improve CI workflow #23

Merged
Bogdanp merged 3 commits from master into master 2 years ago
Bogdanp commented 2 years ago (Migrated from github.com)
Owner

I finally had a bit of time to look into bogdanp/setup-racket#29.

It looks like there were two issues here:

  1. Sometimes xvfb-run doesn't relinquish the display port in time, which leads to xvfb-run: error: Xvfb failed to start errors. This is fixed in 3dc7aa3 by only starting a single framebuffer for all tests.

  2. The tests are SIGABRTed on 6.7BC, 7.{7,8,9}CS. I think this is a genuine issue with those versions of Racket, because, across many runs (1, 2, 3), it's always those versions that fail. The fix here is to allow failures on those versions.

I finally had a bit of time to look into bogdanp/setup-racket#29. It looks like there were two issues here: 1) Sometimes `xvfb-run` doesn't relinquish the display port in time, which leads to `xvfb-run: error: Xvfb failed to start` errors. This is fixed in 3dc7aa3 by only starting a single framebuffer for all tests. 2) The tests are `SIGABRT`ed on 6.7BC, 7.{7,8,9}CS. I think this is a genuine issue with those versions of Racket, because, across many runs ([1](https://github.com/Bogdanp/beautiful-racket/actions/runs/823468578), [2](https://github.com/Bogdanp/beautiful-racket/actions/runs/823483250), [3](https://github.com/Bogdanp/beautiful-racket/actions/runs/823494290)), it's always those versions that fail. The fix here is to allow failures on those versions.
mbutterick commented 2 years ago (Migrated from github.com)
Owner

Thanks for the insight!

Thanks for the insight!
The pull request has been merged as 216536ef29.
Sign in to join this conversation.
No reviewers
No Milestone
No project
No Assignees
1 Participants
Notifications
Due Date

No due date set.

Dependencies

No dependencies set.

Reference: mbutterick/beautiful-racket#23
Loading…
There is no content yet.