Connect to the Tutorial Server
Live tutorial
Your credential card contains your username, password, and server.
Angle-bracket values need editing
Amber text inside angle brackets, such as <USERNAME> or <SERVER>, marks a value you must replace. The rest of the command remains unchanged.
macOS and Linux
Open Terminal and connect with:
ssh <USERNAME>@<SERVER>
Replace <USERNAME> and <SERVER> with the values printed on your card.
Windows — unverified
These Windows instructions have not been verified yet. We will remove this notice after testing password login and SSH tunneling on a Windows computer.
Open PowerShell or Windows Terminal and check whether the OpenSSH client is available:
ssh -V
If that prints an OpenSSH version, connect with:
ssh <USERNAME>@<SERVER>
Replace <USERNAME> and <SERVER> with the values printed on your card.
The first connection may ask whether you want to trust the server. Enter:
yes
Then enter your password.
Your password will not appear on screen while you type. This is normal.
1. Check your account
After logging in, run:
whoami
hostname
pwd
whoami should show the username printed on your card, and your home directory should be:
/home/<USERNAME>
Now confirm that your tutorial workspace is writable:
printf 'access works\n' > ~/tutorial/access-test.txt
cat ~/tutorial/access-test.txt
rm ~/tutorial/access-test.txt
You should see:
access works
Your home directory and tutorial workspace belong to your account. Always use the username and server assigned to you.
2. Find the tutorial repository and examples
All tutorial instructions, scripts, and examples are already available in your workspace. Do not clone another copy of the repository on the live server.
cd ~/tutorial
ls
You should see the main repository files and directories:
README.md docs examples mkdocs.yml scripts
3. Activate the tutorial tools
The tutorial software is installed in a shared, read-only environment.
Activate it with:
source /opt/tutorial/activate.sh
Confirm that the environment is active:
python --version
echo "$CONDA_PREFIX"
CONDA_PREFIX should point somewhere under:
/opt/tutorial/
4. Check TaskVine and Floability
Run:
vine_worker --version
python -c "import ndcctools.taskvine; print('TaskVine: OK')"
floability --help >/dev/null && echo "Floability: OK"
sciunit --version >/dev/null && echo "Sciunit: OK"
You should finish with:
TaskVine: OK
Floability: OK
Sciunit: OK
You are ready
Your setup now has two separate areas:
/opt/tutorial/ shared tutorial software
~/tutorial/ your exercises and files
Later, Floability will create and cache backpack-specific software environments inside your own user space.
These environments are separate from the shared tutorial tools. Repeated runs of the same backpack may reuse your cached environment, but environments are not shared between participants.
SSH tunneling for notebook exercises
Later notebook exercises will use SSH tunneling so Jupyter does not need a public network port. The instructions will provide the local and remote port numbers and use this pattern:
ssh -L <LOCAL_PORT>:localhost:<REMOTE_PORT> <USERNAME>@<SERVER>
The Windows version is expected to use the same options in PowerShell, but it remains unverified until tested on a Windows computer.
Continue to TaskVine Quickstart.