How to generate a Jupyter Notebook configuration file

Jupyter Notebook stores persistent server settings in a user configuration directory, so generating the base file is the first step before changing ports, authentication, browser launch behavior, working directory, or HTTPS settings. Notebook 7 runs on Jupyter Server behind the Notebook interface, so server settings belong in jupyter_server_config.py.

The jupyter server command creates the config file in the active Jupyter config directory, usually ~/.jupyter for a user account. The generated file is a Python config file with commented c.ServerApp settings that can be uncommented later when a specific setting needs to persist.

Generating the file does not change how Notebook starts until a setting is uncommented or added. If jupyter_server_config.py already exists, Jupyter prompts before overwriting it, so keep a backup before replacing a file that already contains custom settings.

1. Show the active Jupyter configuration directory.

   $ jupyter --config-dir
   /home/user/.jupyter

2. Generate the Jupyter Server configuration file.

   $ jupyter server --generate-config
   Writing default config to: '/home/user/.jupyter/jupyter_server_config.py'

   If the file already exists, choose overwrite only after saving any custom settings that should remain.

3. Confirm the generated file contains port settings for later edits.

   $ grep -n "c.ServerApp.port" ~/.jupyter/jupyter_server_config.py
   491:# c.ServerApp.port = 0
   496:# c.ServerApp.port_retries = 50

   ServerApp is the Jupyter Server configuration class used by Notebook 7. Leave generated lines commented until a setting needs to be enabled.

   Related: How to change the Jupyter Notebook port

4. Check another generated server setting from the same file.

   $ grep -n "c.ServerApp.root_dir" ~/.jupyter/jupyter_server_config.py
   523:# c.ServerApp.root_dir = ''

   Related: How to change the Jupyter Notebook startup directory