Radiasoft Public JupyterHub

FAQ

Spawn Failed: There are no more servers available.

You may get this error if our cluster has too many active users:

Spawn failed

The latest attempt to start your server has failed. There are no more servers available. Please wait a few minutes before trying again. Would you like to retry starting it?

You can only retry by clicking on the home page link and then clicking on Start My Server. Refreshing your browser page will not allow you to retry.

You can also submit an Issue so that we know you are waiting for a free server.

Customizations

pre_jupyter_bashrc

The file in ~/jupyter/pre_jupyter_bashrc, and it will be run once at container start. This will allow you to customize your server environment such as running pip install. ~/.bashrc will run before this runs so pyenv is setup.

bashrc

You can create a file in ~/jupyter/bashrc, and it will be read each time a terminal window starts. This will allow you to customize your shell environment. It will also be read when your server starts so if you would like a change to show up in notebooks, you will need to restart your server.

programs

The directory ~/jupyter/bin is automatically inserted in your $PATH so you can create programs/scripts and chmod +x <file> in order to make them executable. You may need to type hash -r after you type a new commands.

radia-run.sh

This repo provides customizations for end-user containers started on jupyter.radiasoft.org. This repo's radia-run.sh is executed first then the user's jupyter.radiasoft.org radia-run.sh, allowing the user to customize the output of this customization.

The file radia-run.sh will be executed at every container start as follows:

. ~/.bashrc
git clone https://github.com/radiasoft/jupyter.radiasoft.org
cd jupyter.radiasoft.org
. ./radia-run.sh
cd ..
rm -rf jupyter.radiasoft.org

Since the script is run in a subshell, it has access to all the functions in the RadiaSoft downloader. For now, do not assume much about this context, but eventually an API will be refined.

Right now, it executes in the master branch. We'll eventually install on the channel, but you can expect that your personal juypter.radiasoft.org repo will always install off of master.

Customization tips

Avoid hardwiring paths. Use $HOME or ~ in scripts. Don't assume anything about $USER, because it may change.

You can assume ~/jupyter is where the notebooks are stored and where the terminal boots from. This directory persists across container restarts so you should be careful not to stomp existing files.

Importing the Warp code in an IPython Notebook

Simply trying 'import warp' or 'from warp import *' will generate errors. Instead, use the following:

import sys
del sys.argv[1:]
from warp import *

European XFEL: the WaveProperGator (WPG) notebooks for running SRW

1. Point your browser to the RadiaSoft JupyterHub server

2. Login with your GitHub credentials.

3. Click the "Terminal" button in the "Launcher" tab.

4. In the new terminal window, type the following command:

jupyter$ git clone https://github.com/samoylv/WPG.git ./WPG

5. Browse to a notebook, by doing (for example) the following:

Double click "WPG"

Double click "samples"

Double click "Tutorials"

Double click "Tutorial_case_1.ipynb"

6. In the "Select Kernel" popup window, you can keep the default choice of "Python 2" by clicking the "SELECT" button.