Table of Contents

1. First time setting
2. Accessing data on /project or /scratch/pgfs
3. Uploading Files
4. Downloading Files
5. Troubleshooting: rebuilding the environment from scratch

First time setting

• Go to https://jupyter.rc.princeton.edu
• You will need authenticate with your Princeton netid and password.
• In Spawner options select Anaconda3 2019.07 in home directory - Climate Modeling. Note that the building the image the first time will take about 3 min. Afterthe first build it will be faster.

Accessing data on /project or /scratch/pgfs

• The server jupyter.rc is independent from the cluster tiger but has access to both /tigress and /scratch/gpfs on tiger. For convenience, create some symbolic links in your home directory.

  You can do this by opening a terminal by clicking on New then Terminal.

  $ ln -s /projects/GEOCLIM projects
  $ ln -s /tiger/scratch/gpfs/GEOCLIM gpfs
  

  

• After this, the directories projects and gpfs will be in your home directory. You may need to refresh your browser to see them

Uploading Files

To upload a file click Upload

Downloading Files

To download a file click the file and Download. The file will be downloaded where your browser downloads files.

Troubleshooting: rebuilding the environment from scratch

If your Anaconda environment as been corrupted or you think something is not working properly. You can rebuilt your environment from scratch by following this procedure:

1. from a running notebook on jupyterhub open a Terminal,

2. remove your current anaconda3 directory with:

   rm -Rf $HOME/anaconda3
   

3. log out,
4. click Stop My Server,
5. click My Server,
6. rebuild the environment following the steps described in First time setting.