Support

This page provides information on how to get help, troubleshoot common issues, and access relevant documentation for the HPC Knowledgebase.

Getting Help & Support

If you need assistance, you can reach out to the support team using the following methods:

Common Issues & Fixes

Below are some common issues and their solutions:

1. Unable to SSH into the Cluster

  • Ensure you are using an supported OpenSSH client (e.g., Linux terminal, Windows PowerShell, PuTTY, MobaXterm).

  • Verify your AU NetID through the PI.

  • Ensure your Duo two-factor authentication is set up.

  • Use the correct SSH command:

    $ ssh username@hpc-sub.augusta.edu

2. Job Submission Fails in Slurm

  • Check if your job script is formatted correctly.

  • Make sure the resources requested available in the Slurm partition

  • Check your job errors and logs:

Example: submit_quick_job.sh

#!/bin/bash
#SBATCH --job-name=scitkit_job          # Job name
#SBATCH --output=scitkit_output.txt     # Output log file
#SBATCH --error=scitkit_error.err       # Output error file

$ cat scitkit_error.err

3. Storage Quota Exceeded

  • Remove unnecessary files.

  • Move longer term data to /project storage.

  • Check /home directory quota:

    $ squota -u

  • Check usage with:

    $ du -sh ~/

4. Invalid or Outdated Slurm Partition Name

If a job fails with Invalid partition name specified, check the currently available partitions:

$ sinfo -a

Verify the partition used in your job script:

#SBATCH --partition=cpu_90g

Some partitions have restricted access. Check the required group with:

$ scontrol show partition <partition_name> | grep AllowGroups

Then check your group memberships:

$ groups

If AllowGroups=ALL, the partition is generally available. If a group is listed, you must be a member of that group to use the partition.

Note

Access to restricted partitions is granted based on workload requirements and resource availability. If you are unsure which partition is appropriate for your workload, contact the AUHPCS team at auhpcs_support@augusta.edu.

System Maintenance & Updates

  • Planned Maintenance: Updates and patches are applied on the first Saturday of every month.

  • Unscheduled Outages: Critical issues will be communicated via email and the system status page.

  • Upcoming Changes: Major changes will be announced at least two weeks in advance.

Regular maintenance windows ensure system security, updates, and optimizations. Below is the scheduled maintenance plan.

πŸ“… Monthly Maintenance

  • August 27, 2026

  • September 24, 2026

  • October 29, 2026

  • November 26, 2026

  • December 31, 2026

  • January 28, 2027

  • February 25, 2027

  • March 25, 2027

  • April 29, 2027

  • May 27, 2027

  • June 24, 2027

  • July 29, 2027

  • August 26, 2026

  • September 30, 2026

  • October 28, 2026

  • November 25, 2026

  • December 30, 2026

  • January 27, 2027

  • February 28, 2027

πŸ›  Semiannual Maintenance (Extended Downtime)

  • January 28 - January 30, 2026

  • June 29 - July 1, 2026

  • January 27 - January 29, 2027

πŸ”” Maintenance Details

  • Monthly maintenance will typically occur from 7:30 AM - 11:30 AM EST unless otherwise noted.

  • Semiannual maintenance involves more extensive updates and may include major system patches, hardware updates, and storage maintenance.

  • Emergency maintenance may be scheduled with a 24-hour notice if necessary.

User Guides & Additional Resources

For more detailed information, refer to the HPC Knowledgebase.

If you need further assistance, don’t hesitate to reach out to the support team!