Hello,
There seems to be a problem with submitting batch jobs. When using sbatch with even a simple file like
#!/bin/bash
#SBATCH --job-name=trial
#SBATCH --partition=private-dpt-cpu
#SBATCH --time=00-00:01:00
#SBATCH -N 1 # total number of nodes
#SBATCH --mem=10G
echo Hello
I get an error saying
sbatch: error: Batch job submission failed: I/O error writing script/environment to file
. Somehow running srun echo hello
seems fine and works.
Can you please have a look?
Best
Bharath
1 Like
Hello, I’m getting the same error on shared-gpu partition, while everything seems fine on other —gpu partitions…
This is the error I get now
Batch job submission failed: Unable to contact slurm controller (connect failure)
2 Likes
also getting this error with sbatch, salloc, and squeue
1 Like
I am also getting the same error with sbatch on shared and public-cpu. Batch job submission failed: Unable to contact slurm controller (connect failure)
Dear all, the issue was this one and it is now solved: [2024] Current issues on HPC Cluster - #11 by Yann.Sagon
Best
Yann