Hi, I am trying to run the nf-core/scrnaseq pipeline on Dardel. Below is the command I used: ./nextflow run nf-core/scrnaseq --input ./samplesheet.csv --outdir ./results --fasta /cfs/klemming/projects/supr/naiss2025-23-256/indranil/Genome_V2/genome.fa --gtf /cfs/klemming/projects/supr/naiss2025-23-256/indranil/Genome_V2/genes.gtf --aligner cellranger -profile pdc_kth --project=naiss2025-22-666 Before running this, I loaded the following modules: module load PDC java singularity I also downloaded Nextflow following the instructions on this page: https://nf-co.re/configs/pdc_kth/ Initially, the pipeline started correctly. When I ran ‘squeue --me’, I noticed several processes were queued and two were actively running. Nextflow also created several files in results and work directory. However, after 1–2 hours, the pipeline appeared to stop without any error messages. When I checked ‘squeue --me’ again, it still listed several queued processes, but none were running. Now its almost 6-7 hours but the queue status is still the same. I would appreciate your help in understanding what might have gone wrong or how I can debug this. Thank you!

Nothing is necessarily wrong, there's nothing unexpected about waiting a few hours (or days).

On thing though. Use

module load PDC apptainer

rather than

module load PDC singularity

. Check your

projinfo

to see if you're over quota on core hours. On Dardel, generally core hours are used up quicker than other swedish clusters because lower memory per cpu allocated. It's not uncommon to wait a while though sometimes.

Hi, I would like to follow up regarding the issue I reported earlier with running the nf-core/scrnaseq pipeline on Dardel. As suggested, I waited for 7 days, but the jobs are still stuck in the same state. The

squeue

output shows the following:

JOBID     PARTITION  NAME     USER      STATE    TIME  TIME_LIMI  NODES NODELIST(REASON)  
11702102  shared     nf-NFCOR indrasin PENDING  0:00  7-00:00:00   1   (AssocGrpCPURunMinutesLimit)  
11702103  shared     nf-NFCOR indrasin PENDING  0:00  7-00:00:00   1   (AssocGrpCPURunMinutesLimit)  
11702104  shared     nf-NFCOR indrasin PENDING  0:00  7-00:00:00   1   (AssocGrpCPURunMinutesLimit)

Initially, I tested with 6 scRNA-seq samples (~250 GB total), but later reduced it to 3 samples. However, the pipeline still gets stuck after the initial steps and never progresses. Could this be a specific issue related to the nf-core/scrnaseq pipeline on Dardel, or does it look more like a queueing/resource allocation issue? Any advice on how to resolve this would be greatly appreciated. Thanks a lot for your help.

Ok, that reason implies that your allocation is too small for those jobs. I have no idea how long they should need, but you can try shortening them and see how much shorter they need to be before they change reason. E.g.

scontrol update job=11702102 timelimit=6-0:0:0

, to set them to six days rather than seven, then trying with even less if needed until they change (you should still expect to wait somewhat, it's the reason you need to check).