help
help > RE: slurm submission error
Sep 18, 2017 09:09 PM | Alfonso Nieto-Castanon - Boston University
RE: slurm submission error
Hi Jake,
That error indicates some problem with the specification of slice order or slice timings (typically a mismatch between the number of slices found in the functional data vs. the number of slices indicated in your "slices order" vector). Are you using the "manually defined" option for slice-acquisition order, and if so what values are you entering there?
Sorry some times the error messages reported by remote jobs are significantly less informative than those obtained when run locally (e.g. in this case CONN is detecting some problem with the slice orders and it is trying to give the user a chance to change those values, but unfortunately there is no GUI or user in the slurm nodes and CONN is failing to realize this, so this attempt fails with the node just waiting indefinitely for some user interaction). In any way, in general I would suggest to first try out the preprocessing steps on a single subject locally on your computer, and once everything seems to be working fine simply submit the same procedure for the remaining subjects to your slurm queue.
Hope this helps
Alfonso
Originally posted by Jake Rieke:
That error indicates some problem with the specification of slice order or slice timings (typically a mismatch between the number of slices found in the functional data vs. the number of slices indicated in your "slices order" vector). Are you using the "manually defined" option for slice-acquisition order, and if so what values are you entering there?
Sorry some times the error messages reported by remote jobs are significantly less informative than those obtained when run locally (e.g. in this case CONN is detecting some problem with the slice orders and it is trying to give the user a chance to change those values, but unfortunately there is no GUI or user in the slurm nodes and CONN is failing to realize this, so this attempt fails with the node just waiting indefinitely for some user interaction). In any way, in general I would suggest to first try out the preprocessing steps on a single subject locally on your computer, and once everything seems to be working fine simply submit the same procedure for the remaining subjects to your slurm queue.
Hope this helps
Alfonso
Originally posted by Jake Rieke:
Hey Alfonso,
I was able to kick-off the job in slurm by adding the MCRROOT variable, but it seems that the job fails during the slice-timing correction step. The error generated is here:
Preparing functional Slice-timing correction. Please wait...
Warning: uiwait is no longer supported when MATLAB is started with the -nodisplay or -noFigureWindows option or there is no display. For more information, see "Changes to -nodisplay and -noFigureWindows Startup Options" in the MATLAB Release Notes. To view the release note in your system browser, run web('www.mathworks.com/help/matlab/release-notes.html#br5ktrh-3', '-browser')
> In warnfiguredialog (line 21)
In uiwait (line 40)
In conn_msgbox (line 26)
In conn_setup_preproc (line 1413)
In conn_process (line 25)
In conn_jobmanager (line 666)
In conn (line 8679)
The job seems to get hung up at this point without explicitly failing. Could you please provide further guidance? Thanks in advance.
Best regards,
Jake
I was able to kick-off the job in slurm by adding the MCRROOT variable, but it seems that the job fails during the slice-timing correction step. The error generated is here:
Preparing functional Slice-timing correction. Please wait...
Warning: uiwait is no longer supported when MATLAB is started with the -nodisplay or -noFigureWindows option or there is no display. For more information, see "Changes to -nodisplay and -noFigureWindows Startup Options" in the MATLAB Release Notes. To view the release note in your system browser, run web('www.mathworks.com/help/matlab/release-notes.html#br5ktrh-3', '-browser')
> In warnfiguredialog (line 21)
In uiwait (line 40)
In conn_msgbox (line 26)
In conn_setup_preproc (line 1413)
In conn_process (line 25)
In conn_jobmanager (line 666)
In conn (line 8679)
The job seems to get hung up at this point without explicitly failing. Could you please provide further guidance? Thanks in advance.
Best regards,
Jake
Threaded View
| Title | Author | Date |
|---|---|---|
| Jake Rieke | Sep 15, 2017 | |
| Alfonso Nieto-Castanon | Sep 15, 2017 | |
| Jake Rieke | Sep 18, 2017 | |
| Jake Rieke | Sep 18, 2017 | |
| Alfonso Nieto-Castanon | Sep 18, 2017 | |
| Jake Rieke | Sep 20, 2017 | |