59

Is there a maximum size for an HTTP POST? And if there is a max size, is it determined by the protocol or is it at the discretion of the server?

4 Answers4

48

The HTTP specification doesn't impose a specific size limit for posts. They will usually be limited by either the web server or the programming technology used to process the form submission.

Justin Scott
  • 8,748
  • 1
  • 27
  • 39
18

No limit by specification. The limit is MIN(browser_limit,server_limit).

lrosa
  • 1,657
  • 14
  • 15
9

POST method does not have any limit on the size of data. But Server & Browsers has.

For Example

IE: 2GB
Firefox: 2GB
Chrome: 4 GB
Opera: 4 GB

For More Details check below URL
https://www.motobit.com/help/scptutl/pa98.htm

MukeshKoshyM
  • 191
  • 1
  • 2
2

I don't think that there is a specific limit in the protocol, BUT there are cases where things get squirrelly if you're transferring huge files. It depends on the size of the file as to what may work a little better for your specific case. There's not any support for checking that the file wasn't corrupted in transfer, for example, unlike using something like a torrent for the transfer.

Bart Silverstrim
  • 31,092
  • 9
  • 65
  • 87
  • 4
    If TCP does its job, the file should be intact. – Justin Scott Jun 14 '10 at 17:24
  • 3
    If TCP does its job, we wouldn't need md5 sums to verify ISO images and other large files :-) – Bart Silverstrim Jun 14 '10 at 18:08
  • 9
    The md5 sums are used to check if source file matches the publisher content file to avoid fakes. TCP does have a package check and chance of error pass without detections is minimum. – cavila Aug 20 '14 at 19:22
  • 1
    Note: md5 is not collision resistant anymore. Don't trust md5sums to verify the creator of a file. – Florian Wendelborn Aug 30 '16 at 17:31
  • 1
    @cavila: The TCP check is a 32 bits checksum, so we can quantify that chance: it's 1 in 2^32 _per packet_. Of course, most packets aren't faulty to begin with. – MSalters Apr 06 '17 at 08:33