On Fri, 4 Jan 2008, autrelandes-curl_at_yahoo.fr wrote:

> Implementors MAY include "keep-alives" in their TCP
> implementations, although this practice is not universally
> accepted. If keep-alives are included, the application MUST
> be able to turn them on or off for each TCP connection, and
> they MUST default to off.

> It may be better to use a --keep-alive option rather than a --no-keep-alive
> option.

I disagree.

RFC1122 is from October 1989, today all modern operating systems support the
keep-alive (as was already discussed before on the libcurl list I believe).

Besides that, the source code in curl only requires that the API for dealing
with the keep-alives work not that the actual implementation to deal with them
is there! So, if keep-alive doesn't actually exist in the TCP stack that curl
runs on, nothing needs to happen...

(Off-topic, but man is those subjects of your mails annying! Why include names
in the subjects? The mails not only have no proper attribution to what thread
they belong to, but it invents new subjects which also makes them impossible
to group with others based on that...)

-- 
  Commercial curl and libcurl Technical Support: http://haxx.se/curl.html