Vuori User's Guide > Batch jobs and the batch system > Monitoring and displaying job and system status
Sisältö
 
Tehdyt toimenpiteet

Monitoring and displaying job and system status

This section gives examples how to monitor and display job, queue, host and SLURM server status

sinfo

View information about SLURM nodes and partitions. Shows, for example which nodes are allocated and which are free:

$ info
PARTITION AVAIL  TIMELIMIT  NODES  STATE NODELIST
parallel     up 7-00:00:00      1  drain n231
parallel     up 7-00:00:00    219  alloc n[4-29,32-66,68-79,87-109,111-122,129-230,232-240]
parallel     up 7-00:00:00     16   idle n[30-31,67,80-86,110,123-128]
serial*      up 7-00:00:00      1  drain n231
serial*      up 7-00:00:00    219  alloc n[4-29,32-66,68-79,87-109,111-122,129-230,232-240]
serial*      up 7-00:00:00     16   idle n[30-31,67,80-86,110,123-128]
longrun      up 21-00:00:0      1  drain n231
longrun      up 21-00:00:0    219  alloc n[4-29,32-66,68-79,87-109,111-122,129-230,232-240]
longrun      up 21-00:00:0     16   idle n[30-31,67,80-86,110,123-128]
interacti    up    4:00:00      1  alloc n1
interacti    up    4:00:00      2   idle n[2-3]

Examples of the sinfo command with different display options is given in Checking hardware composition and Resource monitoring (system and user).

squeue


By default command squeue shows all jobs which are submitted to the sheduler. If you want to see status of your own jobs, command squeue -u $USER -l is quite useful. You can also specify <jobid> with -j switch and with -p <partition> will display only jobs on that SLURM partition.