You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've noticed fmriprep is continually failing on IU Quartz. The log files from the app suggest that there's an issue with the identification of the T1w input (see attached log file).
This appears to be an issue with the copying and/or symbolic linking of the T1w input data file. Specifically, to identify this, I went to the process directory on quartz and attempted to interact with the data, including using fslinfo and mrinfo from fsl and mrtrix3 respectively to pull header info. This lead to "No such file or directory" errors and other weirdness.
I then attempted to both download the data directly from the failed process on brainlife, and an rsync of the data from quartz to my local machine. With the brainlife download, only the .json file was returned. With rsync, I again received similar "No file" errors.
Fortunately, I had a successful job on a different resource (Bridges 2) with the same exact inputs to compare against. On Bridges2, the input data seemed to copy/link just fine.
I don't have much time today to look into the exact issue, so I'm just going to remove fmriprep from Quartz until we can get the issue resolved.
Hello,
I've noticed fmriprep is continually failing on IU Quartz. The log files from the app suggest that there's an issue with the identification of the T1w input (see attached log file).
This appears to be an issue with the copying and/or symbolic linking of the T1w input data file. Specifically, to identify this, I went to the process directory on quartz and attempted to interact with the data, including using fslinfo and mrinfo from fsl and mrtrix3 respectively to pull header info. This lead to "No such file or directory" errors and other weirdness.
I then attempted to both download the data directly from the failed process on brainlife, and an rsync of the data from quartz to my local machine. With the brainlife download, only the .json file was returned. With rsync, I again received similar "No file" errors.
Fortunately, I had a successful job on a different resource (Bridges 2) with the same exact inputs to compare against. On Bridges2, the input data seemed to copy/link just fine.
I don't have much time today to look into the exact issue, so I'm just going to remove fmriprep from Quartz until we can get the issue resolved.
Happy to provide the process data
slurm-848186.log
.
The text was updated successfully, but these errors were encountered: