slurm - sinfo command does not show updated information about allocated nodes -




i wonder why sinfo command shows different information returned squeue. have experienced on several occasions number of allocated nodes returned sinfo did not match allocated nodes showed in squeue.

has noticed issue?

i'm using version 17.02, in previous versions i've experienced behavior.

for instance, i've executed squeue , sinfo @ same moment @ get:

thu aug 24 10:12:50 2017   jobid partition             name st       time  nodes nodelist(reason)   57011   dmrtest        jacobi-98  r       3:02      4 s07r2b[15-18]   57012   dmrtest            cg-99  r       3:02      4 s07r2b[11-14]   57010   dmrtest        jacobi-98  r       3:24     32 s01r1b[39,42,45-48],s07r2b[09-10],s09r2b[54-58,63-64,67],s14r1b[61-62,65-67],s14r2b[49,54- 57,61-63,65],s24r2b[57,59]   57008   dmrtest         nbody-96  r       3:52      8 s09r1b[53-55,57-59,61-62]   57009   dmrtest         nbody-97  r       3:52      4 s10r2b[49-50],s14r2b[69,72]   57007   dmrtest         nbody-95  r       4:21      4 s24r2b[49-52]   57006   dmrtest            cg-94  r       4:26      4 s14r1b[49-52]   57004   dmrtest            cg-92  r       4:42      4 s01r1b[33-35,37]   57003   dmrtest         nbody-91  r       4:45      4 s10r2b[61,63-65]   57001   dmrtest         nbody-89  r       4:46      8 s10r2b[51-52],s14r2b[50-53],s24r2b[53-54]   57002   dmrtest            cg-90  r       4:46      2 s24r2b[62-63]   57000   dmrtest         nbody-88  r       5:00      8 s09r2b[49,51-53],s14r1b[57-60]   56999   dmrtest         nbody-87  r       5:02      4 s07r2b[19-20,22-23]   56997   dmrtest            cg-85  r       5:36      8 s01r1b[29-32],s10r2b[53-56]   56995   dmrtest            cg-83  r       7:08      2 s10r2b[66-67]   56988   dmrtest        jacobi-76  r       8:28      2 s14r1b[53-54]   56985   dmrtest            cg-74  r       8:36      4 s09r2b[68-70,72]   56978   dmrtest            cg-69  r       9:37      2 s14r1b[55-56]   56976   dmrtest            cg-67  r       9:40      2 s14r1b[63-64]   56974   dmrtest            cg-65  r      10:49      4 s10r2b[69-70],s24r2b[64-65]   56973   dmrtest            cg-64  r      10:58      4 s09r1b[49-52]   56971   dmrtest            cg-62  r      11:22      2 s14r2b[67-68]   56969   dmrtest            cg-60  r      11:35      2 s24r2b[55-56] 

if count nodes of squeue command, see 114 allocated nodes, if check sinfo, have is:

partition avail  timelimit  nodes  state nodelist dmrtest*       infinite     94  alloc s01r1b[29-35,37,47-48],s07r2b[09-20,22-23],s09r1b[49-55,57-59,61-62],s09r2b[49,51-53,68-70,72],s10r2b[49-56,61,63-67,69-70],s14r1b[49-60,63-64],s14r2b[50-53,67-69,72],s24r2b[49-56,62-65] dmrtest*       infinite     34   idle s01r1b[39,42,45-46],s09r1b[63-66],s09r2b[54-58,63-64,67],s14r1b[61-62,65-67],s14r2b[49,54-57,61-63,65],s24r2b[57,59-61] 

only 94 allocated nodes.





wiki

Comments

Popular posts from this blog

Asterisk AGI Python Script to Dialplan does not work -

python - Read npy file directly from S3 StreamingBody -

kotlin - Out-projected type in generic interface prohibits the use of metod with generic parameter -