Fmriprep hangs with no error

Summary of what happened:

multiple docker instances, one for each subject, for fmriprep preprocessing. but all threads hang with no further processing. please refer to the uploaded log file for details.

Command used (and if a helper script was used, a link to the helper script or the command generated):

parallel run following commands (command for s013 was used as an example) by replacing the value of participant-label:

In Matlab:
parfor i=1:num_subject
cmd="docker run --rm -v E:\BIDS:/input:ro -v E:\fmriprep-out:/opt/fmriprep/subjects/ -v D:\license.txt:/usr/local/freesurfer/.license:ro -v E:\fmriprep-interm:/fmriprep-interm nipreps/fmriprep:23.1.4 /input /opt/fmriprep/subjects/ participant --participant-label s013 --fs-license-file /usr/local/freesurfer/.license --cifti-output --output-spaces fsaverage5 fsaverage6 MNI152NLin6Asym:res-2 --ignore slicetiming -w /fmriprep-interm"



logs.txt (367.7 KB)

By the way, I got enough memory (110GB left, as shown in docker) and disk space (2TB left).

Hi @Shimmering, some questions:

  1. How many subjects do you have in your dataset?
  2. How long did you wait before cancelling the job?
  3. Did you try only using a single fMRIPrep command and specify multiple participants in the --participant-label argument?
  4. Does it hang when you run fewer participants at a time (or just one participant)?