Skip to content

21centuryweather/node-status

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 

Repository files navigation

node-status

Summarise PBS node status and resources from pbsnodes -v / pbsnodes -av.

pbs_node_status.sh --help:

Modes:
    (default) Per-node listing
    --summary  Aggregate counts of ONLINE/OFFLINE and INUSE/IDLE nodes by a grouping key

Options:
  --online     Show only nodes considered ONLINE
  --offline    Show only nodes considered OFFLINE
  --summary    Print counts instead of per-node rows
  --by         Grouping key for --summary (default: class)
               class   = SUBMIT queue(s) (from TYPE mapping) if known, else TYPE
               queue   = SUBMIT queue(s) inferred from TYPE using the mapping below
               pbsqueue= PBS node attribute "queue" (rarely present on this system)
               type    = derived from node name (e.g. cpu-clx, gpu-v100, gpu-h200)
  -h, --help   Show this help

Notes:
  - STATUS is derived from PBS state. If state contains offline/down/unknown, STATUS=OFFLINE.
  - TYPE is derived from the node name (fallback to topology if needed).
  - INUSE/IDLE are computed from assigned resources:
      INUSE = resources_assigned.ncpus > 0 OR resources_assigned.ngpus > 0 (ONLINE nodes only)
      IDLE  = neither (ONLINE nodes only)

Example usage

pbs_node_status.sh --summary --by queue:

SUBMIT_QUEUE        ONLINE  OFFLINE  INUSE  IDLE  TOTAL
normal/express      2692    332      1217   1475  3024
normalbw/expressbw  782     22       294    488   804
normalsr/expresssr  720     0        187    533   720
normalsl            180     12       28     152   192
gpuvolta            160     0        160    0     160
hugemem             47      3        18     29    50
gpuhopper           29      1        29     0     30
NA                  12      0        0      12    12
analysis            10      0        3      7     10
hugemembw           9       1        8      1     10
biodev              9       0        0      9     9
copyq               6       0        6      0     6
gpursaa             2       2        2      0     4
megamem             4       0        3      1     4
dgxa100             2       0        2      0     2
megamembw           2       0        0      2     2

About

Summarise Gadi node's status and resources

Resources

License

Stars

Watchers

Forks

Packages

 
 
 

Contributors

Languages