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

R_CMD_CHECK runs very slow in MacOS latest #63

Open
cnluzon opened this issue Aug 10, 2020 · 3 comments
Open

R_CMD_CHECK runs very slow in MacOS latest #63

cnluzon opened this issue Aug 10, 2020 · 3 comments

Comments

@cnluzon
Copy link
Collaborator

cnluzon commented Aug 10, 2020

Latest tests added made Check step in MacOS very slow (took 24 minutes full run whereas Ubuntu runs in 11).

My guess is that it has something to do with bw_bins that was untested before for efficiency.

I can either skip the bw_bins test for certain cases if they slow down this, and approach the testing of these functions through mocking and testing the underlying functions.

Note that this also affects regular installation time in MacOS, most likely.

@cnluzon
Copy link
Collaborator Author

cnluzon commented Aug 10, 2020

Further investigation into this shows that it's not the tests, but some accumulation on things that run slower on MacOS, apparently, which are BioConductor installation (4min vs 8min) and the whole R_CMD_CHECK step list being slightly slower in each step, accumulating to ~11minutes vs 4 minutes in Ubuntu.

I am not sure how easy it is to fix this. It slows down the merging of pull requests, though.

@cnluzon
Copy link
Collaborator Author

cnluzon commented Aug 11, 2020

This seems to depend on other external factors. Now, including quite a bit more testing, times are again around 10 minutes and it is evenly distributed across Windows, Ubuntu and MacOS, so I am going to close this for now.

@cnluzon cnluzon closed this as completed Aug 11, 2020
@cnluzon
Copy link
Collaborator Author

cnluzon commented Oct 5, 2020

Reopening this because it takes way too long.

Sometimes there is some kind of timeout crash when there are no nodes available for running. This seems to happen more often with MacOS nodes.

One option is to run Ubuntu only at this time since it caches package installing and it will still allow me to catch R-CMD-Check fails (automated tests, installation to some degree and documentation warnings). If ran into other type of problems (installation and so on) I can reinstate the testing for several platforms.

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

No branches or pull requests

1 participant