2

It seems that when I use a web browser to download a large zip file (2-4gb), I go to unzip it and viola, corrupted. This happens so often. Are browsers not reliable means to download a file (http) ? Is there a better way? I need to download such a file and I do not seem to be able to get it successfully using chrome.

1
  • 1
    what is more interesting to me is that the browser doesn't notify right after downloading that it is corrupt but it finishes the download without any checks or warnings in year 2019.
    – aycanadal
    May 27, 2019 at 19:52

4 Answers 4

5

There are a few points you could consider:

  • Choice of Browser. Some time ago, Internet Explorer often caused this kind of problem.

  • FTP. If possible, you can try to get download from an FTP server via FTP (File Transfer Protocol) rather than HTTP (Hypertext Transfer Protocol). From the name you can see that FTP is dedicated to transferring files.

  • Command-line download utilities. Dedicated software is often better than a browser for such tasks.

    Windows: BITSAdmin is the first one I found, but surely there is more.

    Linux: wget, etc.

  • Internet connection. First of all, see if your internet is not breaking off every 10 minutes – while it does not affect web surfing, it can become a problem for downloading via some browsers (then command-line utilities could be a temporary solution). Also check if your cable is not damaged.

  • Antivirus. If your antivirus is checking the files you're downloading bit by bit, it could delay the process and cause problems. Not saying you should turn your antivirus off when downloading unknown files, but you can do some tests of this kind if everything else fails.

In general, without further info about the problem you experience, it is hard to advise something.

2

The bigger the file the higher the probability that there is an error in it resulting in the corruption.

The corruption could happen during the transfer or in the storage. Consider that magnetic hard drives have bit error rates (BER) of 1 in 10 to the power of 14. This translates to about one possible error for every 12TB stored. If you store enough large files, you may overcome this limit.

Similarly, the network transmissions encounter errors, especially over wireless networks. Most of these are either corrected or fixed by re-transmitting a chunk of data. However, even there remains space for errors.

2

Considering that TCP is lossless, I would be really surprised if the transfer of the data is causing a corruption. It's more likely that it's either the source of the file or something going on with your system.

4
  • 2
    packets are lossless, but the contents of the packets could get corrupted. TCP alone doesn't have an airtight way of ensuring content.
    – Zombies
    Jan 8, 2013 at 16:01
  • @Zombies - Its worth mentioning that while TCP doesn't have an "airtight" way of ensuring content, it does verify a checksum, and ask for packet retransmission when checksum does not match. May 8, 2018 at 16:55
  • 1
    @ToolmakerSteve See en.wikipedia.org/wiki/… " The TCP checksum is a weak check by modern standards. Data Link Layers with high bit error rates may require additional link error correction/detection capabilities."
    – Étienne
    Dec 7, 2020 at 15:14
  • @Étienne - thanks, I did not realize that. I see that it is only 16-bits. Dec 7, 2020 at 23:02
1

After download with the browser, always wait 5-10sec. before grab the file - there are file buffers to be flushed in the browser

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .