Starting a Jupyter Notebook
The instructions below will walk you through the process of launching a Jupyter Server workflow using the Fuzzball workflow catalog. The workflow catalog will help you generate, validate, and submit a Fuzzfile which runs a Jupyter Notebook. By default, this workflow uses a Jupyter container from quay.io.
You can start this workflow using either the web UI or the CLI.
Open the Catalog tab of the Workflows page to see the workflow templates you use to create Fuzzfiles
and launch workflows. Using the search bar, find the Jupyter Server tile – the CIQ
tag marks it as a template provided by CIQ – and click it to open its detail page.

The template’s detail page shows a Configure panel on the right listing the template parameters, which you can modify. You can also give this run an optional descriptive name in the Run Name field. To accept the default values, click RUN. Fuzzball generates a Fuzzfile from the template and your values, validates it, and submits it.

Submitting the workflow takes you straight to its status page. The screenshot below shows the status page for a Jupyter Notebook workflow submission.

To retrieve logs produced by this workflow, select the jupyter job – either by clicking its node
on the canvas or by choosing it from the stage drop down below the canvas – and then click the
Logs tab in the panel on the right.

To run this workflow through the CLI you will need access to the Fuzzball CLI. You can install it using the Fuzzball CLI installation instructions.
First, you can create a values file values.yaml to populate the Jupyter Server workflow template
using the command below. You can modify the parameters to your desired values or use the defaults
below.
$ cat >values.yaml<<EOF
values:
- name: "Type"
string_value: "r"
- name: "Cores"
uint_value: 1
- name: "Memory"
string_value: "4GiB"
- name: "GPUs"
uint_value: 0
- name: "Exclusive"
bool_value: false
- name: "Volume"
string_value: "volume://user/persistent"
- name: "ServiceScope"
string_value: "user"
EOFNext, you will need to find the workflow template. The command below lists workflow templates
provided by CIQ. The ID shown is an example value: template IDs are specific to a cluster and to
the catalog repository the template came from, so yours will differ. start also accepts the
template name, which is what the example below uses – provided no other catalog repository
publishes a template of that name.
$ fuzzball workflow catalog list --provider CIQ --name '*Jupyter*'
ID | NAME
f94e0a09-8d20-52da-b69d-825ab4824ded | Jupyter ServerYou can start a workflow with the workflow template and the values file (values.yaml) we created in the first step using the command below.
$ fuzzball workflow catalog start --values values.yaml 'Jupyter Server'
0345befe-32b0-42c1-9b82-03d3d1c38434You can monitor the workflow’s status by running the following command:
$ fuzzball workflow describe 0345befe-32b0-42c1-9b82-03d3d1c38434
Name: jupyter test
Email: wresch@ciq.com
UserId: 87145648-b830-4291-ab7e-40880d61334e
Status: STAGE_STATUS_STARTED
Cluster: fuzzball-aws-stable
Created: 2026-04-06 12:55:25PM
Started: 2026-04-06 12:55:26PM
Finished: N/A
Error:
KIND | STATUS | NAME | DURATION
Workflow | Started | 0345befe-32b0-42c1-9b82-03d3d1c38434 | 4m
Volume | Finished | data | 2s
Image | Finished | docker://quay.io/jupyter/r-notebook:latest | 1m59s
Service | Started | jupyter | 30sYou can view the output logged by the workflow by running the following command:
$ fuzzball workflow log 0345befe-32b0-42c1-9b82-03d3d1c38434 jupyter
...
[I 2026-04-06 16:59:31.968 LabApp] JupyterLab extension loaded from /opt/conda/lib/python3.13/site-packages/jupyterlab
[I 2026-04-06 16:59:31.968 LabApp] JupyterLab application directory is /opt/conda/share/jupyter/lab
[I 2026-04-06 16:59:31.969 LabApp] Extension Manager is 'pypi'.
[I 2026-04-06 16:59:32.071 ServerApp] jupyterlab | extension was successfully loaded.
[I 2026-04-06 16:59:32.092 ServerApp] nbclassic | extension was successfully loaded.
[I 2026-04-06 16:59:32.096 ServerApp] notebook | extension was successfully loaded.
[I 2026-04-06 16:59:32.097 ServerApp] Serving notebooks from local directory: /data
[I 2026-04-06 16:59:32.097 ServerApp] Jupyter Server 2.17.0 is running at:
[I 2026-04-06 16:59:32.097 ServerApp] http://jupyter:27760/endpoints/endpoint-0fc78dd85a0b0733d20bbe229b2c823a6ea74bc4/lab
[I 2026-04-06 16:59:32.097 ServerApp] http://127.0.0.1:27760/endpoints/endpoint-0fc78dd85a0b0733d20bbe229b2c823a6ea74bc4/lab
[I 2026-04-06 16:59:32.097 ServerApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[I 2026-04-06 16:59:33.338 ServerApp] Skipped non-installed server(s): basedpyright, bash-language-server, dockerfile-language-server-nodejs, javascript-typescript-langserver, jedi-language-server, julia-language-server, pyrefly, pyright, python-language-server, python-lsp-server, r-languageserver, sql-language-server, texlab, typescript-language-server, unified-language-server, vscode-css-languageserver-bin, vscode-html-languageserver-bin, vscode-json-languageserver-bin, yaml-language-serverTo open this endpoint from the CLI, run
fuzzball workflow connect --name jupyter 0345befe-32b0-42c1-9b82-03d3d1c38434, which launches
it in your browser. The
web UI route is described in Accessing the Jupyter Server UI.