hoelzer-lab/rnaflow

hista2 error using singularity profile

chrissikath opened this issue · 7 comments

I try to run the pipeline with the testdata and singularity:

nextflow run main.nf -profile test,singularity

However, i get this error, although i have permissions to write/read to /tmp/.

`ERROR ~ Error executing process > 'preprocess_illumina:hisat2 (MAQCA_rep2)'

Caused by:
  Missing output file(s) `MAQCA_rep2_summary.log` expected by process `preprocess_illumina:hisat2 (MAQCA_rep2)`

Command executed:

  hisat2 -x hsa_small -U MAQCA_rep2.other.fastq.gz -p 1 --new-summary --summary-file MAQCA_rep2_summary.log  | samtools view -bS | samtools sort -o MAQCA_rep2.sorted.bam -T tmp --threads 1

Command exit status:
  0

Command output:
  (empty)

Command error:
  (ERR): mkfifo(/tmp/40.unp) failed.
  Exiting now ...

Work dir:
  /mnt/ribolution/user_worktmp/christina.kuhn/work/2024_04_t_cell_exhaustion/scripts/rnaflow/rnaflow/work/af/f736891078c200c92ca91d4a18427d

Tip: you can try to figure out what's wrong by changing to the process work dir and showing the script file named `.command.sh`

 -- Check '.nextflow.log' file for details`

Hey @chrissikath can you please try to run the latest release of the pipeline like that:

nextflow pull hoelzer-lab/rnaflow
nextflow run hoelzer-lab/rnaflow -r 1.4.7 -profile test,singularity

@hoelzer

$ nextflow pull hoelzer-lab/rnaflow
Checking hoelzer-lab/rnaflow ...

hoelzer-lab/rnaflow contains uncommitted changes -- cannot pull from repository

Aha, strange.

Do you have a folder

ls ~/.nextflow/assets/hoelzer-lab/rnaflow

and if so, can you delete it and try again?

rm -r ~/.nextflow/assets/hoelzer-lab/rnaflow

I still get the same error using:

nextflow run hoelzer-lab/rnaflow -r 1.4.7 -profile test,singularity

Do you have any other idea, what i can try?

Hey,

hm, I am trying to figure out what's going on. Maybe try to download the pipeline code and run from there:

git clone https://github.com/hoelzer-lab/rnaflow.git
cd rnaflow
nextflow run main.nf -profile test,singularity

Alternitavely, could you also try the mamba or conda profile instead of singularity? Or docker?

I already tried it from the downloaded version and also with using conda, mamba (we do not have docker available on our HPC).

However, I switched to singularity, as i had errors with some python packages, possible due to incorrect setup of the conda environments (caused during the expression_reference_based:tpm_filter process).

Command error: Traceback (most recent call last): File ".command.sh", line 3, in <module> import pandas as pd ModuleNotFoundError: No module named 'pandas'

If you have any idea about the conda/mamba, I am open to try it again with conda/mamba.

When you are on a HPC, maybe also try to directly add the job scheduler. What is running on your HPC? SLURM?

-profile test,singularity,slurm