On 05/13/2016 02:48 PM, Rich Pieri wrote:
> On 5/13/16 9:08 AM, sjones wrote:
>> Yet the system continues to run the previous number, cpus=24.
> Try "NUM_CPUS = 32" to override the detected CPU count. As a rule of thumb, though, you should not allocate more CPUs or cores than you actually have.
Thanks. There are 2 cpus in the system, and each has 8 cores; so, with
hyperthreads, the system sees 32 processors:
# cat /proc/cpuinfo | grep processor | tail -n 1
processor : 31
I.e. 0 .. 31 = 32.
It was set up with cpus=24, but changed to cpus=32 (config file is
below). The trouble is, is still acts as if it has 24; e.g. it will run
3 x 8 core jobs, or 24 x 1 core jobs or anything between, but does not
run (e.g.) 4 x 8 core jobs or 32 x 1 core jobs etc.
# cat /etc/condor/config.d/00-node_parameters
RalNodeLabel = E52630V3
RalScaling = 1.295
NUM_SLOTS = 1
SLOT_TYPE_1 = cpus=32,mem=auto,disk=auto
NUM_SLOTS_TYPE_1 = 1
SLOT_TYPE_1_PARTITIONABLE = TRUE
# processors : 32
# numberOfCpus : 2
Steve Jones sjones@xxxxxxxxxxxxxxxx
Grid System Administrator office: 220
High Energy Physics Division tel (int): 43396
Oliver Lodge Laboratory tel (ext): +44 (0)151 794 3396
University of Liverpool
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
You can also unsubscribe by visiting
The archives can be found at: