error installing / running AI extension under 5.3.1

Hi there - I tried to install the AI extension in order to run the OpenAI prompting examples using a completely fresh 5.3.1 install.
Whenever I try to load a workflow, I get prompted for missing extension nodes with the option to install them. After the install and workflow reload the same error re-appears.

the log shows these types of errors, I am using the bundled python


java.io.IOException: An exception was raised by the Python Proxy. Return Message: Traceback (most recent call last):
  File "C:\Users\MYUSER\AppData\Local\Programs\KNIME\bundling\envs\org_knime_python_llm\Lib\site-packages\py4j\clientserver.py", line 617, in _call_proxy
    return_value = getattr(self.pool[obj_id], method)(*params)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\MYUSER\AppData\Local\Programs\KNIME\plugins\org.knime.python3.nodes_5.3.1.v202407291559\src\main\python\_node_backend_launcher.py", line 1194, in loadExtension
    raise RuntimeError(
RuntimeError: Failed to load extension org.knime.python.llm from knime_llm with error: Traceback (most recent call last):
  File "C:\Users\MYUSER\AppData\Local\Programs\KNIME\plugins\org.knime.python3.nodes_5.3.1.v202407291559\src\main\python\_node_backend_launcher.py", line 1189, in loadExtension
    importlib.import_module(extension_module)
pydantic.errors.PydanticUserError: If you use `@root_validator` with pre=False (the default) you MUST specify `skip_on_failure=True`. Note that `@root_validator` is deprecated and should be replaced with `@model_validator`.

For further information visit https://errors.pydantic.dev/2.8/u/root-validator-pre-skip


	at org.knime.python3.FreshPythonGatewayFactory.create(FreshPythonGatewayFactory.java:86)
	at org.knime.python3.QueuedPythonGatewayFactory$PythonGatewayQueue.enqueueGateway(QueuedPythonGatewayFactory.java:301)
	at org.knime.python3.QueuedPythonGatewayFactory$PythonGatewayQueue.lambda$1(QueuedPythonGatewayFactory.java:267)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
	at java.base/java.lang.Thread.run(Unknown Source)
Caused by: py4j.Py4JException: An exception was raised by the Python Proxy. Return Message: Traceback (most recent call last):
  File "C:\Users\MYUSER\AppData\Local\Programs\KNIME\bundling\envs\org_knime_python_llm\Lib\site-packages\py4j\clientserver.py", line 617, in _call_proxy
    return_value = getattr(self.pool[obj_id], method)(*params)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\MYUSER\AppData\Local\Programs\KNIME\plugins\org.knime.python3.nodes_5.3.1.v202407291559\src\main\python\_node_backend_launcher.py", line 1194, in loadExtension
    raise RuntimeError(
RuntimeError: Failed to load extension org.knime.python.llm from knime_llm with error: Traceback (most recent call last):
  File "C:\Users\ckolve\AppData\Local\Programs\KNIME\plugins\org.knime.python3.nodes_5.3.1.v202407291559\src\main\python\_node_backend_launcher.py", line 1189, in loadExtension
    importlib.import_module(extension_module)
pydantic.errors.PydanticUserError: If you use `@root_validator` with pre=False (the default) you MUST specify `skip_on_failure=True`. Note that `@root_validator` is deprecated and should be replaced with `@model_validator`.

For further information visit https://errors.pydantic.dev/2.8/u/root-validator-pre-skip


	at py4j.Protocol.getReturnValue(Protocol.java:476)
	at py4j.reflection.PythonProxyHandler.invoke(PythonProxyHandler.java:108)
	at jdk.proxy7/jdk.proxy7.$Proxy17.loadExtension(Unknown Source)
	at org.knime.python3.nodes.PythonNodeGatewayFactory$KnimeNodeBackendCustomizer.customize(PythonNodeGatewayFactory.java:170)
	at org.knime.python3.nodes.PythonNodeGatewayFactory$KnimeNodeBackendCustomizer.customize(PythonNodeGatewayFactory.java:1)
	at org.knime.python3.FreshPythonGatewayFactory.create(FreshPythonGatewayFactory.java:82)
	... 5 more
2024-08-27 14:34:36,836 : WARN  : ModalContext :  : Node : MISSING OpenAI Authenticator : 3:106 : Node can't be executed - Node "OpenAI Authenticator" not available from extension "KNIME AI Extension (Labs)" (provided by "KNIME AG, Zurich, Switzerland"; plugin "org.knime.python.llm" is installed)
2024-08-27 14:34:36,838 : WARN  : ModalContext :  : Node : MISSING OpenAI Authenticator : 3:106 : Node can't be executed - Node "OpenAI Authenticator" not available from extension "KNIME AI Extension (Labs)" (provided by "KNIME AG, Zurich, Switzerland"; plugin "org.knime.python.llm" is installed)
2024-08-27 14:34:36,866 : ERROR : ModalContext :  : LoadWorkflowRunnable :  :  : Errors during load: Status: Error: 1.1 - Simple OpenAI Prompt 3 loaded with errors
2024-08-27 14:34:36,866 : ERROR : ModalContext :  : LoadWorkflowRunnable :  :  :   Status: Error: 1.1 - Simple OpenAI Prompt 3
2024-08-27 14:34:36,866 : ERROR : ModalContext :  : LoadWorkflowRunnable :  :  :     Status: Error: Node "OpenAI Authenticator" not available from extension "KNIME AI Extension (Labs)" (provided by "KNIME AG, Zurich, Switzerland"; plugin "org.knime.python.llm" is installed)
2024-08-27 14:34:36,866 : ERROR : ModalContext :  : LoadWorkflowRunnable :  :  :     Status: Error: Node "OpenAI LLM Connector" not available from extension "KNIME AI Extension (Labs)" (provided by "KNIME AG, Zurich, Switzerland"; plugin "org.knime.python.llm" is installed)
2024-08-27 14:34:36,866 : ERROR : ModalContext :  : LoadWorkflowRunnable :  :  :     Status: Error: Node "LLM Prompter" not available from extension "KNIME AI Extension (Labs)" (provided by "KNIME AG, Zurich, Switzerland"; plugin "org.knime.python.llm" is installed)

I tried downgrading to 5.2.5 and using the ai community nodes instead and that seems to work

Hi @ckolve ,

Thanks for reporting this issue. I’m glad you found a workaround by using version 5.2.5. Still, we would be happy to involve our development team in an investigation and hopefully provide you with an update path.

For the analysis, would you be able to share KNIME’s full log files?
Specifically, these are the <knime-workspace>\.metadata\knime\knime.log, but also a .log file one folder above the knime.log file.
You can quickly access the log folder via the Modern UI’s Menu entry:
image

(Side note: if you switch to a new workspace and restart KNIME, it will also start new log files in the respective workspace, so that you can be sure that any old information is not present in those logs.)

Kind regards
Marvin

3 Likes

Maybe one further request, just in case you happen to have conda installed on your system:

Could you run conda list from inside <knime-installation>\bundling\envs\org_knime_python_llm?

This would allow us to confirm that all Python packages were downloaded in the correct versions. :slight_smile:

3 Likes

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.