The ability to do HTTP requests over a UNIX domain socket has been
requested before, in Apr 2008 [0][1] and Sep 2010 [2]. While a
discussion happened, no patch seems to get through. I decided to give it
a go since I need to test a nginx HTTP server which listens on a UNIX
domain socket.
One patch [3] seems to make it possible to use the
CURLOPT_OPENSOCKETFUNCTION function to gain a UNIX domain socket.
Another person wrote a Go program which can do HTTP over a UNIX socket
for Docker[4] which uses a special URL scheme (though the name contains
cURL, it has no relation to the cURL library).
This patch considers support for UNIX domain sockets at the same level
as HTTP proxies / IPv6, it acts as an intermediate socket provider and
not as a separate protocol. Since this feature affects network
operations, a new feature flag was added ("unix-sockets") with a
corresponding CURL_VERSION_UNIX_SOCKETS macro.
A new CURLOPT_UNIX_SOCKET_PATH option is added and documented. This
option enables UNIX domain sockets support for all requests on the
handle (replacing IP sockets and skipping proxies).
A new configure option (--enable-unix-sockets) and CMake option
(ENABLE_UNIX_SOCKETS) can disable this optional feature. Note that I
deliberately did not mark this feature as advanced, this is a
feature/component that should easily be available.
[0]: http://curl.haxx.se/mail/lib-2008-04/0279.html
[1]: http://daniel.haxx.se/blog/2008/04/14/http-over-unix-domain-sockets/
[2]: http://sourceforge.net/p/curl/feature-requests/53/
[3]: http://curl.haxx.se/mail/lib-2008-04/0361.html
[4]: https://github.com/Soulou/curl-unix-socket
Signed-off-by: Peter Wu <peter@lekensteyn.nl>
Prior to this change when no file descriptors were ready on platforms
other than Windows the multi examples would sleep whatever was in
timeout, which may or may not have been less than the minimum
recommended value [1] of 100ms.
[1]: http://curl.haxx.se/libcurl/c/curl_multi_fdset.html
Windows does not support using select() for sleeping without a dummy
socket. Instead use Windows' Sleep() and sleep for 100ms which is the
minimum suggested value in the curl_multi_fdset() doc.
Prior to this change the multi examples would exit prematurely since
select() would error instead of sleeping when called without an fd.
Reported-by: Johan Lantz
Bug: http://curl.haxx.se/mail/lib-2014-11/0221.html
Reworked the two sections that discuss modifying the Visual Studio pre-
processor settings, and vc6libcurl.dsw/vc6libcurl.dsp, to remove the
project files references as they have been superseded by a more thorough
set of project files for VC6 through VC12, but to also give the correct
reference to this setting in later versions of Visual Studio.
docs/THANKS-filter is a new filter file for converting contributor names
we get or have recorded in alternative formats to the one we already use
in THANKS. To help us show individual contributors using a single
presentation of their names.
The removed names also appear as:
Andrés García, François Charlier, Gökhan Şengün, Michał Górny, Sébastien
Willemijns, Christopher Conroy, John E. Malmberg, Luca Altea, Peter Su,
S. Moonesamy, Samuel Listopad, Yasuharu Yamada, Karl Moerder
As implementations are refereed to GSS-API libraries as per the RFC and
GSSAPI typically refers to the SASL authentication mechanism.
...and minor rewording on the same paragraph.
Fixed the ability to use the current log-in credentials with DIGEST-MD5.
I had previously disabled this functionality in commit 607883f13c as I
couldn't get this to work under Windows 8, however, from testing HTTP
Digest authentication through Windows SSPI and then further testing of
this code I have found it works in Windows 7.
Some further investigation is required to see what the differences are
between Windows 7 and 8, but for now enable this functionality as the
code will return an error when AcquireCredentialsHandle() fails.
- Remove SSLv3 from SSL default in darwinssl, schannel, cyassl, nss,
openssl effectively making the default TLS 1.x. axTLS is not affected
since it supports only TLS, and gnutls is not affected since it already
defaults to TLS 1.x.
- Update CURLOPT_SSLVERSION doc
ENABLE_IPV6 depends on HAVE_GETADDRINFO or you will get a
Curl_getaddrinfo_ex error. Enable IPv6 by default, disabling it if
struct sockaddr_in6 is not found in netinet/in.h.
Note that HAVE_GETADDRINFO_THREADSAFE is still not set as it needs more
platform checks even though POSIX requires a thread-safe getaddrinfo.
Verified on Arch Linux x86_64 with glibc 2.20-2 and Linux 3.16-rc7.
Signed-off-by: Peter Wu <peter@lekensteyn.nl>
Rather than always outputting an empty manual page for the '-M' option,
generate a full manual page as done by autotools. For simplicity in
CMake, always generate the gzipped page as it will not be used anyway
when zlib is not available.
