n8n-io/n8n · error · NodeOperationError

Failed to initialize Azure OpenAI client: ${error.message}

Error message

Failed to initialize Azure OpenAI client: ${error.message}

What it means

Top-level catch in LmChatAzureOpenAi.supplyData(). It re-throws NodeOperationError as-is so inner messages (e.g. error 694, 696-699) survive, and wraps any other exception in a NodeOperationError whose message is 'Failed to initialize Azure OpenAI client: ' + error.message, passing the original as the cause. It logs the error first for diagnostics.

Source

Thrown at packages/@n8n/nodes-langchain/nodes/llms/LmChatAzureOpenAi/LmChatAzureOpenAi.node.ts:147

						}
					: undefined,
				onFailedAttempt: makeN8nLlmFailedAttemptHandler(this),
			});

			this.logger.info(`Azure OpenAI client initialized for deployment: ${modelName}`);

			return {
				response: model,
			};
		} catch (error) {
			this.logger.error(`Error in LmChatAzureOpenAi.supplyData: ${error.message}`, error);

			// Re-throw NodeOperationError directly, wrap others
			if (error instanceof NodeOperationError) {
				throw error;
			}

			throw new NodeOperationError(
				this.getNode(),
				`Failed to initialize Azure OpenAI client: ${error.message}`,
				error,
			);
		}
	}
}

View on GitHub (pinned to 5ac6606e81)

Solutions

Read the wrapped error.message/cause — it identifies the underlying failure; fix that root cause.
Verify endpoint, resourceName, apiVersion, and deployment id in the credential.
For Entra OAuth2, confirm the token credential can fetch a token (see error 696/697).
Check network/proxy and Azure OpenAI quota/access from the n8n host.

Defensive patterns

Strategy: try-catch

Validate before calling

// Validate config before constructing AzureChatOpenAI
if (!configCredentials.resourceName && !configCredentials.endpoint) {
  throw new Error('Azure OpenAI resourceName or endpoint is required');
}

Type guard

const isNodeOperationError = (e: unknown): e is NodeOperationError =>
  e instanceof NodeOperationError;

Try / catch

try {
  return { response: model };
} catch (e) {
  if (e instanceof NodeOperationError) throw e; // preserve specific messages
  logger.error(e);
  throw new NodeOperationError(this.getNode(), `Failed to initialize Azure OpenAI client: ${(e as Error).message}`, e as Error);
}

Prevention

Pre-validate endpoint, resourceName, apiVersion before client construction.
Keep specific NodeOperationError checks above the generic wrapper.
Log the original error for diagnostics, then re-throw wrapped.

When it happens

Trigger: Any non-NodeOperationError thrown during Azure client construction — network error instantiating AzureChatOpenAI, missing env, malformed endpoint, SDK type error, or an unhandled configuration issue — bubbles into this catch.

Common situations: Endpoint URL malformed; resourceName empty; apiVersion unsupported by the SDK; proxy/network issue on first request; SDK version mismatch; Entra token credential throwing an unexpected shape.

Related errors

AI-assisted analysis of n8n-io/n8n@5ac6606e81 (2026-08-12). Data as JSON: /api/errors/b582a26daee7e420. Report an issue: GitHub.