And after a recent upgrade to 1.91.3 (cloud), it’s back. In spades.
In brief: All stored workflows work perfectly. However, in-dev workflows (in the active dev UI) throw constant errors, the second time we run any chain (including the simplest debug-style stuff). So, first run is fine, then second run:
“Problem Running Workflow: The expression evaluated to a falsy value: a.ok(to)”
… and the error repeats on every run after that.
This is happening on all functions, all chains, every time we test.
The only way to stop it is to CTRL>SHIFT>R. That clears everything, and after that the run executes perfectly. ONCE. Then the next execution throws the error again.
This makes development extremely unpleasant, to say the least.
If n8n can acknowledge that they’re looking at this (since it’s been around for a while), we’d be grateful. We’re in the middle of an intensive dev cycle and everyone is pretty unhappy about this.
Yes I’ve seen this issue come back now too… I am also on the latest stable version. This is a pretty critical issue, completely unable to run my workflows even after clearing and restarting my development environment.
Hey, I’m having the same issue. My workflow was working perfectly and now I can’t even receive a webhook. I’m doing my tesis with n8n, any ideas on how to fix this? I erased my complete executions history, restarted, but the problem persists.
I have the exact same problem since 1-2h. Only way I can run the workflow ONCE is download it and upload it. I’m self hosted via elest.io running latest version (i tried it with other versions too → same problem)
Not sure if this is helpful but i was able to replicate this by setting the limit parameter to a number beyond that which the input held.
e.g. I used the reddit node to look up a single post. Then i used another reddit node to look for comments on that post, setting the limit param to 100 when there were less than 100 comments on it
