Stuck at proc_func

[hoptman]

• What are you trying to do?
• If you've tried running this with BrainSpace code already, how did you do so and what went wrong?
• Please tell us about your computing environment:
• Other information
  Hi all,
  I'm making progress on the data analysis, but I seem to be stuck at the -proc_func stage. I'm running micapipe v.0.2.2 in docker (ubuntu 20.04), and am running the command:
  docker run -ti --rm -v pwd/bids_mica:/bids -v pwd/micapipe:/out -v pwd/micapipe_tmp:/tmp -v /mnt/software/freesurfer_7.4.1/license.txt:/opt/licence.txt micalab/micapipe:v0.2.2 -bids /bids -out /out -fs_licence/opt/licence.txt -threads 10 -sub sub-A00035606 -ses ses-BAS1 -proc_func -acqStr task-rest_acq-AP_bold -dropTR 8 -NSR -noFIX
  I get the error:
  [ INFO ]..... Registering func MRI to nativepro

root @ 42abcf444798 Thu Feb 1 20:07:26 UTC 2024:
COMMAND --> mri_synthseg --i /tmp/14903_micapipe_proc-func_sub-A00035606_ses-BAS1/sub-A00035606_ses-BAS1_space-nativepro_desc-T1wbold.nii.gz --o /tmp/14903_micapipe_proc-func_sub-A00035606_ses-BAS1/T1bold_synthseg.nii.gz --robust --threads 10 --cpu
terminate called after throwing an instance of 'std::bad_alloc'
what(): std::bad_alloc
/opt/micapipe/functions/utilities.sh: line 1084: 3573 Aborted (core dumped) $l_command

Any thoughts?
Thanks,
Matt

[rcruces]

Hi, can you share the full log file please. It is saved under /logs

[hoptman]

Hi, can you share the full log file please. It is saved under /logs

Please see attached. I had removed the crash, but got the same error in the attached re-run.
proc_func_04-02-2024_se_task-rest_acq-AP_bold.txt

Thanks,
Matt

[rcruces]

Hello Matt,

It appears that the error is occurring due to a memory shortage on your workstation during the pipeline execution. The pipeline is a resource-intensive tool, typically run on robust desktops or SGE clusters. Could you please specify the location where you ran the pipeline? Additionally, consider increasing the allocated virtual memory to address this issue.

Best regards,

Raul

[hoptman]

Hello Matt,

It appears that the error is occurring due to a memory shortage on your workstation during the pipeline execution. The pipeline is a resource-intensive tool, typically run on robust desktops or SGE clusters. Could you please specify the location where you ran the pipeline? Additionally, consider increasing the allocated virtual memory to address this issue.

Best regards,

Raul

Hi Raul,
Thanks for your response!
I was running the command from /mnt/home2/Rockland_sample. How can I allocate more memory to the command? Is it something like --mem-gb? Also, how much memory should I allocate?

Thanks,
Matt

[hoptman]

Hi again,
I just wanted to check in about the issue of insufficient memory allocation:
Could you please tell me how I can allocate more memory to the docker micapipe command? Is it something like --mem-gb? Also, how much memory should I allocate?

Thanks,
Matt