1. Getting set up¶
If you are using the Quantum Mobile virtual machine locally inside VirtualBox, you can skip this section.
1.1. Connect to your virtual machine¶
You should each have received from the instructors:
an IP address
a private SSH key
a public SSH key
The steps below explain how to use these in order to connect to your personal virtual machine (VM) on Amazon Elastic Cloud 2 using the Secure Shell protocol. This VM is closely based on the Quantum Mobile VM and already includes a pre-configured AiiDA installation.
1.1.1. Linux and MacOS¶
It’s recommended for you to place the ssh key you received in a folder dedicated to your ssh configuration, to do so:
If not already present, create a
.sshdirectory in your home (
mkdir ~/.ssh), and set its permissions:
chmod 700 ~/.ssh
Copy the two keys
Set the correct permissions on the private key:
chmod 600 ~/.ssh/aiida_tutorial_NUM. You can check check with
ls -lthat the permissions of this file are now
After that ssh key is in place, you can add the following block your
Host aiidatutorial Hostname IP_ADDRESS User max IdentityFile ~/.ssh/aiida_tutorial_NUM ForwardX11 yes ForwardX11Trusted yes LocalForward 8888 localhost:8888 LocalForward 5000 localhost:5000 ServerAliveInterval 120
replacing the IP address (
IP_ADDRESS) and the
the one you received.
Afterwards you can connect to the server using this simple command:
$ ssh aiidatutorial
Here’s a copy-paste ready command for you to use directly with zero configuration:
ssh \ -i ~/.ssh/aiida-tutorial-max.pem \ -L 8888:localhost:8888 \ -L 5000:localhost:5000 \ -o ServerAliveInterval=120 \ -X -C \ max@IP_ADDRESS
On MacOS you need to install XQuartz in order to use X-forwarding.
If you’re running Windows 10, you may want to consider installing the Windows Subsystem for Linux (and then follow the instructions above). Alternatively:
Install the PuTTY SSH client.
aiida_tutorial_NNprivate key (button “Load”). You may need to choose to show “All files (.)”, and select the file without any extension (Type: File).
In the same window, click on “Save private Key”, and save the key with the name
aiida_tutorial_NN.ppk(don’t specify a password).
It will add a new icon near the clock, in the bottom right of your screen.
Right click on this Pageant icon, and click on “View Keys”.
Click on “Add key” and select the
aiida_tutorial_NN.ppkyou saved a few steps above.
Put the given IP address as hostname, type
aiidatutorialin “Saved Sessions” and click “Save”.
Go to Connection > Data and put
maxas autologin username.
Go to Connection > SSH > Tunnels, type
8888in the “Source Port” box, type
localhost:8888in “Destination” and click “Add”.
Repeat the previous step for port
Go back to the “Session” screen, select “aiidatutorial” and click “Save”
Finally, click “Open” (and click “Yes” on the putty security alert to add the VM to your known hosts). You should be redirected to a bash terminal on the virtual machine.
Next time you open PuTTY, select
aiidatutorial and click “Load” before clicking “Open”.
In order to enable X-forwarding:
1.2. Start jupyter¶
Once connected to your virtual machine, type in the remote terminal
This will enable the virtual environment in which AiiDA is installed, allowing you to use AiiDA. Now type in the same shell:
jupyter notebook --no-browser
This will run a server with a web application called
is used to create interactive python notebooks.
In order to connect to the jupyter notebook server:
copy the URL that has been printed to the terminal (similar to
open a web browser on your laptop and paste the URL
You will see a list of folders on your personal VM.
While keeping the first
ssh connection running, open another
ssh connection in a second terminal and type
workon aiida here too.
This terminal is the one we will actually use in this tutorial.
Our SSH configuration assumes that
jupyter will serve the notebooks on port 8888.
If you want to serve notebooks on different ports, you’ll also need to adjust
the SSH configuration.
1.3. Downloading files¶
Throughout this tutorial, you will encounter links to download python scripts, jupyter notebooks and more.
These files should be downloaded to the environment/working directory you use to run the tutorial.
In particular, when running the tutorial on a linux virtual machine, copy the link address and download the files to the machine using the
wget utility on the terminal:
where you replace
<LINK> with the actual HTTPS link that you copied from the tutorial text in your browser.
This will download that file in your current directory.
If you get errors
ImportError: No module named aiidaor
No command ’verdi’ found, double check that you have loaded the virtual environment with
workon aiidabefore launching
If, while connecting to your VM, you get a warning similar to:
bind [127.0.0.1]:8888: Address already in use channel_setup_fwd_listener_tcpip: cannot listen to port: 8888
your local port 8888 is already occupied - likely because you are running a
jupyter notebookserver locally. We suggest you stop any locally running jupyter notebook servers before connecting to the VM. If necessary, you can start them again after you have connected (
jupyter notebookwill then realize that port 8888 is already taken and simply serve the notebook on a different port).
If your browser cannot connect to the jupyter notebook server, check that you have configured SSH tunneling/forwarding as described above. Keep in mind that you need to start the jupyter server from the terminal connected to the VM, while opening the web browser locally on your laptop.
See the jupyter notebook documentation for compatibility of jupyter with various web browsers.