slurmd
Monitors all tasks running on the compute node, accepts tasks, launches tasks, and kills running tasks upon request. More information: https://slurm.schedmd.com/slurmd.html.
- Report node rebooted when daemon restarted (Used for testing purposes):
slurmd -b
- Run the daemon with the given nodename:
slurmd -N {{nodename}}
- Write log messages to the specified file:
slurmd -L {{path/to/output_file}}
- Read configuration from the specified file:
slurmd -f {{path/to/file}}
- Display help:
slurmd -h
