I don’t know if this has already been considered, but I think restarting an existing run with the first step after the trigger would be nice.
For example, I have a trigger for new files in Google Drive, which are processed in various steps. If one of the steps is running into a temporary error (3rd party API unavailable, etc.), the whole flow fails.
To restart it, I’d have to re-create the trigger’s conditions manually, i.e., delete and upload the same file again to Google Drive.
Hence it would be nice if the runs preserve the parameters of the trigger (file upload, webhook params, etc.) and can be restarted from there as a new run.
Just trying to think with you, how does the flow start? Is it via Webhook of schedule?
Because if I am correct you could create multiple files in Google docs with the same name, it uses a different unique ID.
With the Webhook you can select the specific hook to continue and schedule will just rerun the data.
Do you have some examples of the errors you run into by testing the whole flow?
yes, I could upload the same file, and it will have a different ID.
But the file itself is uploaded by another program (an audio recorder), and it does it automatically.
So, I have to go into Google Drive, download the file, and upload it again.
This happened for a flow during which I used OpenAI, and their API had a temporary issue, so the whole thing failed, and I had to upload the file as I have described manually.
Don’t you mean this? there u can select a previous trigger and run the flow again?
If not, it’s probably me not really understanding the request.