[BUG]: Starting ZenML server hangs

[guzman-raphael]

Contact Details [Optional]

No response

System Information

vscode@machine:/workspaces/zenml-demo$ zenml info -a -s
Initializing the ZenML global configuration version to 0.65.0
Creating database tables
Creating default workspace 'default' ...
Creating default stack in workspace default...
Setting the global active workspace to 'default'.
Setting the global active stack to default.
ZENML_LOCAL_VERSION: 0.65.0
ZENML_SERVER_VERSION: 0.65.0
ZENML_SERVER_DATABASE: sqlite
ZENML_SERVER_DEPLOYMENT_TYPE: other
ZENML_CONFIG_DIR: /home/vscode/.config/zenml
ZENML_LOCAL_STORE_DIR: /home/vscode/.config/zenml/local_stores
ZENML_SERVER_URL: sqlite:////home/vscode/.config/zenml/local_stores/default_zen_store/zenml.db
ZENML_ACTIVE_REPOSITORY_ROOT: None
PYTHON_VERSION: 3.9.19
ENVIRONMENT: vscode_remote_container
SYSTEM_INFO: {'os': 'linux', 'linux_distro': 'debian', 'linux_distro_like': '', 'linux_distro_version': '12'}
ACTIVE_WORKSPACE: default
ACTIVE_STACK: default
ACTIVE_USER: default
TELEMETRY_STATUS: enabled
ANALYTICS_CLIENT_ID: c9298a71-eeee-4eb2-8df3-d1004e47084c
ANALYTICS_USER_ID: 7cee2133-007d-4740-9cec-805002f07f0e
ANALYTICS_SERVER_ID: c9298a71-eeee-4eb2-8df3-d1004e47084c
INTEGRATIONS: ['airflow', 'bitbucket', 'kaniko', 'pigeon']
PACKAGES: {'zenml': '0.65.0', 'widgetsnbextension': '4.0.13', 'validators': '0.18.2', 'urllib3': '2.2.2', 
'tzdata': '2024.1', 'sqlmodel': '0.0.18', 'sqlalchemy': '2.0.32', 'sqlalchemy-utils': '0.41.2', 'smmap': 
'5.0.1', 'secure': '0.3.0', 'rich': '13.8.0', 'requests': '2.32.3', 'pyyaml': '6.0.2', 'pytz': '2024.1', 
'python-dotenv': '1.0.1', 'pyparsing': '2.4.7', 'pymysql': '1.1.1', 'pydantic': '2.7.4', 'pydantic-settings': 
'2.2.1', 'pydantic-core': '2.18.4', 'passlib': '1.7.4', 'pandas': '2.2.2', 'numpy': '1.26.4', 'mdurl': '0.1.2', 
'markupsafe': '2.1.5', 'markdown-it-py': '3.0.0', 'mako': '1.3.5', 'jupyterlab-widgets': '3.0.13', 'ipywidgets':
'8.1.5', 'importlib-metadata': '7.0.0', 'idna': '3.8', 'httplib2': '0.19.1', 'greenlet': '3.0.3', 'gitpython': 
'3.1.43', 'gitdb': '4.0.11', 'docker': '7.1.0', 'distro': '1.9.0', 'cloudpickle': '2.2.1', 'click': '8.1.3', 
'click-params': '0.3.0', 'charset-normalizer': '3.3.2', 'certifi': '2024.8.30', 'bcrypt': '4.0.1', 
'annotated-types': '0.7.0', 'alembic': '1.8.1', 'zipp': '3.20.1', 'wheel': '0.44.0', 'wcwidth': '0.2.13', 
'typing-extensions': '4.12.2', 'traitlets': '5.14.3', 'tornado': '6.4.1', 'tomli': '2.0.1', 'stack-data': 
'0.6.3', 'six': '1.16.0', 'setuptools': '58.1.0', 'pyzmq': '26.2.0', 'python-dateutil': '2.9.0.post0', 
'pygments': '2.18.0', 'pyflakes': '3.2.0', 'pycodestyle': '2.12.1', 'pure-eval': '0.2.3', 'ptyprocess': '0.7.0',
'psutil': '6.0.0', 'prompt-toolkit': '3.0.47', 'platformdirs': '4.2.2', 'pip': '24.2', 'pexpect': '4.9.0', 
'pathspec': '0.12.1', 'parso': '0.8.4', 'packaging': '24.1', 'nest-asyncio': '1.6.0', 'mypy-extensions': 
'1.0.0', 'mccabe': '0.7.0', 'matplotlib-inline': '0.1.7', 'jupyter-core': '5.7.2', 'jupyter-client': '8.6.2', 
'jedi': '0.19.1', 'ipython': '8.18.1', 'ipykernel': '6.29.5', 'flake8': '7.1.1', 'faker': '28.1.0', 'executing':
'2.0.1', 'exceptiongroup': '1.2.2', 'decorator': '5.1.1', 'debugpy': '1.8.5', 'comm': '0.2.2', 'black': 
'24.8.0', 'asttokens': '2.4.1'}

CURRENT STACK

Name: default
ID: ec008200-7676-4720-a962-7bde08752820
Workspace: default / d81c2a07-1b7c-4aa5-ade5-72a8d56c45ae

ORCHESTRATOR: default

Name: default
ID: 927bc017-56f3-4199-9fb7-e1b3696d270f
Type: orchestrator
Flavor: local
Configuration: {}
Workspace: default / d81c2a07-1b7c-4aa5-ade5-72a8d56c45ae

ARTIFACT_STORE: default

Name: default
ID: 7bd8cbfc-6c33-415d-9fda-6bd6063d05ab
Type: artifact_store
Flavor: local
Configuration: {'path': ''}
Workspace: default / d81c2a07-1b7c-4aa5-ade5-72a8d56c45ae

What happened?

We perform neuroscience research and I am evaluating if ZenML can be used to manage our compute pipelines. Not sure if this is a bug or maybe just a misunderstanding on my part but I wanted to start the ZenML server using the Docker deploy instructions but it just hangs. Was expecting to access the dashboard so I could start trying to run a pipeline I created.

Reproduction steps

1. Install ZenML

   pip install ./zenml-0.65.0-py3-none-any.whl

2. Start ZenML server

   docker run --name zenml-server -it -d -e ZENML_SERVER_USE_LEGACY_DASHBOARD=1 -e ZENML_DEFAULT_USER_NAME=admin -e ZENML_DEFAULT_USER_PASSWORD=simple -p 8080:8080 zenmldocker/zenml-server

Relevant log output

INFO:     Started server process [1]
INFO:     Waiting for application startup.
Not writing the global configuration to disk in a ZenML server environment.
Initializing the ZenML global configuration version to 0.65.0
Not writing the global configuration to disk in a ZenML server environment.
Creating database tables
Not writing the global configuration to disk in a ZenML server environment.
Creating default workspace 'default' ...
Creating default stack in workspace default...
Not writing the global configuration to disk in a ZenML server environment.

Code of Conduct

• I agree to follow this project's Code of Conduct

[wjayesh]

Hey @guzman-raphael, the logs look okay to me. You should be able to access the zenml dashboard on your localhost at port 8080 after you run the docker run command. Can you run docker ps? That will show you the running zenml server container and also what port to access it on.
let me know how it goes.