I need to transfer large files (in the 10+ Gigabyte range) across an unreliable network connection that spans large geographies. Remote users need to download these files to their local computers. The server that holds the files is running Linux, and the clients are Windows and Linux. I would like to offer some form of resume support for the file transfer.

Since the users have an SSH account on the Linux server, they also have SFTP access. Is it possible to get reliable resume support from an SFTP transfer so that my remote users can pick up where they left off when the connection drops out? I have heard that both client and server need to support a resume feature but have not found solid information on that topic yet.

  • Consider using CrushFTP which supports resume features and is very user friendly.
    – Wesley
    Dec 17, 2012 at 19:36
  • It would be much faster to encrypt the file (eg openssl enc bf -k SecurePassword < file.bin > file.bin.bfe; decrypt the same way with a -d option) then use "normal" means of transport like FTP, rsync, HTTP, etc.
    – Chris S
    Dec 17, 2012 at 19:37
  • If possible I would strongly look at rsync. It has probably an excellent resume protocol, and can operate over SSH.
    – Zoredache
    Dec 18, 2012 at 7:45
  • Rsync will work, plus it has lots of other features. unix.stackexchange.com/questions/2445/… Oct 6, 2015 at 12:24

3 Answers 3


Winscp supports resume over SCP - http://winscp.net/eng/docs/resume#automatic.

  • Sweet! That looks pretty good so far!
    – cat pants
    Dec 16, 2012 at 2:33

perhaps wget over ssh tunnel might work. i can't find anything about openssh supporting file transfter resume on sftp. Might want try to get them to tinker with the Keep Alive settings too.

hope this helps.

  • I can't find anything in the ssh documentation either. :/ That's why I posted it here, but I guess this question is "off topic"...do you think this question is off topic?
    – cat pants
    Dec 16, 2012 at 2:32

Both server and client need to support resuming. https://datatracker.ietf.org/doc/html/draft-ietf-secsh-filexfer-13#section-8.2.1 The standard talked about reading/writing files at some offsets. The offsets are how resuming works. A link to some guys working on a resuming issue. https://github.com/pkg/sftp/issues/295

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.