If you get an error 500 after clicking on the launcher icon, this could be due to RStudio taking too much time to start, which is interpreted as a failure by JupyterLab. A good place to start is looking at the log file from jupyter, for the current session: $ cat ~/.jupyter/.jupyterhub_$.log Troubleshooting
If this does not work, you will need to investigate the errors. If your RStudio session does not start, try to reload the page, in case the reported failure is just due to the container taking too much time to start. Then restart your Jupyter session and launch a new RStudio session to make use of your container. Once your container is ready, upload it on NeSI and use the configuration file ~/.config/rstudio_on_nesi/singularity_image_path to indicate the path of your container to the RStudio-on-NeSI plugin: $ echo PATH_TO_CONTAINER > ~/.config/rstudio_on_nesi/singularity_image_path To modify the content of the container, you need to adapt the Singularity definition file, found in the conf folder of the repository, and then rebuild the container.
#R studio help code#
The related code is hosted on GitHub, in the rstudio_on_nesi repository. RStudio runs in a Singularity container prepared by the NeSI team to run on.
#R studio help install#
The alternative is to install packages in a terminal session Advanced usage " / nesi/nobackup / / rstudio_tmp/Rtmpjp2rm8" Within RStudio run the command `tempdir()` which should return the following (below), where `Rtmpjp2rm8` is a randomly generated folder name, and is emptied with each new session. $ echo "TMP=/nesi/nobackup//rstudio_tmp " >. $ mkdir - p / nesi/nobackup / / rstudio_tmp These will setup a larger directory that will allow for packages to be installed to your personal library. To avoid read/write issues with a small temorary directory filling up, in a terminal run the following two lines of code. Check that the correct version of R has loaded and that the correct Library Paths are available. Once your configuration file is ready, make sure to restart your Jupyter session and re-launch RStudio for these changes to be taken into account. In the following example, we use the module that is built for R/4.1.0 $ echo "module load R/4.1.0-gimkl-2020a" > ~/.config/rstudio_on_nesi/prelude.bash The module needs to be entered in the configuration file ~/.config/rstudio_on_nesi/prelude.bash. This can be useful if you want to change the version of the R interpreter or use NeSI's R-Geo or R-bundle-Bioconductor modules. You can configure a set of environment modules to preload before starting RStudio. It will be auto-filled using a pre-generated password, unless you disabled javascript in your web browser. Once RStudio is launched, you should briefly see a login screen. In the JupyterLab interface, RStudio can be started using the corresponding entry in the launcher.Ĭlicking on this entry will open a separate tab in your web browser, where RStudio will be accessible. RStudio can be accessed as a web application via Jupyter on NeSI. Your feedback is welcome, please don't hesitate to contact us at to make suggestions. This functionality is experimental and may introduce breaking changes in the future.
0 kommentar(er)
