Hello! I’ve been using the “Continue (using error output)” option on nodes but just recently realized that if I don’t put a stop and error node on the error output it will keep executing my flow with broken data. I want to avoid this. My initial thinking was the feature branched to a new error path and stopped the normal path so I can handle each error individually without it messing up the rest of my flow. Is there anyway to do this? Ideally this is fixed with a stop node but there’s only stop and error which causes my error workflow to run which I don’t want in specific error cases (ie I want to stop the flow without error but it continues instead).
No because (at least from how I understand it) - the workflow continues running with empty data in the “success” branch and then sends the error down the error branch (so a do nothing node will not stop execution of the “success” branch). I’m looking for a way for nothing to come out of the success branch or a way to stop the workflow without a Stop and Error node (since this would cause my error workflow to run) on the error branch thereby stopping the success branch.
Here’s how the error in workflow works. If there is an error in that step, it sends it down the error path. If there is no error, it sends it down the success path. It doesn’t send empty data down the success path. It’s binary, one or the other.
Do you have some sample data where you are seeing the issue?
This doesn’t seem to be the case. In the latest update when setting “Continue (using error output)” the node UI even says it will continue in case of fail.
I do not want execution to continue on the success branch (which it indeed does). See a below screenshot of an example run I did - notice the HTTP Request Node errored but the If node connected to it was still run (with empty data causing it to break).
