fmriprep slice timing correction failed

Hi, I have been running a bunch of subjects using fmriprep (fmriprep/23.1.4) While most of my subjects run fine, slice timing error keeps occuring in some subject even I tried to rerun it in a fresh work directory. I've also checked my slice timings and they all seemed to be correct to me. Acquisition times are kind of the same across all individuals. No specific bad motions and artifacts have been spotted. So I am running out of ideas of what it could cause this. I've attached my slice timing crash report here.

Many thanks, Sarah crash-20231222-145204-sf674-slice_timing_correction-f867cc0a-b263-4618-934f-44a83dd8a5f6.txt

Dec 22 '23 21:12 sarahfan128

It says the job is getting killed. Likely this is an out of memory error.

Dec 22 '23 21:12 effigies

Thanks for the fast reply. I just double check on our server, we have enough memories...

Dec 22 '23 21:12 sarahfan128

Okay. Then I would try changing directory to the working directory of the failing command, then run it manually to see if you get more information.

Dec 22 '23 22:12 effigies

Sorry I am fairly new to fmriprep. Do I understand correctly that I re run this subject by going into the slice_timing_correction working directory (see below)? And what exactly do you mean by running it mannually to see? Like running the usual fmriprep preprocessing command? Many thanks. /home/sf674/palmer_scratch/predcode/mprc/20/work/fmriprep_23_1_wf/single_subject_M085678_wf/func_preproc_ses_A5512_task_kamin_acq_AP_wf/bold_stc_wf/slice_timing_correction

Dec 22 '23 22:12 sarahfan128

Hi Sarah, Can you try binding the tmpdir/scratch to your singularity or docker run like: -B $TMPDIR:$TMPDIR

Dec 23 '23 10:12 amirhusseinab