Skip to content

jupyterhub/jupyterhub-idle-culler

Repository files navigation

JupyterHub Idle Culler Service

GitHub Workflow Status - Test Latest PyPI version GitHub Discourse Gitter

jupyterhub-idle-culler provides a JupyterHub service to identify and stop idle or long-running Jupyter servers via JupyterHub. It works solely by interacting with JupyterHub's REST API, and is often configured to run as a JupyterHub managed service started up by JupyterHub itself.

Setup

Setup involves three parts:

  1. Install the Python package.
  2. Configure JupyterHub permissions to work against JupyterHub's REST API.
  3. Configure how its started up, either as a JupyterHub managed service or as a standalone script.

Installation

pip install jupyterhub-idle-culler

Permissions

Prior to JupyterHub 2.0, the jupyterhub-idle-culler required full administrative privileges, in order to have sufficient permissions to stop servers on behalf of users.

JupyterHub 2.0 introduces scopes to allow for more fine-grained permission control. This means that the configured culler service does not need full administrative privileges anymore. It can be assigned only the permissions it needs.

jupyterhub-idle-culler requires the following scopes to function:

  • list:users - to access to the user list API, our source of information about who to cull
  • read:users:activity - to read the users' last_activity field
  • read:servers - to read the users' servers field
  • delete:servers - to stop users' servers, and delete named servers if --remove-named-servers is passed
  • admin:users (optional) - to delete users if --cull-users is passed

To assign the service the appropriate permissions, declare a role in your jupyterhub_config.py:

c.JupyterHub.load_roles = [
    {
        "name": "jupyterhub-idle-culler-role",
        "scopes": [
            "list:users",
            "read:users:activity",
            "read:servers",
            "delete:servers",
            # "admin:users", # if using --cull-users
        ],
        # assignment of role's permissions to:
        "services": ["jupyterhub-idle-culler-service"],
    }
]

As a hub managed service

In jupyterhub_config.py, add the following dictionary for the idle-culler service to the c.JupyterHub.services list:

c.JupyterHub.services = [
    {
        "name": "jupyterhub-idle-culler-service",
        "command": [
            sys.executable,
            "-m", "jupyterhub_idle_culler",
            "--timeout=3600",
        ],
        # "admin": True,
    }
]

where:

  • "command" indicates that the Service will be managed by the Hub, and
  • "admin": True grants admin permissions to this Service and is only meant for use with jupyterhub < 2.0; see [above][permissions].

As a standalone script

jupyterhub-idle-culler can also be run as a standalone script. It can access the hub's api with a service token.

Register the service token with JupyterHub in jupyterhub_config.py:

c.JupyterHub.services = [
    {
        "name": "jupyterhub-idle-culler-service",
        "api_token": "...",
        # "admin": True,
    }
]

where:

  • "api_token" contains a secret token, e.g. generated by openssl rand -hex 32, and
  • "admin": True grants admin permissions to this Service and is only meant for use with jupyterhub < 2.0; see [above][permissions].

and store the same token in a JUPYTERHUB_API_TOKEN environment variable. Then start jupyterhub-idle-culler manually.

export JUPYTERHUB_API_TOKEN=api_token_above...
python3 -m jupyterhub_idle_culler [--timeout=900] [--url=http://localhost:8081/hub/api]

Command line flags

  --api-page-size                  Number of users to request per page, when
                                   using JupyterHub 2.0's paginated user list
                                   API. Default: user the server-side default
                                   configured page size. (default 0)
  --concurrency                    Limit the number of concurrent requests made
                                   to the Hub.  Deleting a lot of users at the
                                   same time can slow down the Hub, so limit
                                   the number of API requests we have
                                   outstanding at any given time. (default 10)
  --cull-admin-users               Whether admin users should be culled (only
                                   if --cull-users=true). (default True)
  --cull-every                     The interval (in seconds) for checking for
                                   idle servers to cull. (default 0)
  --cull-default-servers           Whether default servers should be culled (only
                                   if --cull-default-servers=true). (default True)
  --cull-named-servers             Whether named servers should be culled (only
                                   if --cull-named-servers=true). (default True)
  --cull-users                     Cull users in addition to servers.  This is
                                   for use in temporary-user cases such as
                                   tmpnb. (default False)
  --internal-certs-location        The location of generated internal-ssl
                                   certificates (only needed with --ssl-
                                   enabled=true). (default internal-ssl)
  --max-age                        The maximum age (in seconds) of servers that
                                   should be culled even if they are active.
                                   (default 0)
  --remove-named-servers           Remove named servers in addition to stopping
                                   them.  This is useful for a BinderHub that
                                   uses authentication and named servers.
                                   (default False)
  --ssl-enabled                    Whether the Jupyter API endpoint has TLS
                                   enabled. (default False)
  --timeout                        The idle timeout (in seconds). (default 600)
  --url                            The JupyterHub API URL.

