One run, two Facebook posts: the retry question to ask first

A recent Make Community case reported one scenario run, one input bundle, and one returned Facebook post ID — but two public Facebook posts.

The first recovery question is not “should I rerun it?” It is “could the provider already have accepted a write that my execution log does not fully explain?”

Before any replay:

1. freeze the scenario and export the execution;

2. record the request, bundle, execution, returned provider ID, and both public post URLs;

3. compare content, timestamps, actor/page identity, and media;

4. inspect retries, error handlers, parallel routes, webhooks, and any second scheduler;

5. retry only the segment proven not to have created a provider object.

That does not assign blame to Make or Facebook. It preserves enough evidence to distinguish a duplicate route from an ambiguous acknowledgement.

I operate ANKK at ANAKONN. ANKK has no built-in AI writer; it connects content prepared by people, external AI tools, or scripts to scheduling, status tracking, and provider-original verification.

The complete checklist, including Make’s retry and rollback boundaries:

https://ankk.app/en/posts/facebook-duplicate-posts-one-automation-run?utm_source=producthunt&utm_medium=product_forum&utm_campaign=ankk_72h_qualified_visitors_20260814&utm_content=en_product_hunt_forum_facebook_duplicate_one_run_v1

What provider-side field would you add before permitting a retry?