I am attempting to download everything down from the Amazon Cloud Drive since my NAS broke earlier in the year. After anywhere from 1GB to 25GB of downloading without problems I receive the following error: “Can’t sync Amazon Cloud Drive. Unexpected error. Please try again later.” I try immediately after hitting the Ok button and the download continues. Unfortunately I do not want to babysit the download process. I have it set to Always download (current). I understand that sometimes it would drop connections but it does not appear that I’m dropping service that often. Also it would be nice that if it came across this error that it would just log the error and then try again and the user could set how many times it would attempt it until it would stop and then the error would appear.
I am just using Amazon Cloud Drive so this is the only one I am seeing it on. It is for all various types of files and folders within the saved storage area. I have restarted the software and the machine and the problem still persists.
Can you provide a screenshot of the error and also send a diagnostic from the tray menu? We do try to retry these types of operations, but, since its an unexpected error, the could be a departure in that behavior.
Attached is a screenshot of the error and the diagnostics were just sent. Thanks for looking into the problem.
Any update on this issue?
I also get this error every time I try to sync from Amazon to my computer. Large files (a few GBs) are never fully synced (this unexpected error stops the sync); some small files make it, but if I try to sync many of them, at some point this error kicks in and stops the others from syncing.
What’s really frustrating is that it never saves it’s progress to retry downloading the rest of the data after the error. If I hit sync again, it starts from scratch and fails with unexpected error again.
Amazon doesn’t support resuming downloads natively, so there is no way to resume.
We are planning on making the “sync all” operation more tolerant of exceptions which should help these cases. Large file operations have been shown to be problematic for some user. It’s actually one of primary reasons we came up with IFS.
The next time you hit this issue, can you ping this thread and send over a diagnostic? I would like to take a look to see if I can identify the exact error that is being hit. It looks like Amazon is literally returning an unexpected error, but it could be something we are just not expecting.
I see this is still happening with large files. I can never remember to send diagnostics when it happens and I suppose it’s useless to send them now. The following may help out more though:
I found that the tempLink Amazon creates when requesting a download for a specific file is resumable! It doesn’t do that when requesting a batch archive, but still, this discovery finally makes ACD useful! Perhaps you can implement something similar in your sync software for Amazon; request a tempLink for a specific file if it’s over 100Mb, start downloading, handle errors and keep resuming since the link supports it, then move on. Easy as pie, right?
I’d also recommend resuming if an error happens for smaller files, since as I recall, currently the software just stops after the first error even if there are hundreds more files to download. It can just take it from where it left off, keep trying as long as there’s progress.
I don’t know if the API you’re using supports this tempLink creation, if not, and it depends on Amazon, perhaps we could pressure them into enabling it, they’re using it already after all.
I have requested, on multiple occasions, that Amazon allow us to use their current resume API (which does exist, but is undocumented), but they have not given us access to it yet. This is something we are still looking into, though, since upload resume would be a huge boon to our users.
What are you looking at to determine resume support on your side?
I’m just doing some http requests in the browser. The tempLink url can be found in the GET request response after starting a download on a file. Aside from the empirical evidence of being able to pause and resume a download from where it left off, you can send a custom request to the tempLink url with the Range header and Amazon complies:
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.12; rv:50.0) Gecko/20101001 Firefox/50.0
Accept-Encoding: gzip, deflate, br
Cache-Control: public, max-age=7200
Content-Disposition: attachment; filename="movie.mkv"
Content-Range: bytes 250000-567599836/567599837
Date: Sun, 27 Nov 2016 18:50:10 GMT
The GET request for the tempLink looks like this:
An API library for XYZ language may not have that option, but if you can send a custom request with tempLink=true as a query parameter, maybe it’ll work
Appreciate the follow-up. Thanks!