Hello,

the docu for heterogeneous jobs [1] says that the envVar SLURM_JOB_ID should be different for each component. However, I cannot reproduce this on a fresh slurm-19.05.1 installation.

$ salloc -pcompute -N1 : -pcompute2 -N1
[...]
salloc: Granted job allocation 108453
[...]
bash-4.1$ squeue -h -j 108453 -o %i,%P
108453+0,compute
108453+1,compute2

bash-4.1$ srun --pack-group=0 printenv SLURM_JOB_ID
108453
bash-4.1$ srun --pack-group=1 printenv SLURM_JOB_ID
108453

As you can see, for both components the SLURM_JOB_ID is the same. Does someone know what might be wrong in my setup?

Moreover, I would like to know how to distinguish different components of a heterogeneous job in the epilog script. As far as I can see there is only a limited set of envVars available and none can be used to determine the component - again the SLURM_JOB_ID is the same in all epilog!

Thanks for your help,
Hendryk

[1] https://slurm.schedmd.com/heterogeneous_jobs.html

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

Reply via email to