Make processed empty CSV rows

Hi everyone! hopefully someone can help me out here, I have a scenario attached to my app, which processes data updates from an end user csv update. Today, one of the users uploaded a csv with +1.000.000 empty rows (only 10 rows had data to process).

The user uploaded the file, unaware of the damage she was causing, twice, since the file took so long to process. Now I’m exceeded 5276% on my quota, paying for additional operations to cover for this would cost north of $1.600 dollars.

So my questions are:

  • Has anyone had an issue with this and has Make/Integromat been helpful about it?
  • Is there a way to stop integromat from going overboard like this?

My app is now useless with integromat being completely frozen. I already wrote to support but have yet to receive a response.

Best regards.

Hi @Alejandro_Estela,

Sorry to hear about your situation :face_holding_back_tears:

To answer your questions:

  1. I have never had an issue like that so I have no idea how Make will handle this. I imagine it is rare for them too so I don’t think they have a standard procedure.

  2. I am not sure how this happened for the following reasons:

  • how did you not reach your operations limit? This should have stopped the process.
  • how did you face the 40 minutes limit for one execution before processing 1m rows.

To protect yourself against cases like this, you should always assume that users are going to do unexpected things. Some ideas:

  • set a maximum number of rows the users can upload. If a CSV is over the limit you void the process
  • filter out empty rows

Keep us posted

1 Like

I wonder if the max out on operations pausing a scenario can be triggered during one modules operation. Seeing a history on this scenario run would be fascinating. @Alejandro_Estela Can you take a snapshot so we can see 1M ops on one google sheet update?

Hi @Alejandro_Estela, welcome to the community :wave:

So sorry to hear about this situation of yours, that’s really unfortunate :cry:

You said you’ve already logged a support ticket which is really the best thing to do in such a case. When things like this happen, our team judges each case individually and tries to see if there’s anything that can be done.

Good luck with this!