Parsing Genesys Cloud Webhook Payloads in Node.js Lambda

NICE CXone API & Integrations
1

  • Does anyone know how to correctly parse the nested JSON structure when a Genesys Cloud webhook triggers a Node.js Lambda function? I am experiencing consistent failures in my CI pipeline where the Terraform apply succeeds, but the downstream Lambda handler throws a runtime error due to unexpected payload formatting.

  • I have configured a webhook in Genesys Cloud to POST to my AWS API Gateway endpoint. The webhook is triggered by a conversation event. The Lambda function is deployed via our GitHub Actions pipeline using the Terraform CX as Code provider. The goal is to extract the conversation_id and event_type from the payload.

  • The Lambda handler code is as follows:

exports.handler = async (event, context) => {
 try {
 console.log('Received event:', JSON.stringify(event, null, 2));
 const body = JSON.parse(event.body);
 const conversationId = body.conversationId;
 const eventType = body.eventType;
 
 if (!conversationId) {
 throw new Error('Missing conversationId in payload');
 }
 
 // Process conversation...
 return {
 statusCode: 200,
 body: JSON.stringify({ message: 'Processed successfully' })
 };
 } catch (error) {
 console.error('Error processing webhook:', error);
 return {
 statusCode: 500,
 body: JSON.stringify({ error: error.message })
 };
 }
};

  • The error returned by CloudWatch Logs is: TypeError: Cannot read properties of undefined (reading 'conversationId'). This indicates that body is not an object with conversationId. However, when I inspect the logs, event.body is a stringified JSON object. I am using JSON.parse(event.body) to convert it. Why is conversationId undefined after parsing?

  • I have verified the webhook configuration in Genesys Cloud. The payload sent by Genesys Cloud is:

{
 "id": "12345678-1234-1234-1234-123456789012",
 "eventType": "conversation:updated",
 "conversationId": "87654321-4321-4321-4321-210987654321",
 "timestamp": "2023-10-27T10:00:00Z"
}

  • Is there a specific header or content type issue that prevents proper parsing? I am using the default Lambda runtime settings. The Terraform configuration for the API Gateway stage is set to passthrough_behavior as WHEN_NO_MATCH. Could this affect the body parsing?

  • I have also tried using event.body directly without parsing, but that results in a string, which does not have conversationId as a property. I need to access the properties as an object. How can I ensure the body is correctly parsed into an object with the expected properties?