At the first glance nothing indicates it wouldn’t work as expected.
Just in case, did you try to refresh the Sheet? Updating 100+ rows mught result in lags on Google side.
Another possibility is that 200+ requests hit a per second/minute Google API calls rate. Try to update the sheet in batches by using the Loop node and Wait node within the loop.
This might help to pinpoint the root cause.
Example:
Yet another reason: something goes wrong on n8n or Google servers sides due to temporary service unavailability. Just retry your workflow after some time.
Could you paste your workflow with some sample data from “read: sheet” node pinned? Just select all nodes on the canvas, Ctrl/Cmd-C it to the clipboard, then in forum click the </> button and paste workflow json between triple backticks it creates for you.
Let’s take a look into data transformations you have.
Just in case: try scrolling your target sheet all the way down to see if the data has been appended somehere. A couple of Ctrl-Down should get you there.
There were reports of this sort behaviour, reasons are not clear though. My guess would be that there still can be some “invisible” data (e.g. formulas that resolve to “” and hence not visible but formulas still count as data) on the sheet causing n8n/Google to pick the “wrong” row to start adding new data at.
“Invisible” data could be remnants from not thorough enough cleaning.
Do you mind granting access to the Google Sheet? Should be safe if data is anonimized. Just requested access.
Not sure if this is the root cause, but at a quick glance this expression previousAddresses[positionB] and other expressions of that kind may resolve not as expected. This will create a key with contents of positionB variable. You may wanted to use previousAddresses["positionB"] or previousAddresses.positionB instead.
E.g. if positionB === 100 then previousAddresses[positionB]++ will try to increment previousAddresses["100"], not previousAddresses["positionB"].
Disregard this if you truly intend relative addressing like this and items like previousAddresses["100"] are defined and hold values that make sense. I just do not see the underlying data to test the hypothesis.
You’ve got a plenty of data on the n8n-sheet. See rows 545+
There are some values in column Numer wpisu that match the Update condition, hence no new rows added.
I added a column to store Timestamp for debugging purposes. This helped me to make sure the workflow works. Just not as expected.
And n8n will try to append any new data past last row in use. It won’t fill in gaps.
So try to delete obsolete data by selecting all rows (not just cells) from row 2 till the very end and delete via context menu action. Or simply set up a new sheet to collect outputs.
Hope this solves your case
P.S. Please feel free revoking my access to the spreadsheet.