure it does. I still have to try
this on the NetBSD tests to see how things go.

@jmmv

This comment has been minimized.

Copy link
Owner Author

@jmmv jmmv commented Feb 22, 2014

From jmmv@julipedia.org on June 05, 2011 11:46:09

The psshfs test cases in NetBSD used to trigger the aforementioned behavior. After some intensive testing, I can't reproduce the problems any more, so let's consider this fixed for now :-)

Status: Fixed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
1 participant
You can’t perform that action at this time.