Hi,
when there is an error in a workflow, it is sometimes caused by some logic error done in previous nodes rather than in the node where the error actually happens. Would it be possible to add a Retry from beginning (of the currently saved workflow) option with the initial input of the first node? This would be useful for those cases in which events cannot be easily retriggered (e.g. saved list which gets erased after sending data to n8n)
Yeah this is very much needed.
Especially when webhooks and such are used. Starting from the node that error’d is pretty useless, as most often it is something earlier in the flow that made the data wrong for this node to continue for example.
Also when simply rebuilding part of the workflow because it turned out to be faulty would benefit from an easy retry from the start.
3 Likes
Absolutely need this! I have a long list of failures I’m having to copy the webhook request JSON and run manually because during my flow I have to make authentication callouts that have tokens that expire.
1 Like
Especially in cases when I’ve purposely failed a workflow with the stop node because a previous action has been detected as failed. No amount of retrying from the failed node (stop and error) will ever succeed, however if I reran the whole workflow with the same trigger data it would succeed.
1 Like
We’d need two additional options to re-execute a workflow that failed:
- Retry with original worflow (from the start of the workflow)
- Retry with currently saved workflow (from the start of the workflow)
Of course if the workflow was triggered by a webhook call, re-executions need to use the payload data received by the webhook when the execution failed.
1 Like
Would be very helpful for proper debugging! Optional ideas: Seeing the number of retries of individual nodes (eg API Calls) and running duration of nodes. Still doable now already, with effort tho.
This is a must. It makes many things soo harder