Troubleshooting

This page lists common problems and possible solutions.

1 Cannot connect by SSH

Check:

  • Are you connected to the correct network?
  • Is the hostname correct?
  • Is your username correct?
  • Is the workstation online?

Try:

ssh username@mod-linux-ws01

2 Permission denied

Check your username:

whoami

Check your groups:

groups

You may need to be added to the correct group.

3 Conda command not found

Check:

which conda

If nothing appears, conda may not be loaded in your shell.

Try logging out and logging in again.

4 Environment not found

List environments:

conda env list

Check the exact environment name.

5 Job is pending

Check queue:

squeue

Check job details:

scontrol show job JOBID

Common reasons:

  • not enough requested resources available
  • Slurm configuration issue
  • node is busy

6 Job ran but produced no output

Check your Slurm output file:

ls -lh
cat job-name.out
cat job-name.err

Check whether your script actually writes output.

7 Out of memory

If the error suggests memory problems, increase memory request:

#SBATCH --mem=8G

Use only as much as needed.

8 Disk full

Check disk space:

df -h

Check folder size:

du -sh .

Find large files:

find . -type f -size +1G

9 Command not found

Check whether the software is installed:

which command_name

Example:

which R
which python
which quarto

You may need to activate the correct environment.