I gave this a try with the binary rpm, it installed ok but when I run
it, the usage totals in the state summary at the top are wrong for me.
Anyone else seeing this odd behavior where clearly more than 0 nodes and
procs are used since there are many jobs running.
Usage Totals: 0/5360 Procs, 0/304 Nodes, 4641/9905 Jobs Running
Node States: 272 busy 7 down,unknown 23
free 2 offline
I am running slurm 14.11.3
This utility is the only way I know of getting a quick summary on the
usage totals, similar to the way showq in pbs/torque would display
something like this:
197 Active Jobs 197 of 1200 Processors Active (16.42%)
23 of 139 Nodes Active (16.55%)
Is there another way to get this in slurm?
Thanks
Marcin Sliwowski | SysAdmin@RENCI | 919-445-0479
On 08/25/2015 12:38 PM, Dennis McRitchie wrote:
Hi all,
This is just to announce that the long-lived and well-loved cluster
monitoring tool for PBS known as pbstop has been ported to SLURM. Based on
pbstop v4.16 released by Fedora, schedtop v5.01 supports both SLURM and PBS,
and also contains significant functional enhancements and bug fixes. When
built under SLURM it creates slurmtop.
Details and downloads can be found here:
http://bugs.schedmd.com/show_bug.cgi?id=1868
For those of you not familiar with pbstop, you can follow the above link to
see attached screenshots of the main window and the interactive help screen.
Best,
Dennis McRitchie
