Slurm offers several tools to inspect job activity, track resource usage, and check cluster status. This is especially useful when troubleshooting failed jobs, optimizing job submissions, or understanding resource consumption.
View job history with sacct:
sacct lets you inspect completed jobs. It shows details such as memory usage, start/end time, and job status.
[user@rocky-login011 ~]$ sacct -j JOBID
Customize output with additional fields:
[user@rocky-login011 ~]$ sacct -j JOBID --format="User,JobID,JobName,Partition,State,Time,Start,End,Elapsed,NodeList"
View jobs starting from a specific date:
[user@rocky-login011 ~]$ sacct --starttime=2025-06-21 --format="User,JobID,JobName,Partition,State,Time,Start,End,Elapsed,NodeList"
For a full list of sacct format options, see the Slurm sacct documentation.
Check current job queue with squeue:
Used to view jobs currently in the queue.
[user@rocky-login011 ~]$ squeue -u NETID
More options and filters can be found in the Slurm squeue documentation.
View available nodes with sinfo:
Used to check the availability and status of compute nodes in the cluster.
[user@rocky-login011 ~]$ sinfo
See more about sinfo in the Slurm sinfo documentation.