Salesforce CaseUpdated trigger not always firing

Describe the problem/error/question

Hello! I have a flow starting with a Salesforce CaseUpdated node. While there are multiple updates on cases (status updates, ownership updates) the node is not always firing. It only fires sometimes and my poll time is every minute. Any idea why this is happening?

Please share your workflow



Information on your n8n setup

  • n8n version: Version 2.13.3
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app): Cloud
  • Operating system:

could be a polling timing issue — case updates might happen faster than your 1-minute poll cycle catches them. does your flow check case.updatedAt or maybe a status field? also salesforce triggers sometimes get swamped by other updates, might need to filter more specifically to just case changes.

Thank you Benjamin for your quick reply, really appreciate it! Okay, that makes sense!
The flow checks the status, but later on. Could you maybe help me on how to filter more specifically? Basically, I want the flow to run when the status changes to “Response Received”.

So this is actually a known bug with the Salesforce trigger — once a case ID fires the workflow, it gets stored in dedup data and wont fire again for the same record. Check if its always the same cases getting missed: Salesforce Trigger fires only once for repeated Account updates · Issue #21028 · n8n-io/n8n · GitHub

yeah, that dedup thing achamm mentioned is probably why the same cases get missed. for the ‘Response Received’ filter: add a conditional check right after the trigger on the Status field, and only continue if it matches. wont fix the dedup issue for repeat updates on the same case tho.

Thank you both for your responses! Hope the bug gets fixed soon.

It’s strange behavior, as it fires in the first update, then it will miss a few and then it will fire again for some cases.
I have a filter, but since the trigger doesn’t fire in every update, it’s not really useful.
Appreciate your time both!

yeah that pattern — fires once, misses a bunch, fires again — is exactly what you’d expect from dedup TTL. the dedup record eventually expires, so the case becomes “new” again and triggers. explains why it’s not completely dead, just inconsistent.

not much to do workflow-side until the bug gets patched, but at least the behavior makes sense now

Yes, it finally makes sense now, I thought I was going crazy :sweat_smile:
Thank you Benjamin!

no problem! glad the dedup behavior makes sense now. hopeful the n8n team gets that patched soon — looks like it’s been a known issue for a bit. good luck with your workflow!"

1 Like