Specialised Environments

Page tree

After executing the "jupyter.ini.sh" command as explained in the preceding sections, a file named "client_cmd" will be generated in the $PBS_O_WORKDIR directory. This file contains instructions for redirecting network traffic from the specified port number on the Gadi worker node to an external client machine (such as your desktop) through the Gadi login node at gadi.nci.org.au. You should execute these commands on your client machine.

Client_cmd file

The contents of client_cmd file may vary on the way to access the Jupyterlab session and whether a pre-defined Dask/Ray cluster is set up when you run jupyter.ini.sh script with different flags.

Job with token access to Gadi JupyterLab ( i.e. after running "jupyter.ini.sh -J -T" ).

ssh -N -L 8676:gadi-cpu-clx-2310.gadi.nci.org.au:8676 jpf777@gadi.nci.org.au     # ssh port forwarding command towards the remote JupyterLab session.
http://127.0.0.1:8676/lab?token=44c79c8a8ac2d5d5238f69508a3d4fa9350c325a615f7abf # url link with token access to the remote JupyterLab session.

Job with token access to Gadi JupyterLab session together with a pre-defined Dask/Ray cluster ( i.e. after running "jupyter.ini.sh -J -T -D"  or

"jupyter.ini.sh -J -T -R").

ssh -N -L 8828:gadi-cpu-clx-2310.gadi.nci.org.au:8828 jpf777@gadi.nci.org.au     # ssh port forwarding command towards the Dask/Ray dashboard.
ssh -N -L 8676:gadi-cpu-clx-2310.gadi.nci.org.au:8676 jpf777@gadi.nci.org.au     # ssh port forwarding command towards the remote JupyterLab session.
http://127.0.0.1:8676/lab?token=44c79c8a8ac2d5d5238f69508a3d4fa9350c325a615f7abf # url link with token access to remote JupyterLab session.

Job without token access to Gadi JupyterLab session ( i.e. after running "jupyter.ini.sh -J" ).

ssh -N -L 8676:gadi-cpu-clx-2310.gadi.nci.org.au:8676 jpf777@gadi.nci.org.au  # ssh port forwarding command towards the remote JupyterLab session.
http://127.0.0.1:8676/lab                                                     # Url link to access the remote Jupyter Lab server. You may need the password to login.

Job without token access to Gadi JupyterLab session together with a pre-defined Dask/Ray cluster. (i.e. after running "jupyter.ini.sh -J -D"  or "jupyter.ini.sh -J -R")

ssh -N -L 8828:gadi-cpu-clx-2310.gadi.nci.org.au:8828 jpf777@gadi.nci.org.au   # ssh port forwarding command towards the Dask/Ray dashboard.
ssh -N -L 8676:gadi-cpu-clx-2310.gadi.nci.org.au:8676 jpf777@gadi.nci.org.au   # ssh port forwarding command towards the remote JupyterLab session.
http://127.0.0.1:8676/lab                                                      # Url link to access the remote Jupyter Lab server. You may need the password to login.

Whenever you set up a Dask cluster, a file called scheduler.json will be produced. It contains the information for the Dask scheduler. The file should be kept existing during the job running time.

 Connect Gadi JupyterLab session from your client machine

Now you can connect to the Gadi Jupyterlab session from your own client machine or a VDI session

  1. Run each SSH port forwarding command except for the bottom line of the file "client_cmd" in a seperate command line terminal of your client computer and keep them alive.
  2. Copy the last line of client_cmd file, which is the JupyterLab url, into your client machine web browser to access Gadi JupyterLab session.

Note: If you haven’t set up SSH keys on your client machine to enable the non-password SSH login to Gadi, instructions on how to do this can be found here. Alternatively, you should enter your Gadi password when prompted.

  • No labels