Issue with workflow being unable to run after upgrading to version 1.11.1

[henrybjsoft]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:)
• Please do not modify this template :) and fill in all the required fields.

Content

I upgraded from version 1.10.1 to 1.11.1, and my workflow encountered an error and could not run. I had to downgrade to version 1.10.

Below is the error message:

{
"error_message": "1 validation error for VariableMessage\n Value error, Only basic types and lists are allowed. [type=value_error, input_value={'variable_name': 'tail_i…: None, ‘stream’: False}, input_type=dict]\n For further information visit Redirecting... ",
"error_type": "WorkflowNodeError"
}

The node is a workflow TOOL. I checked the TOOL's running logs, which were normal, but the error occurred in the upper-level workflow.

Please help me identify the issue.

[dosubot[bot]]

Hi @henrybjsoft! I'm Dosu and I’m helping the dify team.

The error happens because Dify 1.11.1 enforces stricter type validation for TOOL node parameters and outputs. Now, tool parameters defined as string or select must receive a string, not a dict or other structure—passing a dict to a string parameter triggers the "Only basic types and lists are allowed" error. This stricter check did not exist in 1.10.1, so workflows that previously worked may now fail after the upgrade. Also, if any workflow output contains None/null, it will cause this validation error in 1.11.1, while older versions were more permissive.

To fix this, make sure your workflow and agent configurations pass values that match the expected types for each tool parameter. For string/select parameters, pass a string; if you need to pass a dict, change the parameter type to OBJECT in your workflow tool config. Also, ensure that workflow outputs do not contain null values—provide default values or filter them out before passing to the TOOL node. If your workflow was created in an older version, check that every output variable includes a value_type field (like "string", "number", "object", etc); missing this field can also cause validation errors after upgrade. You may need to manually update your workflow node data to include value_type for each output.

As a temporary workaround, you can set the tool_node_version field in your workflow or agent node configuration to null to restore legacy permissive behavior, but this may not be supported in future versions. If you need a quick fix for missing value_type, try re-importing the end node and setting variables again, which may auto-populate the missing field. Always back up your workflow configs before making changes and test thoroughly after applying fixes.

References:

• Stricter type validation and solutions
• Schema change requiring value_type field
• Workaround for None/null values
• Compatibility flag tool_node_version

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

   

[henrybjsoft]

Thank you. I'll check as you said.