Interactive apps
Available apps
- Accelerated visualization
- Desktop
- Jupyter
- Julia on Jupyter
- Jupyter for courses
- MATLAB
- MLflow
- RStudio
- TensorBoard
- Visual Studio Code
Launching an interactive app
- The interactive apps can be found in the navigation bar under Apps, or on My Interactive Sessions page.
- To launch an app on a compute node, select it from the menu.
- After selecting an interactive app from the list you will be presented with
a form to configure the session.
- Fill in appropriate billing project, partition, resources and app-specific settings.
- After submitting the app form, and the Slurm job for the app has finished queuing, the app will be started and you will be able to connect to the application on the My Interactive Sessions page (see below).
Avoid idle interactive sessions
Note that apps keep running and consuming resources even if you close the browser tab for the app session. To stop the app, you can cancel the session from the My Interactive Sessions page.
Troubleshooting
Several factors may cause your interactive session to fail to launch. Common reasons for failures are temporary issues with the shared file system (Lustre), job running out of memory, or invalid settings in the session configuration form (e.g. invalid Python installation for Jupyter). Below are some general tips for troubleshooting:
- File system issues come and go. The level of disk lag displayed on the web interface dashboard may give a hint if Lustre issues might be the reason for why your job failed to start. In this case there's not much that can be done, and the best option is typically to wait for a moment and then try re-launching the session.
- To rule out errors due to limited computing resources, try launching a new session requesting more memory or cores.
- Make sure to double-check that all the other settings in the configuration form are correct, e.g. path to a custom Python interpreter is correct, and that the environment contains all packages needed to launch the session (e.g. Jupyter).
- Information about a session is stored in a log file
output.log
, which can be viewed from the card of your session on the My Interactive Sessions page. Checking the contents of this file for error messages is a good starting point to understand why your job failed to launch. If you're unable to interpret the contents of the log file, don't hesitate to contact CSC Service Desk. Please include the log file of your failed session in your support request. This will help us to solve your issue faster. Other ways in which you can speed up the processing of your support request are discussed on the How to write good support requests page.