iflytek/astron-agent · error · CustomException

ENG_PROTOCOL_VALIDATE_ERROR

ENG_PROTOCOL_VALIDATE_ERROR

Error message

Current workflow does not support node type: {node.get_node_type()}

What it means

Raised by NodeFactory.create when the workflow engine's tool_classes registry (built from workflow.engine.nodes.cache_node) has no node class registered for the given node's type string. The engine cannot instantiate a node whose type it does not recognize, so it fails protocol validation with ENG_PROTOCOL_VALIDATE_ERROR. This usually means the workflow definition references a node type from a newer/older version or a plugin node that is not loaded.

Solutions

1. Check the node_type value in the workflow definition against the registry keys in workflow/engine/nodes/cache_node.tool_classes and correct any typo
2. Upgrade or fix the deployment so the plugin/module registering that node class is loaded (verify cache_node imports all node modules)
3. Remove or replace the unsupported node in the workflow DSL with a supported equivalent type
4. If the workflow came from another version, re-export it from a matching engine version

Example fix

// before (workflow DSL)
{"type": "agentNode_v3"}
// after
{"type": "agent"}

from workflow.engine.nodes.cache_node import tool_classes
def node_type_supported(node_type: str) -> bool:
    return node_type in tool_classes

try:
    node = NodeFactory.create(node_data)
except CustomException as e:
    if e.err_code == CodeEnum.ENG_PROTOCOL_VALIDATE_ERROR:
        log.warning("unsupported node type skipped: %s", node_data.get("type"))
        return None
    raise

When it happens

Trigger: Calling NodeFactory.create (directly or via workflow instantiation) with a node whose get_node_type() returns a string not present in tool_classes — e.g. an imported workflow DSL containing a node type this deployment does not support, or a typo in a hand-edited workflow JSON.

Common situations: Importing an exported workflow from a different product version that contains node types not yet released in this engine; disabling a plugin module so its node class is never registered; hand-editing workflow definitions and misspelling the node type.

Understand the failure class

Background: Invalid enum value errors: "Unknown type", "Invalid scope", "must be one of" — when a string is not on the library's allowed list — this error's family across 23 libraries.

Related errors

• WORKFLOW_VERSION_PUBLISH_FAILED
• 8008
• 8008
• WORKFLOW_PROTOCOL_NODE_INFO_CANNOT_EMPTY
• APPID_CANNOT_EMPTY

AI-assisted analysis of iflytek/astron-agent@5e758547a8 (2026-09-12). Data as JSON: /api/errors/d6e81024ca2ab365. Report an issue: GitHub.

Appendix: source

Thrown at core/workflow/engine/node.py:773

class NodeFactory:
    """Factory class for creating workflow engine nodes."""

    @staticmethod
    def create(node: Node, span_context: Span) -> SparkFlowEngineNode:
        """
        Create a node instance.

        :param node: Node data
        :param span_context: Span context for tracing
        :return: Created node instance
        :raises CustomException: When node type is not supported
        """

        from workflow.engine.nodes.cache_node import tool_classes

        node_class = tool_classes.get(node.get_node_type())
        if not node_class:
            raise CustomException(
                CodeEnum.ENG_PROTOCOL_VALIDATE_ERROR,
                err_msg=f"Current workflow does not support node type: {node.get_node_type()}",
                cause_error=f"Current workflow does not support node type: {node.get_node_type()}",
            )

        if not node.data:
            return node_class(span=span_context)

        # Get basic configuration
        inputs = node.data.inputs
        outputs = node.data.outputs

        # Build retry configuration
        retry_config = node.data.retryConfig
        retry_config.custom_output = NodeFactory._check_custom_output(
            retry_config.custom_output, outputs, span_context
        )

View on GitHub (pinned to 5e758547a8)