I really don’t know anything about pdf or tracking what data is flowing, but have you tried making a copy of the file and seeing if the same thing happens when you open it. I have occasionally run across things in other software that seems like its got garbage lying in the file, and when you make a copy of it, it removes the garbage. If a copy doesn’t have the same issue then switch to using the copy.

I’m less and less impressed with Adobe reader - getting too bloated and too buggy.

Is it possible to recreate the file?

Is the content, say, 40 pages? - 40 lots of 10Mb maybe?

.

There are a few things that come to my head. The is the list of them:
1. Is the form active (having the input fields)?
2. Is the form created in PDF Distiller or LifeCicle Designer?
3. If the form was not created using any Adobe tools what software was used to create the form?
4. Is there any script in the form?
5. What version of Acrobat the form was created first time?
6. Are they any page templates?

Each of those questions can affect the document separately and it could be possible that there is more the one fitting to the scenario. I would suggest answer for them doing a simple investigation. If there is something that create complication to answer let me know. Only if those questions are ruled out we can look somewhere else. The information that you posted about the form is not enough to make any solid judgement. However I can believe that when you complete to answer for above questions at least one will open the door to the end of the problem

There is no form, it is simply a document with content on 154 pages. There are no scripts that we know of. Recreating it is not possible for us as it was sent to us from a reputable inernational company.

We have extensively analysed this ourselves, and simply don’t know what is happening, as such we are returning the document to the company that sent it to us with information about its activity and that we see it as a security risk.

I did try saving it using kpdf in Linux, then opening it again on windows, it still takes about 400mb of data from the file server.
And sending it back to its source isn’t as easy as we thought as they don’t know where in their company it was produced…

I’m hoping that you have the Acrobat Professional and the file has no passwords. If so try to open it in the design mode to see whether it is going to open in Acrobat or tend to open in LifeCycle. I could also check for Templates since they can be visible and invisible. Are there any pictures (images)?
You see without physical investigation it is impossible to point the problem. It is going to be only a guessing. I believe there is something with file compression. It could be a part of file that build your document or whole document itself. It also could be a triggered activity but again without physical investigation it is impossible pinpoint the problem. There are tools on the market that image the PDF document and create the new one. It could be the idea to get off the problems. You can also rescan the document and make it your own way. If there is a security matter I would do it and forget the problem especially when you say that the file is sensitive and you can’t show to any third party. By the wise, if the document is not protected divide it on pages (make several or as many as it is necessary files) and see which page is creating the data flow. Those options can save you from problems and give you some idea.

[...] A user is running into some unusual networking issues when downloading PDFs. Can you help solve the [...]