Skip to content

Conversation

@JamesonRGrieve
Copy link
Owner

No description provided.

- Linting workflows (test-py-lint, test-ts-lint): singlethreaded
- Testing workflows (test-*): multithreaded for parallel test execution
- Other workflows: singlethreaded
- Updated all workflows to use fromJSON(inputs.runs_on) for array support
Change parallel_workers default from 'auto' to '4'. The 'auto' setting
detects host/hypervisor CPUs (e.g., 20) rather than container-allocated
resources, causing OOM kills when running on singlethreaded runners
with 1 CPU and limited RAM.
- Multithreaded runner: 6 workers (matches 8 CPU / 8GB RAM)
- Singlethreaded runner: 1 worker (matches 1 CPU / 1GB RAM)

Default is now empty string which triggers auto-detection from the
runs_on input. Explicit values can still be passed to override.
@JamesonRGrieve JamesonRGrieve merged commit 7896bf6 into main Dec 22, 2025
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.

3 participants