[RESOLVED] Hcp_freesurfer preprocessing issues

Issues
1

Description:

For certain subjects, hcp_freesurfer keep stuck at tmp error or just end up showing error. This issue is not resolved even after deleting the hcp folder and rerun again.

Call:

qunex_container hcp_freesurfer \
  --batchfile="/vast/palmer/pi/pushkarskaya/BlackThorn/processing/batch_hcp_03132926.txt" \
  --sessionsfolder="/vast/palmer/pi/pushkarskaya/BlackThorn/sessions" \
  --parsessions="1" \
  --overwrite=“yes” \
  --sessions="pb4074,pb7058,pb7399" \
  --container="/gpfs/gibbs/pi/n3/software/Singularity/qunex_suite-1.3.0.sif" \
  --scheduler="SLURM,time=1-00:00:00,ntasks=1,cpus-per-task=2,mem-per-cpu=20000,partition=day"

qunex_container hcp_freesurfer \
  --batchfile="/vast/palmer/pi/pushkarskaya/PVDM/processing/batch_hcp_04232026.txt" \
  --sessionsfolder="/vast/palmer/pi/pushkarskaya/PVDM/sessions" \
  --parsessions="1" \
  --overwrite=“yes” \
  --sessions=“pb10043,pb10416" \
  --container="/gpfs/gibbs/pi/n3/software/Singularity/qunex_suite-1.3.0.sif" \
  --scheduler="SLURM,time=1-00:00:00,ntasks=1,cpus-per-task=2,mem-per-cpu=20000,partition=day"

Logs:

/vast/palmer/pi/pushkarskaya/BlackThorn/processing/logs/comlogs/error_hcp_freesurfer_pb4074_2026-04-29_16.31.11.214475.log
/vast/palmer/pi/pushkarskaya/BlackThorn/processing/logs/comlogs/tmp_hcp_freesurfer_pb7399_2026-04-29_16.31.11.218038.log
/vast/palmer/pi/pushkarskaya/BlackThorn/processing/logs/comlogs/error_hcp_freesurfer_pb7058_2026-04-29_16.31.11.215028.log
/vast/palmer/pi/pushkarskaya/PVDM/processing/logs/comlogs/tmp_hcp_freesurfer_pb10043_2026-05-11_06.39.41.895100.log
/vast/palmer/pi/pushkarskaya/PVDM/processing/logs/comlogs/tmp_hcp_freesurfer_pb10416_2026-05-11_06.39.41.904401.log
2

Hi,

I no longer have access to your infrastructure. Could you please upload the mentioned logs here as attachments.

The ones that are stuck in tmp probably hit the walltime, try allocating 2 days for the job instead of 1, so time=2-00:00:00.

Are these sessions with poor quality or not healthy data? We had some issues with such data ourselves and using FreeSurfer 8 seems to resolve most of them. We are in the process of adding FreeSurfer 8 support to QuNex. Hopefuly in May 2026, but it is not completely in our hands unforunately, so this could take a bit more than that.

Best, Jure

3

Hello,

Thank you for your reply. I believe these sessions are fine data as they were previously preprocessed but we are preprocessing them again with newer version of QuNex.

Please find the attached files for the logs. One of the error log is too big for upload but the error is the same as the uploaded one. I will try to specify longer schedule meanwhile to try. Thank you for your kind help.

Best,
Po-Kai

tmp_hcp_freesurfer_pb10416_2026-05-11_06.39.41.904401.log (114.9 KB)

tmp_hcp_freesurfer_pb10043_2026-05-11_06.39.41.895100.log (116.9 KB)

tmp_hcp_freesurfer_pb7399_2026-04-29_16.31.11.218038.log (160.6 KB)

error_hcp_freesurfer_pb4074_2026-04-29_16.31.11.214475.log (6.6 MB)

4

Hm,

The log keeps saying that a file is missing, maybe there are some issues with the study. Can you reprocess from scratch. Going back and forth between post freesurfer and freesurfer should not be done and will produce invalid results. So if you want to redo structural processing, you need to repeat the whole freesurfer chain (hcp_pre_freesurfer, hcp_freesurfer and hcp_post_freesurfer).

It also says additional details can be found at

/vast/palmer/pi/pushkarskaya/BlackThorn/sessions/pb4074/hcp/pb4074/T1w/pb4074/scripts/recon-all.log

can you please upload that. There is nothing that would point to a QuNex issues in the attached logs.

Best, Jure

5

Hello,

I did reprocess it but it still showed error. I will try to reprocess again.

Please find the attached file for the log, thank you.

recon-all.log (6.6 MB)

Best,
Po-Kai

6

Hi,

Unfortunately, there is nothing new in this log that would help us out. Seems like some files are missing, that are needed by a downstream command but it is hard to say which ones.

The only thing that pops to my mind is that you zip and upload this session in an unprocessed state and share it with me (if you legaly can), so I can try this on my end. It looks like we won’t be able to resolve this “remotely”. You can also ask if someone in Cho lab can help you out, they are the QuNex developers at Yale now.

Best, Jure

7

Hi,

Any updates here?

Best, Jure