Support Ticket

• Submit requests via email to itservicedesk@surrey.ac.uk.

• Please use the templates below to ensure we have all the necessary information to help you quickly.

• Choose the template’s subject for your email subject to help us route your ticket faster.

Ticket Templates

1. Select the most relevant template for your issue to structure your ticket.

2. Copy, paste, and fill in the details.

3. Don’t forget to set the appropriate subject line

Requesting Access

Please also CC your supervisor/line manager when you submit this kind of ticket.

Subject: [HPC] - Access Request

Full name
Surrey username
Faculty name:
Department:
Research group:
Supervisor or line manager:
Cluster(s) requested:
Brief description of intended use:

Storage and Dataset Request

If you’d like to request research data storage for your project, please use the storage request form.

If you’d like to have a copy of a read-only dataset on the WEKA storage area or would like to purchase some writable, persistant scratch space (similar to the user areas), then please use the following format:

Subject: [HPC] - Datasets on WEKA for AI Surrey

Project or dataset name:
Existing directory name:
Expected size (GB or TB):
Access type (read-only / read-write):
Cost code to be charged (for writable scratch spaces):
Expected duration / retention:
Brief description of dataset/purpose:

NB: WEKA storage currently costs £500/TB/year.

Software Request

Subject: [HPC] - New Software Request

Software name and version:
Source (download link / git repository):
Target cluster(s):
Have you tried installing in user space? (Yes/No - if yes, what issues occurred)
License information (if applicable):
Number of expected users:
Dependencies required:

Something Not Working

Subject: [HPC] - Problem Running Jobs

What did you do? (commands/scripts run):
What happened? (error messages, unexpected behavior):
What did you expect to happen?:
Job IDs and submission times:
Script paths and configuration files:
Which cluster:
Modules loaded:
Container images used (if any):
Hostname(s) where issue occurred:
When did this last work? (if it ever did)

GitLab / Container Registry

Subject: [GitLab/Registry] - Issue Description

Project/repository/image name:
Action that failed: (push, pull, login, clone, etc.)
Error messages (full text):
When did this issue first occur?:
What was the last successful operation?:
Browser/client used (if applicable):
Connection method (SSH/HTTPS):

RemoteLabs / OpenOnDemand

Subject: [HPC] - RemoteLabs/OpenOnDemand Issue

Which service?:
How are you connecting? (VPN, direct browser, SSH tunnel):
Location: (On-campus/Off-campus/specific building if on-campus/abroad):
Operating system and browser (if web-based):
Error message or screenshot:
When did it last work?:
What changed since it last worked?:

Research Software Engineering (RSE) Support

Use this for help with code optimization, debugging custom scripts, or software design advice.

Subject: [RSE Support] - Brief Description of Code/Issue

Project Context
- What is the code supposed to do?:
- Programming languages, frameworks & versions:
- Code location (GitLab/GitHub link or path):

The Problem / Request
- Type of help: (Debugging / performance optimization / design advice)
- Describe the issue or goal:

Reproducibility (Crucial for debugging)
- Can we run this code ourselves to test it? (Yes/No):
- Command/script to reproduce the issue:
- Exact error messages / output:

Environment
- Which cluster, machine, or compute environment are you on?:
- Modules or environments (e.g. Conda) loaded:

Software License Issue

Subject: [Research Computing] - License Issue

Software name and version:
License type: (floating/node-locked/site license/unknown):
License server (if known):
Error messages (exact text):
Number of users affected:
System type: (laptop/cluster/workstation):
Hostname of affected system:
When did the license last work?:
Are you within your license terms? (user count, etc.):

General Issue

Subject: [HPC/Gitlab/Research Computing] - Brief Issue Description

Issue Summary:
- What is the problem?
- What did you do?
- What happened?
- What did you expect?

Relevant Info:
- Job IDs, submission scripts, storage paths, hostnames, error messages

Additional Notes:
- Have you checked docs/Google?
- Any recent changes?

Seeking Advice

Subject: [Research Computing] - Seeking advice

What are you trying to achieve? (specific goal):
What type of input do you need? (technical advice/best practices/resource recommendations):
Your current approach or plan:
Constraints or requirements:
Timeline or deadlines:
Your experience level with this type of work:

Equipment Purchase

Subject: [Research Computing] - Equipment Purchase

Type of equipment:

Research details:
- Research group/department:
- Principal investigator:

Technical requirements:
- Primary use case: (ML/AI training, simulations, data analysis, etc.)
- Software to be used:
- Expected number of users:
- Storage needs:
- Special requirements: (GPU type, memory, networking, etc.)

Budget and timeline:
- Approximate budget range:
- Purchase timeline/deadline:

Integration needs:
- Does this need to integrate with existing HPC systems? (Yes/No)
- Network/infrastructure requirements:

Any other specific requirements or constraints:

Grant / Bid Preparation Support

Process & expectations for this ticket type

• Who helps you: This ticket can reach both Research Computing Admins and Research Software Engineers (RSEs).

• The timeline: We aim to provide an initial response within a few working days.

• The workflow: Triage ➔ Consultation (email/Teams) ➔ Iteration.

• Lead time: Please engage us as early as possible before your deadline. Late requests may limit the depth of support we can provide.

Subject: [Research Computing] - Bid Support - Project Name / Acronym

Project Details
- Funding Body / Call Name:
- Submission Deadline:
- Principal Investigator:
- Brief Project Summary:
- Does this project involve sensitive or personal data (e.g. medical, identifiable)? (Yes/No/Unsure):
- Are there known compliance requirements? (e.g. TRE, DPIA, GDPR):

Infrastructure Needs (Admins)
(Leave blank if not applicable)
- Compute needs (e.g. HPC / GPU / High Memory / Standard):
- Storage requirements (Volume size & Type):
- Computing environment (Local cluster, Cloud, TRE, or Unsure):

Research Software Needs (RSEs)
(Leave blank if not applicable)
- Type of help (Design / Development / Optimization / Maintenance):
- Languages / tools (e.g. Python, C++, specific frameworks):
- Code status (Starting from scratch / Existing code):
- Do you need help drafting technical / work packages? (Yes/No):
- Do you need a software management plan / open science statement? (Yes/No):

Costing & Budget
(Please indicate what needs to be formally costed in the grant)
- Dedicated hardware/compute nodes? (Yes/No):
- Cloud computing costs? (Yes/No):
- RSE staffing time? (Yes/No):
- If RSE time: Approx % FTE or months needed (if known):
- Any funder budget caps/constraints?:

Etc
- Anything else you deem relevant.

Best Practices

You can help us and yourself by following these best practices when writing your ticket submission.

📖 Do your research before submitting a support request

• What does the documentation or user guide say?

• Have you searched the internet for key phrases from your jobs log files / error messages?

• What have you changed since the last time your job succeeded?

🎯 Be precise in your description of your problem

• Use the correct template and fill in all the requested information

• Include exact error messages (don’t paraphrase)

• Provide enough detail for us to recreate or understand your issue: job IDs, file paths, commands run

• Never include passwords, API keys, or sensitive data in tickets

🤔 Manage your expectations

• Consultants can address system issues and answer cluster/job scheduler questions, but they cannot teach parallel programming.

• They may offer advice that might help you to build, debug or optimise your code but, you should not expect these things to be done for you.

⏱ Be patient

• It may take a day for a consultant to respond to your support request, especially if your issue is complex.

• It may take a few messages before you and the consultant are on the same page.

• If your jobs are paused/held by the admins, it is not punitive. When systems are being impacted or in danger of crashing the admins may not have time to notify you before acting.