pbsnodes
the command shows the state of the nodes:
n1
state = job-exclusive,busy
np = 48
properties = dell
ntype = cluster
jobs = 0/88182.master, 1/88182.master, 2/88182.master, 3/88182.master, 4/88182.master, 5/88182.master, 6/88182.master, 7/88182.master, 8/88182.master, 9/88182.master, 10/88182.master, 11/88182.master, 12/88182.master, 13/88182.master, 14/88182.master, 15/88182.master, 16/88182.master, 17/88182.master, 18/88182.master, 19/88182.master, 20/88182.master, 21/88182.master, 22/88182.master, 23/88182.master, 24/88182.master, 25/88182.master, 26/88182.master, 27/88182.master, 28/88182.master, 29/88182.master, 30/88182.master, 31/88182.master, 32/88182.master, 33/88182.master, 34/88182.master, 35/88182.master, 36/88182.master, 37/88182.master, 38/88182.master, 39/88182.master, 40/88182.master, 41/88182.master, 42/88182.master, 43/88182.master, 44/88182.master, 45/88182.master, 46/88182.master, 47/88182.master
status = rectime=1446026760, varattr=, jobs=88182.master, state=busy, netload=12814551583, gres=, loadave=29.06, ncpus=48, physmem=198088144kb, availmem=196312616kb, totmem=198088144kb, idletime=6280885, nusers=1, nsessions=1, sessions=10168, uname=Linux n1 2.6.18-238.el5 #1 SMP Thu Jan 13 15:51:15 EST 2011 x86_64,opsys=linux
gpus = 0
...
n5
state = busy
np = 48
properties = dell
ntype = cluster
jobs = 0/88209.master, 1/88209.master, 2/88209.master, 3/88209.master, 4/88209.master, 5/88209.master, 6/88209.master, 7/88209.master, 8/88209.master, 9/88209.master, 10/88209.master, 11/88209.master, 12/88209.master, 13/88209.master, 14/88209.master, 15/88209.master, 16/88210.master, 17/88210.master, 18/88210.master, 19/88210.master, 20/88210.master, 21/88210.master, 22/88210.master, 23/88210.master, 24/88210.master, 25/88210.master, 26/88210.master, 27/88210.master, 28/88210.master, 29/88210.master, 30/88210.master, 31/88210.master
status = rectime=1446026758, varattr=, jobs=88209.master, state=busy, netload=13549296340, gres=, loadave=32.08, ncpus=48,physmem=198088144kb, availmem=116519628kb, totmem=198088144kb, idletime=6280923, nusers=1, nsessions=2, sessions=10416 10421, uname=Linux n5 2.6.18-238.el5 #1 SMP Thu Jan 13 15:51:15 EST 2011 x86_64,opsys=linux
gpus = 0
n6
state = free
np = 48
properties = dell
ntype = cluster
status = rectime=1446026751, varattr=, jobs=, state=free, netload=4343365038, gres=, loadave=0.02, ncpus=48, physmem=198088144kb, availmem=197636740kb, totmem=198088144kb, idletime=1782512, nusers=0, nsessions=? 0, sessions=? 0, uname=Linux n6 2.6.18-238.el5 #1 SMP Thu Jan 13 15:51:15 EST 2011 x86_64,opsys=linux
gpus = 0
n7
state = down
np = 48
properties = dell
ntype = cluster
status = rectime=1446026755, varattr=, jobs=, state=down, netload=4343365038, gres=, loadave=0.02, ncpus=48, physmem=198088144kb, availmem=197636740kb, totmem=198088144kb, idletime=1782512, nusers=0, nsessions=? 0, sessions=? 0, uname=Linux n6 2.6.18-238.el5 #1 SMP Thu Jan 13 15:51:15 EST 2011 x86_64,opsys=linux
gpus = 0
...
n11
state = offline
np = 48
properties = fast
ntype = cluster
status = rectime=1446031059, varattr=, jobs=, state=free, netload=2955488077, gres=, loadave=0.04, ncpus=48, physmem=264279116kb, availmem=263563112kb, totmem=264279116kb, idletime=95718, nusers=0, nsessions=? 0, sessions=? 0, uname=Linux n11 2.6.18-238.el5 #1 SMP Thu Jan 13 15:51:15 EST 2011 x86_64,opsys=linux
gpus = 0
...
In the above example:
- node n1 is occupied by one job using all cores
- node n5 is occupied by two jobs, each of them uses 16 cores
- node n6 is free
- node n7 is turned off (down); this occurs when you can not communicate with the node due to shutdown or suspension,
- node n11 is disabled (offline) from the queue system