| Summary: | Job just needs 1 core but won't run unless node is empty | ||
|---|---|---|---|
| Product: | Slurm | Reporter: | Torkil Svensgaard <torkil> |
| Component: | Configuration | Assignee: | Director of Support <support> |
| Status: | RESOLVED DUPLICATE | QA Contact: | |
| Severity: | 4 - Minor Issue | ||
| Priority: | --- | CC: | rkv |
| Version: | 20.11.0 | ||
| Hardware: | Linux | ||
| OS: | Linux | ||
| Site: | DRCMR | Alineos Sites: | --- |
| Atos/Eviden Sites: | --- | Confidential Site: | --- |
| Coreweave sites: | --- | Cray Sites: | --- |
| DS9 clusters: | --- | HPCnow Sites: | --- |
| HPE Sites: | --- | IBM Sites: | --- |
| NOAA SIte: | --- | OCF Sites: | --- |
| Recursion Pharma Sites: | --- | SFW Sites: | --- |
| SNIC sites: | --- | Linux Distro: | --- |
| Machine Name: | CLE Version: | ||
| Version Fixed: | Target Release: | --- | |
| DevPrio: | --- | Emory-Cloud Sites: | --- |
Submit never returned for 10332 but seems it was submitted *** This ticket has been marked as a duplicate of ticket 10332 *** |
From slurm.conf: " NodeName=bigger9 CPUs=48 Boards=1 SocketsPerBoard=1 CoresPerSocket=24 ThreadsPerCore=2 RealMemory=257552 Gres=gpu:1 PartitionName=debug Nodes=ALL MaxTime=INFINITE Shared=NO LLN=YES State=UP PartitionName=HPC Nodes=ALL Default=YES MaxTime=INFINITE Shared=NO LLN=YES State=UP PartitionName=application Nodes=ALL MaxTime=INFINITE Shared=YES LLN=YES State=UP " I then start a terminal on the application partition and submit a batch job, and the latter won't run until I clear the application queue even though it only needs a single core. What am I missing? torkil@joe:/home/torkil$ sinfo PARTITION AVAIL TIMELIMIT NODES STATE NODELIST debug up infinite 1 mix bigger9 HPC* up infinite 1 mix bigger9 application up infinite 1 mix bigger9 torkil@joe:/home/torkil$ squeue JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON) 746 HPC test_gpu torkil PD 0:00 1 (Resources) 745 applicati xfce4-te torkil R 57:51 1 bigger9 " Batch job submitted by sbatch: " torkil@joe:/home/torkil/slurm$ cat test_gpu.sh #!/bin/bash #SBATCH --partition=HPC #SBATCH --gres=gpu:1 #SBATCH --cpus-per-task=1 #SBATCH --ntasks=1 nvidia-smi " Mvh. Torkil