PR for #139: Improving logging to avoid concurrency issues.

[liaochris]

In this issue, we implemented a new logging procedure that replaces subfolder-widesconscript.log files with script-specific files in a log folder that mirrors source and is not version-controlled.

[zhizhongpu]

I think we're just waiting for @arewolinski to take a look.

[zhizhongpu]

@toogiii @arewolinski I hope you can both make sure that scons runs well both clean (after scons --clean) and unclean, before you approve the pull.

This is especially important as my machine sometimes runs into a known issue of matlab hanging

@liaochris do you think we want to execute a run on SLURM too, just in case? If so, we might as well have @arewolinski and @toogiii do it since running the Template is typically the onboarding task for SLURM.

[toogiii]

happy to try the slurm run tomorrow. the scons and scons --clean calls still work for me locally

[liaochris]

@liaochris do you think we want to execute a run on SLURM too, just in case?

@zhizhongpu works when I compile Template on slurm.

FYI @arewolinski @toogiii you'll probably encounter some missing LaTeX .sty file errors when you attempt to compile on the RC. The RC has instructions on how you can install LaTeX packages to overcome those errors, but if you encounter difficulty finding that documentation please let me know.

[liaochris]

@toogiii @arewolinski I think your reviews are the last thing blocking this pull. Here's what I think we need to do in order to merge the pull.

1. Both of you still need to compile the Template on the RC. This should be a similar procedure to compiling the Template on a local machine.

2. @arewolinski I think you still need to leave review comments - please see the pull review guidelines below

Template/docs/workflow.md

Lines 20 to 23 in 8d79103

	* _Reviewers_ **review** the deliverable.
	* Check for violations of our standards for code clarity and data integrity (production deliverables).
	* Check for clear errors or reasons reproducibility will fail (all deliverables.)
	* _Reviewers_ **approve** the pull when all pull threads are resolved.

Compiling the template on the RC should cover point (2) above.

Let me know if you have any questions about this, and when you have done the tasks above, you can approve the pull.

[toogiii]

only one thing @liaochris: should we add .sconsign.dblite to .gitignore? is there a reason we would want to git add this ever?

[toogiii]

resolved other review.

[liaochris]

/run-actions-all --post

[github-actions]

Triggered by @liaochris on this comment

Check Results (run details)

Check	Result	Time
SCons DAG	✅	0.037s
Newlines	✅	0.045s
EPS data	✅	0.024s
Log failures	SKIP

[github-actions]

@liaochris Issue summary

Thanks for closing this pull.

Before leaving the pull, please be sure you have completed all the required steps in the workflow.

This includes filling in the issue summary linked at the top of this comment.

[zhizhongpu]

@liaochris is the SKIP here #167 (comment) expected?