Hi there!
I get strange errors when calling nextflow run helloworld.nf
on two different hosts where I installed it via micromamba.
On the Debian 12 it says:
library initialization failed - unable to allocate file descriptor table - out of memoryUnable to initialize nextflow environment
(Same error when trying nextflow info
).
On the CentOS 7.9 it says:
N E X T F L O W ~ version 0.24.2
Launching `src/helloworld.nf` [high_mirzakhani] - revision: b1230ea5c8
ERROR ~ Unknown process directive: `path`
Did you mean of these?
each
-- Check script 'helloworld.nf' at line: 5 or see '.nextflow.log' file for more details
But nextflow info
works.
To me it looks like the Java/Groovy dependencies that should have been installed by micromamba are not used.
Steps to reproduce:
-
Install micromamba:
curl -Ls https://micro.mamba.pm/api/micromamba/linux-64/latest | tar -xvj bin/micromamba ./bin/micromamba shell init -s bash -r ~/micromamba source ~/.bashrc # or restart terminal
I configure it like this:
micromamba config append channels bioconda micromamba config append channels conda-forge micromamba config set channel_priority_strict
-
micromamba create -n testnextflow nextflow
which apparently installs openjdk.It also creates a “classpath” file for nextflow at
$HOME/.nextflow/tmp/launcher/nextflow-one_24.04.4/$HOSTNAME/classpath-8269d21....
-
micromamba activate testnextflow
-
Save the tutorial workflow from here: Your first script — Nextflow v24.08.0-edge documentation to
helloworld.nf
-
nextflow run helloworld.nf
I don’t know if it’s a nextflow problem or a micromamba problem.
I have succeeded in installing nextflow using the recommended default method and also using conda
with --solver libmamba
.
I suppose the environment set up by the micromamba executable is incorrect.
Any ideas?
Thanks in advance!