Caveats

  1. JupyterHub's last_activity data about user servers is not updated with high frequency, so cull timeout should be greater than the sum of:

    • single-user websocket ping interval (default: 30 seconds)
    • JupyterHub.last_activity_interval (default: 5 minutes)
  2. If you want to use --cull-users with a different culling interval for the user servers and users, you must start two idle culler services. This is because both are configured via --timeout and --max-age. To do so, configure this service to start twice with different configuration, where one has the --cull-users option.

  3. By default jupyterhub-idle-cullers HTTP requests to JupyterHub's REST API timeouts after 60 seconds. This can be changed by setting the JUPYTERHUB_REQUEST_TIMEOUT environment variable.

How it works

JupyterHub's REST API is used to acquire information about activity, and if the idle culler service based on configuration thinks a server should be stopped or deleted it also does so via JupyterHub's REST API.

In depth

jupyterhub-idle-culler relies on permission to work against JupyterHub's REST API is provided via the JUPYTERHUB_API_TOKEN, that is set automatically for managed services started by JupyterHub.

jupyterhub-idle-culler lists available users and their server's reported last_activity via JupyterHub's /users REST API and makes decisions based on that. User's default servers can be stopped via /users/{name}/server, named servers can be stopped and optionally removed via /users/{name}/servers/{server_name}, and users can optionally be deleted via /users/{name}.

JupyterHub's reported last_activity for user servers is updated by JupyterHub at a regular interval in the update_last_activity function that relies on two sources of information.

  1. The proxy's routes data

    The configurable proxy class for JupyterHub is an interface for JupyterHub to request routing of network traffic to user servers. Through this interface, JupyterHub be informed on network activity if the proxy class provides it, specifically via the get_all_routes function.

    The configurable-http-proxy used in https://z2jh.jupyter.org provides information about network routes activity, but traefik-proxy used in https://tljh.jupyter.org currently does not.

  2. The user server's activity reports

    The update_last_activity function also reads JupyterHub's database that keeps state about servers last_activity. These database records are updated whenever a server notifies JupyterHub about activity, as they are required to do.

    Servers has before JupyterHub 4 notified JupyterHub about activity by being started by the jupyterhub-singleuser script made available by installing jupyterhub (or jupyterhub-singleuser on conda-forge). With JupyterHub 4+ and jupyter_server 2+ a jupyter_server server extension can be used instead.

    The jupyterhub-singleuser script launches a modified server application that keeps JupyterHub updated with the server activity via the notify_activity function.

    The notify_activity function in turn make use of the server applications last_activity function (see implementation in NotebookApp and ServerApp respectively) that that combines information from API activity, kernel activity, kernel shutdown, and terminal activity. This activity also covers activity of applications like RStudio running via jupyter-server-proxy.

Here is a summary of what's described so far:

  1. jupyterhub-idle-culler collects information and acts entirely through JupyterHub's REST API.
  2. jupyterhub-idle-culler makes decisions based on information provided by JupyterHub, that collects activity reports from the user servers and polls the proxy class for information about user servers' network activity.

Now, as the server's kernel activity influence the activity that servers will notify JupyterHub about, the kernel activity in turn influences jupyterhub-idle-culler. Due to this, it can be relevant to also learn a little about a mechanism to cull idle kernels as well even though jupyterhub-idle-culler isn't involved in that.

The default kernel manager, the MappingKernelManager, can be configured to cull idle kernels. Its configuration is documented in ServerApp's and NotebookApp's respective documentation, and here are some relevant kernel culling configuration options:

  • MappingKernelManager.cull_busy

  • MappingKernelManager.cull_idle_timeout

  • MappingKernelManager.cull_interval

  • MappingKernelManager.cull_connected

    Note that cull_connected can be tricky to understand for JupyterLab as a browser having a web-socket connection to a kernel or not isn't as obvious as it was in the classical Jupyter notebook UI. See this issue for more details.

    Also note that configuration of MappingKernelManager should be made on the user server itself, for example via a jupyter_server_config.py file in /etc/jupyter or /usr/local/etc/jupyter rather than where JupyterHub is running.

Finally, note that a Jupyter server can shut itself down without intervention by jupyterhub-idle-culler if ServerApp.shutdown_no_activity_timeout is configured.

Caveats

Pagination

JupyterHub 2.0 introduces pagination to the /users API endpoint. This pagination does not guarantee a consistent snapshot for consecutive requests spread over time, so it is possible for a highly active hub to occasionally miss culling users crossing page boundaries between requests. This is expected to be an infrequent occurrence and only result in delaying a server being culled by one cull interval in realistic scenarios, so of minor consequence in JupyterHub.

The issue can be mitigated by requesting a larger page size, via e.g. --api-page-size=200, but feel free to open an issue if this is causing a problem for you.