Strange behavior with binary data extraction from AWS file download

[Thomas_Goirand]

Describe the problem/error/question

I am running a workflow in order to extract files from AWS before translating them through the Deepl API. At some part of the workflow I get an file URL which from AWS and simply download it with a HTTP GET query.

What is the error message (if any)?

For large files > 15-20 Mo, I have issues when n8n tries to get the data from AWS for no reason. I did a few tests using google drive, no issue. And when I try downloading the file from AWS to chrome or postman, I have no issue.

Since the links have a 2 min lifespan with AWS, I cannot really give you something you can try yourself, but I give the json export of the node on which I have issues hereby.

Here is the download step in the workflow (I get the AWS URL from Asana API)

Here is what happens when the file is accessed from a Googledrive link:

Here is a json file resulting in the data export from AWS (data too large, I had to host it on Googledrive): https://drive.google.com/file/d/1jsDbwvH2tMfWd5glSsDhsa_j6Jz_Vy_q/view?usp=sharing

NB: In case you think about that reason, there is no login or credentials required with AWS since Asana API is giving all the access keys in the generated link and as I mentionned, the workflow works most of the time but would experience download issues for large files only.

Please share your workflow

Information on your n8n setup

• **n8n version:**1.36.4
• **Database (default: SQLite):**Default
• **n8n EXECUTIONS_PROCESS setting (default: own, main):**Default
• **Running n8n via (Docker, npm, n8n cloud, desktop app):**Docker
• **Operating system:**Debian

[barn4k]

hello @Thomas_Goirand

So what will happen if you click on the “Show data anyway” button? Is there a binary panel?

For now, it seems that the data is converted into json, so it is not a file.

[Thomas_Goirand]

Hey @barn4k thanks for the support!

I tried, and with +30 Mb, it’s impossible to handle. It would make crash Firefox or Chrome anyway. The only thing I got looks like a bug, but here it is from a .pptx download from AWS, making my app freezing after I get this:

Indeed it looks like there is no filetype or anything else in the data. What can we suppose? There is an issue from AWS? Maybe the file is chunked?

Any hint would help me at this point.

[barn4k]

Try to set the File output option

[Thomas_Goirand]

Hey @barn4k it worked thanks to your help!