Workato Developer API - Trigger only works in one recipe

[saraiva]

[UPDATE 27-Aug-2025] Workato's support team is already looking at this situation. [/UPDATE]

Hello everyone,

After several attempts I'm not able to understand what's going on, hence my question here.

I've built a connector to access Workato Developer API and it has a trigger named "Deployment complete", which uses the list deployments endpoint (GET /api/deployments) to pool for new deployments. According to documentation (FAQ 5), this endpoint only works from DEV and I've created an API client with all permissions to DEV environment.

I've created a recipe and started it in DEV environment. It works pretty well, triggering correctly when there's a new deployment (it polls every 15min).

The problem appeared when I created a second recipe, with the same trigger, same connection config...and it never triggers.
Does anyone have idea why? Can it be some Workato limitation?

I appreciate your time and effort to help me sort this out.

Thanks!

[Bhagya_pola]

Hi @saraiva,
I noticed the same issue when we kept On Prem as the trigger for two recipes and the second one never triggered. We didn’t use any workarounds for this, but I feel it might be a Workato limitation since it creates an ambiguity.

[saraiva]

Hello @Bhagya_pola !

Thanks for your reply.

How could it create an ambiguity? I can use other apps' triggers in more than one recipe - still, if that's the case, can you point me to any documentation around that?

Also, I tried the following: I've stopped both recipes and started just one and the trigger worked, then I stopped it and started the other one, and the trigger didn't work (of course, I ensured a new deployment happened for the trigger to happen).

[Bhagya_pola]

Which field are you using in the dedup block of your connector code? I think that might be the issue since polling relies on the dedup to decide what is already been processed, it could be why the other recipe is not receiving the trigger.

[saraiva]

dedup: lambda do |record|
  "#{record['id']}@#{record['updated_at']}"
end

According to documentation "To prevent a job from being repeated (this might happen due to a bug on the API end), use the dedup key which tells your connector how to create a unique signature for each record. This signature is stored for each recipe and if a record with the same signature is found, no job will be created." (highlight is mine)
So duplications across all recipes should not be the issue.