Oracle JD Edwards and WooCommerce integration

What moves between them

The main data flow is from WooCommerce into Oracle JD Edwards. When customers place orders in WooCommerce, the connector receives webhook notifications, extracts the order amount and customer information, and posts revenue GL transactions into the Oracle JD Edwards GL account designated for online sales. Customer records from WooCommerce are also synced into the Oracle JD Edwards Address Book (F0101) so the GL is reconciled with the customer master. The connector tracks each order by ID to avoid duplicate GL postings. Because Oracle JD Edwards cloud is pull-only and has no webhooks, the connector also polls for any GL-side changes and validates them against the WooCommerce record.

How ml-connector handles it

ml-connector stores the AIS Server base URL and credentials encrypted and obtains a session token by calling POST /jderest/v2/tokenrequest at startup and before expiry. WooCommerce webhooks are configured to push order.created and customer.created events to an ml-connector ingest endpoint; the connector validates the HMAC-SHA256 signature using the Consumer Secret and processes the webhook payload. For each WooCommerce order, the connector calls the Oracle JD Edwards GL posting orchestration endpoint to create a revenue transaction in the designated GL account for online sales, with the order reference number as the batch ID to prevent duplicate posts if the webhook retries. Customer records flow into the Oracle JD Edwards Address Book (F0101) so the GL matches the customer master. Session tokens are cached and refreshed when a request returns HTTP 444 (invalid token). Because AIS Server response times depend on customer infrastructure, retries use exponential backoff with jitter. Order numbers are stored in the audit log so any failed GL post can be replayed once the downstream system is restored. IP allowlist configuration at the AIS Server is common, so the customer must confirm that the connector egress IP is whitelisted.

A real-world example

A mid-sized retailer runs WooCommerce on their WordPress site for e-commerce and Oracle JD Edwards EnterpriseOne on-premises for accounting, inventory, and customer management. Before the integration, the accounting team exported order summaries from WooCommerce daily, manually entered the revenue amounts into Oracle JD Edwards, and periodically reconciled the customer records between the two systems. At month-end, discrepancies between WooCommerce sales and the GL took days to track down. With WooCommerce and Oracle JD Edwards connected, each order is automatically posted to the GL within minutes of creation, with the customer record and revenue amount both synced. The GL stays current with actual sales without manual entry, and month-end close starts with the sales accounts already reconciled.

What you can do

• •Post WooCommerce orders into Oracle JD Edwards GL as revenue transactions under the correct GL account and cost center.
• •Sync WooCommerce customer records into the Oracle JD Edwards Address Book (F0101) so the GL reconciles with the customer master.
• •Validate WooCommerce webhook signatures using HMAC-SHA256 and the Consumer Secret to prevent fraudulent posts.
• •Authenticate Oracle JD Edwards via the AIS Server session token API and handle token refresh when tokens expire.
• •Track each order by ID and audit every GL post, with replay capability if a downstream API call fails.

Questions

Which direction does data move between WooCommerce and Oracle JD Edwards?

The main flow is WooCommerce into Oracle JD Edwards. Orders and customers from WooCommerce are posted as GL transactions and customer master records into Oracle JD Edwards. GL postings are created via orchestration endpoints, which support named batch processing to prevent duplicates. Oracle JD Edwards writes do not flow back to WooCommerce.

How does the integration handle the fact that Oracle JD Edwards has no native webhooks?

WooCommerce has native webhook support and pushes order.created and customer.created events to the ml-connector ingest endpoint in real time. ml-connector also polls the Oracle JD Edwards GL on a schedule to validate that all orders have been successfully posted. If a GL post fails, the audit log stores the order number so it can be replayed once the connection is restored.

What happens when the AIS Server session token expires?

Session tokens are typically valid for 30 to 60 minutes. ml-connector caches the token and refreshes it before expiry or when an API call returns HTTP 444. The connector also stores the AIS Server base URL and credentials encrypted, so token renewal happens transparently without re-configuration.