Connecting to Instances

Connection Methods

SynpixCloud instances support multiple connection methods:

• SSH - Command-line access via terminal
• VSCode Remote - Full IDE experience
• Jupyter Notebook - Browser-based Python environment

SSH Connection

Basic SSH Connection

1. Get your connection details from the instance dashboard
2. Open your terminal and run:

ssh -p <port> root@<host>

3. Enter the root password when prompted

SSH on Different Systems

macOS / Linux

Use the built-in Terminal app:

ssh -p 12345 root@gpu.synpixcloud.com

Windows

Use PowerShell, Windows Terminal, or install an SSH client:

ssh -p 12345 root@gpu.synpixcloud.com

Or use PuTTY for a graphical SSH client.

SSH Key Authentication

For passwordless login, add your public key:

# On your local machine
cat ~/.ssh/id_rsa.pub

# On the remote instance
mkdir -p ~/.ssh
echo "your-public-key" >> ~/.ssh/authorized_keys
chmod 600 ~/.ssh/authorized_keys

Keeping SSH Sessions Alive

To prevent disconnection during long operations, configure SSH keep-alive:

# Add to ~/.ssh/config on your local machine
Host *
    ServerAliveInterval 60
    ServerAliveCountMax 3

Running Background Processes

Use screen or tmux to keep processes running after disconnection:

Using Screen

# Start a new screen session
screen -S training

# Run your command
python train.py

# Detach: Press Ctrl+A, then D

# Reattach later
screen -r training

Using tmux

# Start a new tmux session
tmux new -s training

# Run your command
python train.py

# Detach: Press Ctrl+B, then D

# Reattach later
tmux attach -t training

VSCode Remote Connection

Connect to your instance directly from Visual Studio Code for a full IDE experience.

Setup

1. Install VSCode
2. Install the Remote - SSH extension

Connect

1. Press Ctrl+Shift+P (or Cmd+Shift+P on macOS)
2. Type Remote-SSH: Connect to Host
3. Enter: root@<host> -p <port>
4. Enter password when prompted

SSH Config for Easy Access

Add to your ~/.ssh/config:

Host synpix-gpu
    HostName gpu.synpixcloud.com
    Port 12345
    User root

Now connect with just:

ssh synpix-gpu

Or select "synpix-gpu" in VSCode's Remote-SSH hosts.

Troubleshooting VSCode Connection

If VSCode fails to connect:

1. Check SSH access - Verify you can connect via terminal first
2. Update Remote-SSH - Ensure the extension is up to date
3. Check server requirements - Instance needs sufficient memory for VSCode Server
4. Clear server cache - Run "Remote-SSH: Kill VS Code Server on Host"

Jupyter Notebook

Access Jupyter Notebook through your browser for interactive Python development.

Accessing Jupyter

If your instance has Jupyter pre-installed:

1. Find the Jupyter URL in your connection details
2. Open the URL in your browser
3. Enter the token/password if prompted

Manual Jupyter Setup

If Jupyter is not pre-installed:

# Install Jupyter
pip install jupyterlab

# Start Jupyter with remote access
jupyter lab --ip=0.0.0.0 --port=8888 --no-browser --allow-root

Jupyter Security

When running Jupyter on a remote server:

# Set a password
jupyter lab password

# Generate config
jupyter lab --generate-config

# Edit config to require password
# Set c.ServerApp.password in jupyter_lab_config.py

Port Mapping

Map internal ports to external access for web services, APIs, or custom applications.

Common Port Mappings

Configuring Port Mapping

1. Go to instance settings
2. Navigate to Port Mapping
3. Add your internal port
4. Note the assigned external port

Access your service at: http://<host>:<external-port>

Connection Troubleshooting

Cannot Connect via SSH

1. Check instance status - Ensure it's "Running"
2. Verify credentials - Double-check host, port, and password
3. Check network - Ensure your network allows outbound SSH
4. Wait for startup - New instances may take 1-2 minutes

Connection Drops Frequently

1. Configure SSH keep-alive (see above)
2. Use screen or tmux for important processes
3. Check your local network stability

Slow Connection

1. Check your local network bandwidth
2. Consider using a VPN if you're far from the server region
3. For large file transfers, use rsync with compression:

rsync -avz --progress local_file root@host:remote_path -e "ssh -p <port>"

Support

Connection issues? Contact us at support@synpixcloud.com