FAQ

Frequently asked questions about SynpixCloud GPU rental

Account & Billing

How do I add funds to my account?

  1. Go to Settings > Credits
  2. Click Top Up
  3. Enter the amount and complete payment via Stripe
  4. Credits are added instantly

What payment methods are accepted?

We accept credit and debit cards through Stripe, including Visa, Mastercard, and American Express.

Is there a minimum top-up amount?

Yes, the minimum top-up is $5.00 USD.

Can I get a refund?

Unused account credits are non-refundable. If you experience service issues, please contact support for resolution.

How does billing work?

SynpixCloud uses prepaid billing. The full rental amount is deducted when you create an instance. See our Billing Guide for details.


Instance Management

How long does it take to create an instance?

Most instances are ready within 1-2 minutes after creation.

Can I change my instance configuration after creation?

Yes, you can:

  • Upgrade CPU/Memory - Available anytime
  • Expand disk - Available anytime
  • Change GPU - Requires creating a new instance

What happens when my rental period ends?

  • With auto-renewal ON: Instance renews automatically if you have sufficient balance
  • With auto-renewal OFF: Instance status changes to "Expired" and will be stopped

How do I extend my rental period?

  1. Go to your instance dashboard
  2. Click Extend
  3. Select additional duration
  4. Confirm payment

What's the difference between Stop and Terminate?

ActionData PreservedCan ResumeBilling
StopYes (with retention)YesStorage fees only
TerminateNoNoNo further charges

Connection Issues

I can't connect via SSH. What should I check?

  1. Instance status - Must be "Running"
  2. Credentials - Verify host, port, and password
  3. Network - Check if your network blocks SSH (port 22)
  4. Firewall - Corporate networks may block custom ports
  5. Wait - New instances may take 1-2 minutes to become accessible

My SSH connection keeps dropping. How do I fix it?

Add to your ~/.ssh/config:

Host *
    ServerAliveInterval 60
    ServerAliveCountMax 3

How do I keep processes running after disconnecting?

Use screen or tmux:

# Start a screen session
screen -S mysession

# Run your command
python train.py

# Detach: Ctrl+A, D

# Reattach later
screen -r mysession

VSCode Remote won't connect. What should I do?

  1. Verify SSH works from terminal first
  2. Update the Remote-SSH extension
  3. Try: Remote-SSH > Kill VS Code Server on Host
  4. Check instance has enough memory (>2GB recommended)

GPU & Performance

How do I check if my GPU is working?

# Check GPU status
nvidia-smi

# Test CUDA
python -c "import torch; print(torch.cuda.is_available())"

My GPU is not being detected. What's wrong?

  1. Check drivers: Run nvidia-smi
  2. CUDA mismatch: Ensure PyTorch/TensorFlow version matches CUDA version
  3. Restart instance: Sometimes a reboot helps
  4. Contact support: If issue persists

How do I monitor GPU usage during training?

# Real-time monitoring
watch -n 1 nvidia-smi

# Or use nvtop
apt install nvtop
nvtop

What CUDA versions are available?

Pre-configured images include various CUDA versions (11.8, 12.1, 12.4). Check the image description when creating an instance.


Storage & Data

How do I upload large files?

Use rsync for reliable large file transfers:

rsync -avz --progress large_file.zip root@host:/root/ -e "ssh -p <port>"

My disk is full. What should I do?

  1. Check usage: du -sh /* | sort -hr
  2. Clear caches:
    pip cache purge
    conda clean --all
    apt clean
  3. Remove unnecessary files
  4. Expand disk if needed

How do I download HuggingFace models faster?

# Set mirror (especially useful in China)
export HF_ENDPOINT=https://hf-mirror.com

# Then download
huggingface-cli download model-name

Will I lose my data if I stop the instance?

  • With disk retention: Data is preserved
  • Without disk retention: Data is lost when terminated

Software & Environment

What software is pre-installed?

Depends on the image you choose. Common pre-installed packages include:

  • NVIDIA drivers and CUDA
  • Python 3.x with pip
  • PyTorch or TensorFlow (image-dependent)
  • Jupyter Lab
  • Common utilities (git, wget, vim, etc.)

How do I install additional packages?

# Python packages
pip install package-name

# System packages
apt update && apt install package-name

# Conda packages
conda install package-name

Can I use Docker?

Yes, Docker is available on most images:

# Check Docker status
docker --version

# Run a container
docker run -it --gpus all nvidia/cuda:12.1-base nvidia-smi

Security

Is my data secure?

  • Instances are isolated from each other
  • SSH connections are encrypted
  • We don't access your data without permission

How do I change my instance password?

# On the instance
passwd root

Or use SSH key authentication for better security.

Can I restrict access to my instance?

Yes, you can:

  • Change the root password
  • Use SSH key authentication
  • Configure firewall rules within the instance

Support

How do I contact support?

Email us at support@synpixcloud.com

What information should I include in support requests?

  • Instance ID
  • Error messages or screenshots
  • Steps to reproduce the issue
  • What you've already tried

Where can I report bugs?

Please email support@synpixcloud.com with details about the bug.