A community effort to collect a curated set of analysis pipelines built using Nextflow.

nf-core

hello, if i’m running out of space on the volume that i’m using for my workdir and out dir, is there any way i can split my samplesheet and free up space without rerunning the whole pipeline? i’d really like to avoid rerunning the whole pipeline if possible

Probably not sorry. But this might be useful: <https://nextflow.io/blog/2024/experimental-cleanup-with-nf-boost.html|https://nextflow.io/blog/2024/experimental-cleanup-with-nf-boost.html>

You can probably rerun with resume using that plugin and it'll clean up the work directories for you. 

It'll delete those files though so it won't be resumable after that, so it's a bit all-or-nothing

Thank you. Sounds like if I have some free space left, I can split my sample sheet into groups A/B/C, resume A to completion, cleanup A, resume B to completion, cleanup B, then resume C?

Using `nextflow clean -f -before $( nextflow log -q | tail -n 1 )` might already go a long way to cleaning up space. This cleans up work directories that are not associated with the last runs, hopefully freeing up space.

&gt; Sounds like if I have some free space left, I can split my sample sheet into groups A/B/C, resume A to completion, cleanup A, resume B to completion, cleanup B, then resume C?
Possibly, though it wasn't what I was suggesting. It might be that changing the samplesheet invalidates the entire run cache, but _maybe_ Nextflow is clever enough to find the cached tasks for the individual steps :thinking_face:  Not sure

If the samplesheet goes through a process, it will likely restart the whole pipeline. But if samples are read through a `fromPath(...).splitCsv(...)` chain or similar, then it won't restart the dropped samples ( I think )