I have a nextflow user who cannot manage to get any pipeline to run, always hitting the same bug when initiating the nextflow environment and more specifically when downloading the required dependencies. It precisely always happen when trying to download the nextflow-23.04.1.jar file, which puzzles me as it does not need it, the command being executed form that same very file to start with.
Why would nextflow try to download itself?
Annoyingly, it seems to be a user-specific issue as I can't reproduce this error with my own user, but I can't find anything specific in their environment that would cause such an issue. Disk space for downloading the artifacts does not seem to be an issue, even though I might be underestimating the size of the nextflow-23.04.1.jar file.
Can you please help me diagnose what is the issue here?
Expected behavior and actual behavior
that the pipeline runs, or at least that they could get the Nextflow usage message when doing nextflow -h
from an environment that works fine (under my own user), we should get this:
[fl4@farm5-head2 test]$ nano-rave --help
N E X T F L O W ~ version 23.04.1
Launching `/software/pathogen/custom_installs/nano-rave/v1.0.1/nano-rave/nano-rave` [agitated_bose] DSL2 - revision: c2f00dc80c
Usage:
nextflow run nano-rave
Options:
--sequencing_manifest Manifest containing paths to sequencing directories and sequencing summary files (mandatory)
--reference_manifest Manifest containing reference identifiers and paths to fastq reference files (mandatory)
--results_dir Specify results directory [default: ./nextflow_results] (optional)
--variant_caller Specify a variant caller to use [medaka (default), medaka_haploid, freebayes] (optional)
--min_barcode_dir_size Specify the expected minimum size of the barcode directories, in MB. Must be > 0. [default: 10] (optional)
--help Print this help message (optional)
[fl4@farm5-head2 test]$ nano-rave -h
Execute a pipeline project
Usage: run [options] Project name or repository url
Options:
-E
Exports all current system environment
Default: false
-ansi-log
Enable/disable ANSI console logging
[...]
user deleting their .nextflow folder seem to have solved the issue. Previous content of folder might have been corrupted or held artifacts from runs with previous versions of Nextflow?
Bug report
I have a nextflow user who cannot manage to get any pipeline to run, always hitting the same bug when initiating the nextflow environment and more specifically when downloading the required dependencies. It precisely always happen when trying to download the nextflow-23.04.1.jar file, which puzzles me as it does not need it, the command being executed form that same very file to start with.
Why would nextflow try to download itself?
Annoyingly, it seems to be a user-specific issue as I can't reproduce this error with my own user, but I can't find anything specific in their environment that would cause such an issue. Disk space for downloading the artifacts does not seem to be an issue, even though I might be underestimating the size of the nextflow-23.04.1.jar file.
Can you please help me diagnose what is the issue here?
Expected behavior and actual behavior
that the pipeline runs, or at least that they could get the Nextflow usage message when doing
nextflow -h
from an environment that works fine (under my own user), we should get this:
Steps to reproduce the problem
This bug is observed on the Sanger Institute LSF HPC. The command actually run is nano-rave, executing the main.nf workflow; see here https://github.com/sanger-pathogens/nano-rave#running-on-the-farm-sanger-hpc-clusters
with the following commands, we observe the respective outputs in the section below:
Program output
Environment
Additional context
(Add any other context about the problem here)