For running a parallel job in a supercomputer, the login host IP is different from the head node IP assigned to the job. From the GUI set up, the $command is exported as $c onto the script slurm.sh and then the system runs the job.
I submitted a job this way, but it failed without generating a log file. I could not find where a copy of the $command was saved. Did any of you know where to look?
Thanks.
Hi,
I am not sure i completely understand your question? How the job is run depends on your run profile, what are you using as a run command?
Tim
Hello Tim,
Currently, We are trying to run CTFFIND. Finally we figure out how to run the jobs in slurm mode but still having problem with connection timeout and jobs stop but when we restart the job it start from scratch and redo all the finished one as well. Is there a way we can start where we lost the connection so we do not have to redo the finished one something similar as in relion "--only_do_unfinished".
Thanks
Gaya
How we change the scale for the 1D CTF fit results? I want to change the maximum Y axis to 0.5 from default 1.0 ?
Thanks
Gaya
Hi Gaya,
There is no way to do only unfinished, the only thing you can do is break your job into a few groups and run them seperately. Are the jobs crashing?
You cannot change the scale for the 1D fits in the GUI. The text files that are plotted with the data are available in the Assets/CTF folder though, so you can replot them yourself if you would like.
Thanks,
Tim
Thanks for all the information.
Is there a way to export selected images list after the CTF fitting. I can export particles but cannot find any option to generate the list of micrographs. Actually I want to remove the micrographs which are not good to free some space. Though, I was able to sort out from the particles star file, but if we can export that will make it very handy.
Thanks
Gaya
You can only do this by extracting the information directly from the database. Some examples of how to do this (including geting all images in a certain group) is shown here :-
https://cistem.org/documentation#tab-1-16