Gather additional debug info when the test suite ssh server fails to start.

data sent in a post.

previously the second %s would wrongly get the numerical argument that is used
for the variable precision for the first %s...

are good/compatible enough to run the test suite ssh server and
socks tests

from the other day. It is time to setup the internal SSL libs and treat them
with a "handler" struct similar to how we deal with the protocols these days...

when sshd is not being identified as an OpenSSH daemon

updated external library versions to latest.

Improve wording of a couple of debug messages.

found or the OpenSSH version found

the lib was built staticly.

start so that subsequent tries are simply skipped.

callback was used, as it could wrongly pass on a bad size for the outgoing
HTTP header. The bad size would be a very large value as it was a wrapped
size_t content. This happened when the whole HTTP request failed to get sent
in one single send.  http://curl.haxx.se/mail/lib-2007-11/0165.html

#ifdef CURL_DOES_CONVERSIONS anyway! I turned it into a DEBUGASSERT() instead.

ipv6 server can't run.

do_init() and do_complete() which now are called first and last in the DO
function. It simplified the flow in multi.c and the functions got more
sensible names!

for the cases where there's nothing to do in here, like for SFTP directory
listings that already is complete when this function gets called. The init
stuff clears byte counters which isn't really desired.

When looking up in DNS and then in the hosts file, return the error code from DNS if both fail, instead of returning the error code from the hosts file, as today. Patch from the Google tree.

Return TTL data from ares_parse_{a,aaaa}_reply, if the user is so inclined. Patch from the Google tree.

been used for other protocols in between. I found this when test 2004 started
to fail for me!

huge send buffer sizes

persistent connection. Mentioned by Immanuel Gregoire on the mailing list.

forwarded from the Gentoo bug tracker by Daniel Black and was originally
submitted by Robin Johnson, pointed out that libcurl would do bad memory
references when it failed and bailed out before the handler thing was
setup. My fix is not done like the provided patch does it, but instead I
make sure that there's never any chance for a NULL pointer in that struct
member.

This also defines it for WinCE even though it is a subset of WIN32.