How can I recover my files from Dropbox Backup?

Thank you for the reply. But it's not just that it's not ideal--I would be happy to have a less than ideal solution--it's that it seems to be impossible: I've been trying for the better part of a week with no success. How can it be that I've been paying for this service for this long and was never told (until now) that many of my files are un-recoverable?

Would you mind clarifying what you mean by "unrecoverable", z1717?

I mean, it will definitely be hard to break them up in batches and download a few at a time, but have you had issues downloading them that way?

Sure, I will explain what I mean. Maybe someone else is having the same problem and this might help them, although hopefully they find a better way to do this than what I came up with.

Since my last post I have fortunately been able to download nearly all of the stuff in my backups, although it has not been easy. The file size/number limitations meant that I often had to go many levels deep to find units small enough to download. This meant keeping clear records of what I had or had not yet downloaded so as not to risk losing my place.

My worries about unrecoverability were mainly about a particular folder, one with no subfolders and that contains around 50,000 files. If I had compressed them into an archive I would have had a much easier time, but the Dropbox.com interface does not handle long filelists well.

The basic problem is that every task involved in downloading a batch of files in this way strains the site to its breaking-point. Getting through the whole process without the page freezing up (forcing me to refresh and start over) is difficult.

It takes a long time for the page to load the full list of files. Scrolling through them is dicey: it takes several seconds for the page to render the new text after each scrolling action, and the list will often go black (in my experience, this means I have to reload the page and start over). Checking the box to select a file is also dicey: the page often locks up during this action, which is another reload. After shift-clicking to select a batch of files, I then have to get back to the top of the page, a long scrolling action during which it often freezes (reload). If I make it, the download button sometimes stops responding (reload).

When it does respond, I hit another roadblock. The official limit for each download is 10,000 files, but I have found that for files at the same directory-level the limit is more like 2,500 (at least that's the most I've been able to get). Any more, and the page returns an error saying "attempted to zip too many files". On occasion I get this error while trying to download fewer than 1,500 files. I have to intuit roughly how much downward scrolling will get me the right number of files, since I don't think you can batch-unselect so if I select too many that's a reload. The "real" limit seems to vary from moment to moment, so I am conservative with my selections. Better to get a few files downloaded than to risk making it all the way to the download button only to find out I've selected too many.

When I was able to get a batch of files downloaded, I then had to get back to where I left off to download the next batch. This took some time. It was often quicker to reload the page than to try to un-select the current batch of files, since this action often froze the page.

I was able to come up with some workflow improvements to speed things up. One was to keep several browser windows open at all times so they could load silently in the background and I wouldn't have to wait for the filelist each time. Another was to avoid releasing the scrollbar during long scrolls: in my experience, click-dragging and releasing several times is more likely to result in one of the lock-up events I described above. For finer location adjustments, the mouse scrollwheel works better than the browser scrollbar.

Every other backup tool I've used has automated this sort of thing, or at least provided API access so that if I am willing to do a bit of reading, I can set something up myself. I suppose I should have paid more attention to the fact that Dropbox Backups was officially in beta before I used it in this way; my mistake was that I never stopped to consider that a backup tool might lack this functionality. I have found a new backup service, and this time I read the fine print.

(I was mostly doing all this on a Core Ultra 9 Ubuntu machine but I also tried it on an M4 Pro Mac to see if that would make a difference [it didn't, for me]. I used both Chrome and Firefox at different points; they behaved the same.)