Remote Access

Modified

2026-05-14

The recommended way to work on Bertha is via Positron or VS Code with a remote SSH connection. This gives you a full IDE experience — editor, terminal, file browser, extensions — running on the workstation while the UI runs on your laptop.

Positron is the preferred IDE, particularly for R and Python data science workflows (data viewer, variable explorer, plots pane). It includes the Remote SSH extension by default.

Prerequisites

  • An account on Bertha (contact the admin)
  • SSH key-based authentication set up
NoteSSH key setup

If you haven’t set up SSH keys before, follow the SSH key setup guide in the BIPS cluster docs. The instructions cover both macOS/Linux and Windows step by step. The process is identical — just use Bertha’s hostname instead of the cluster’s.

SSH config

Add Bertha to your local SSH config (~/.ssh/config on macOS/Linux, %USERPROFILE%\.ssh\config on Windows) so you can connect with a short name:

Host bertha
    HostName bertha.bips.de
    User <username>

Replace <username> with your Bertha username. After this, you can ssh bertha from a terminal or select bertha in your IDE’s Remote SSH connection dialog.

See the cluster docs SSH config section if you need help creating this file on Windows.

Connecting with Positron

  1. Open Positron
  2. Open the Command Palette (Ctrl+Shift+P on Linux/Windows, Cmd+Shift+P on macOS)
  3. Search for Remote-SSH: Connect to Host
  4. Select bertha from the list
  5. Positron opens a new window connected to the workstation

On the first connection, the IDE installs a server component on Bertha. Subsequent connections are faster.

TipInstall extensions on the remote side

When connected remotely, install extensions like “R”, “Python”, or “Jupyter” on the remote side. They persist across sessions.

Connecting with VS Code

  1. Install VS Code and the Remote - SSH extension (ms-vscode-remote.remote-ssh)
  2. Click the green button in the bottom-left corner (or Ctrl+Shift+P → Remote-SSH: Connect to Host)
  3. Select bertha

Working with Python

The Python extension automatically detects interpreters on the workstation. To use a uv-managed environment:

  • Open a project folder that contains a .venv — the extension detects it automatically
  • Or select the interpreter manually: Command Palette → Python: Select Interpreter → point to /home/<username>/project/.venv/bin/python

Working with R

R versions installed via rig are available at /opt/R/<version>/bin/R. Positron detects these automatically and lets you switch between versions. In VS Code, install the R extension and set r.rpath.linux if needed.

See R & RStudio for more details on R version management.

Port Forwarding

The IDE automatically forwards ports when a process on the workstation starts listening. This is useful for:

  • Jupyter: uv run jupyter lab — the forwarded URL appears in the terminal
  • Shiny apps: shiny::runApp() or shiny run — the local browser opens automatically
  • Streamlit, Gradio, etc.: same behavior

You can also manually forward ports via the Ports panel (Command Palette → Ports: Focus on Ports View).

Tips

  • Terminal: the integrated terminal runs on the workstation. All CLI tools (uv, rig, module, bertha, etc.) are available directly.
  • File editing: all file operations happen on the workstation. There is no sync step.
  • Long-running processes: if your SSH connection drops, the IDE reconnects automatically. But long-running computations should still be wrapped in tmux or screen to survive disconnects.
  • Connection keepalive: if connections drop frequently, add ServerAliveInterval 60 to your SSH config entry for Bertha.

RStudio Server (legacy)

RStudio Server is still available at http://bertha.bips.de:443 but is considered legacy. New users should prefer Positron for a more modern experience with better Python support and a similar R workflow.