Linux command
slurmd 命令
文本
复制后可按需替换文件名、目录或参数。
常用示例
Report
slurmd -b
Run
slurmd -D -v
Run
slurmd -N [nodename]
Write
slurmd -L [path/to/output_file]
Read
slurmd -f [path/to/file]
Clear
slurmd -c
说明
slurmd is the Slurm compute node daemon. It monitors all tasks running on the compute node, accepts new tasks from the central controller, launches tasks, and terminates running tasks upon request. Each compute node in a Slurm cluster runs slurmd to participate in workload scheduling and execution. The daemon communicates with slurmctld (the central controller) for job management.
参数
- -b
- Report node rebooted when daemon restarts
- -N _nodename_
- Run with specified node name
- -L _logfile_
- Write log to specified file
- -f _config_
- Read configuration from specified file
- -c
- Clear previous node state
- -D
- Run in foreground (don't daemonize)
- -v
- Verbose logging
- -M
- Run the daemon with multiple node names (must have unique port and NodeAddr)
- -h
- Display help information
- -V
- Display version information
FAQ
What is the slurmd command used for?
slurmd is the Slurm compute node daemon. It monitors all tasks running on the compute node, accepts new tasks from the central controller, launches tasks, and terminates running tasks upon request. Each compute node in a Slurm cluster runs slurmd to participate in workload scheduling and execution. The daemon communicates with slurmctld (the central controller) for job management.
How do I run a basic slurmd example?
Run `slurmd -b` in a terminal, then adjust file names, paths, flags, or remote targets for your system.
What does -b do in slurmd?
Report node rebooted when daemon restarts