Re: [HTCondor-users] Problem in running parallel program

[Carsten Aulbert <carsten.aulbert@xxxxxxxxxx>]

Hi

On 22.09.21 13:57, Rajagopala Reddy Seelam wrote:
> I have installed HTCondor in an workstation with 20 cores. The 
> installation went well and *condor_status* returns 20 free slots. I 
> could successfully run the examples given in the quick start manual. 
> Thank you very much for the program.

just a random guess, if you see 20 free slots, you probably have defined 
 20 slots with 1 CPU core each. This would explain, why request_cpus = 
1 works but request_cpus = 16 does not.

Try using a single partition-able slot, e.g.

NUM_SLOTS_TYPE_1                 = 1
SLOT_TYPE_1                      = cpus=100%, ram=100%, swap=0%
SLOT_TYPE_1_PARTITIONABLE        = True

I am a bit unsure about the "local" universe you use as this usually 
means the job is not run via startd but by the schedd  (others please 
correct me if I am wrong).

For initial testing, I would try using

universe = vanilla

I hope/guess it will work then.

Cheers

Carsten
--
Dr. Carsten Aulbert, Max Planck Institute for Gravitational Physics,
CallinstraÃe 38, 30167 Hannover, Germany, Phone +49 511 762 17185

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature