so https://github.com/roc-lang/roc/actions/runs/5886655302/job/15976575327?pr=5764#step:10:1481 happens pretty often, and it's a tax on everyone's time having to re-run CI on PRs that have this flake...does anyone have any insight into why this happens, and what we can do about it? :sweat_smile:
just double checking if anyone has any context here - it seems like maybe it's a race condition?
e.g. perhaps things are being generated in the same folder with the same name (main.o
for example, or main.wasm
) and we could fix it by using --output
to a tempdir per thread (per thread instead of per test so we can reuse intermediate things potentially)
Last updated: Jul 06 2025 at 12:14 UTC