On Wed, 18 Sep 2013, Michele Mor wrote:

> On Windows (win 7) - using latest version of cUrl.

Latest being 7.32.0 I presume?

> I've been using curl in combination with a batch file to download files from
> a ftp site but sometimes the download fails.
>
> In those instances I revert to a manul process and use Filezilla. I have
> noticed that Filezilla is more reliable - the downloads are not cut off by
> the server - and quicker.

We would need lots of more details to be able to acknowledge that - the
reliability part I mean. The speed part we can probably help out with as we've
fixed some FTP related "slowness" in git and if you try the latest git code
you could help us verify that.

> I wonder if I can improve the reliability and performance of cUrl using some
> options that I am not aware of.

It depends on what the "reliability" part consists of. Is perhaps filezilla
retrying things on its on when it detects disconnected transfers in a way curl
doesn't?

> Note: I'm not too worried about donwload performance since it's an automatic
> process, but I am more concerned about interrupted downloads.

Give us some logs of such interrupts and maybe we can see what it is and what
we can do about it. Use --trace-ascii for example to get a good log, and edit
out sensitive info in there before you post anything here!

-- 
  / daniel.haxx.se
-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-users
FAQ:        http://curl.haxx.se/docs/faq.html
Etiquette:  http://curl.haxx.se/mail/etiquette.html