Signed-off-by: Peter Wu <peter@lekensteyn.nl>
Option --pinnedpubkey takes a path to a public key in DER format and
only connect if it matches (currently only implemented with OpenSSL).
Provides CURLOPT_PINNEDPUBLICKEY for curl_easy_setopt().
Extract a public RSA key from a website like so:
openssl s_client -connect google.com:443 2>&1 < /dev/null | \
sed -n '/-----BEGIN/,/-----END/p' | openssl x509 -noout -pubkey \
| openssl rsa -pubin -outform DER > google.com.der
For private keys, use the first match from: user-specified key file
(if provided), ~/.ssh/id_rsa, ~/.ssh/id_dsa, ./id_rsa, ./id_dsa
Note that the previous code only looked for id_dsa files. id_rsa is
now generally preferred, as it supports larger key sizes.
For public keys, use the user-specified key file, if provided.
Otherwise, try to extract the public key from the private key file.
This means that passing --pubkey is typically no longer required,
and makes the key-handling behavior more like OpenSSH.
As we weren't using the correct phonetic description and doing it correctly
involves funny letters that I'm sure will cause problems for people in a text
document so I instead rephrased it and link to a WAV file with a person
actually saying 'curl'.
Reported-By: Dimitar Boevski
Added support for Kerberos 5 to the email protocols following the recent
additions in 7.38.0.
Removed Kerberos 4 as this has been gone for a while now.
Only minor edits to make it generate nice HTML output using markdown, as
this document serves both in source release tarballs as on the web site.
URL: http://curl.haxx.se/docs/sslcerts.html
When enabling CURLOPT_NOBODY, libcurl effectively switches off upload
mode and will do a download (without a body). This is now better
explained in this man page.
Bug: http://curl.haxx.se/mail/lib-2014-08/0236.html
Reported-by: John Coffey
I re-ran contributors.sh on all changes since 7.10 and I found these
contributors who are mentioned in the commits but never were added to
THANKS before!
I also removed a couple of duplicates (mostly due to different
spellings).
Problem: if CURLOPT_FORBID_REUSE is set, requests using NTLM failed
since NTLM requires multiple requests that re-use the same connection
for the authentication to work
Solution: Ignore the forbid reuse flag in case the NTLM authentication
handshake is in progress, according to the NTLM state flag.
Fixed known bug #77.
Updated "FTP 4.6 GSSAPI via Windows SSPI" and "SASL 14.1 Other
authentication mechanisms" following recent additions.
Added SASL 14.2 GSSAPI via GSS-API libraries.
This repeats what has already been documented in both the curl manpage
and CURLOPT_USERPWD documentation but is provided here for completeness
as someone may not especially read the latter when using libcurl.
Added information about Kerberos V5 requiring the domain part in the
user name.
Mentioned that the user name can be specified in UPN format, and not
just in Down-Level Logon Name format, following the information
added in commit 7679cb3fa8 reworking the exisitng information in the
process.
1 - fixes the warnings when built without http2 support
2 - adds CURLE_HTTP2, a new error code for errors detected by nghttp2
basically when they are about http2 specific things.
This is just fundamentally broken. SPNEGO (RFC4178) is a protocol which
allows client and server to negotiate the underlying mechanism which will
actually be used to authenticate. This is *often* Kerberos, and can also
be NTLM and other things. And to complicate matters, there are various
different OIDs which can be used to specify the Kerberos mechanism too.
A SPNEGO exchange will identify *which* GSSAPI mechanism is being used,
and will exchange GSSAPI tokens which are appropriate for that mechanism.
But this SPNEGO implementation just strips the incoming SPNEGO packet
and extracts the token, if any. And completely discards the information
about *which* mechanism is being used. Then we *assume* it was Kerberos,
and feed the token into gss_init_sec_context() with the default
mechanism (GSS_S_NO_OID for the mech_type argument).
Furthermore... broken as this code is, it was never even *used* for input
tokens anyway, because higher layers of curl would just bail out if the
server actually said anything *back* to us in the negotiation. We assume
that we send a single token to the server, and it accepts it. If the server
wants to continue the exchange (as is required for NTLM and for SPNEGO
to do anything useful), then curl was broken anyway.
So the only bit which actually did anything was the bit in
Curl_output_negotiate(), which always generates an *initial* SPNEGO
token saying "Hey, I support only the Kerberos mechanism and this is its
token".
You could have done that by manually just prefixing the Kerberos token
with the appropriate bytes, if you weren't going to do any proper SPNEGO
handling. There's no need for the FBOpenSSL library at all.
The sane way to do SPNEGO is just to *ask* the GSSAPI library to do
SPNEGO. That's what the 'mech_type' argument to gss_init_sec_context()
is for. And then it should all Just Work™.
That 'sane way' will be added in a subsequent patch, as will bug fixes
for our failure to handle any exchange other than a single outbound
token to the server which results in immediate success.
With all the new individual option man pages created, this now refers to
each separate one instead of duplicaing the info. Also makes this page
easier to overview.
