Michael Kaufmann [Wed, 14 Nov 2018 10:52:45 +0000 (11:52 +0100)]
openssl: support session resume with TLS 1.3
Session resumption information is not available immediately after a TLS 1.3
handshake. The client must wait until the server has sent a session ticket.
Use OpenSSL's "new session" callback to get the session information and put it
into curl's session cache. For TLS 1.3 sessions, this callback will be invoked
after the server has sent a session ticket.
The "new session" callback is invoked only if OpenSSL's session cache is
enabled, so enable it and use the "external storage" mode which lets curl manage
the contents of the session cache.
A pointer to the connection data and the sockindex are now saved as "SSL extra
data" to make them available to the callback.
This approach also works for old SSL/TLS versions and old OpenSSL versions.
Reviewed-by: Daniel Stenberg <daniel@haxx.se>
Fixes #3202
Closes #3271
Romain Fliedel [Mon, 5 Nov 2018 10:01:19 +0000 (11:01 +0100)]
ares: remove fd from multi fd set when ares is about to close the fd
When using c-ares for asyn dns, the dns socket fd was silently closed
by c-ares without curl being aware. curl would then 'realize' the fd
has been removed at next call of Curl_resolver_getsock, and only then
notify the CURLMOPT_SOCKETFUNCTION to remove fd from its poll set with
CURL_POLL_REMOVE. At this point the fd is already closed.
By using ares socket state callback (ARES_OPT_SOCK_STATE_CB), this
patch allows curl to be notified that the fd is not longer needed
for neither for write nor read. At this point by calling
Curl_multi_closed we are able to notify multi with CURL_POLL_REMOVE
before the fd is actually closed by ares.
In asyn-ares.c Curl_resolver_duphandle we can't use ares_dup anymore
since it does not allow passing a different sock_state_cb_data
Remove the redundant USE_OPENSSL along with #else (it's not possible to
reach it anyway). The removed construction is a leftover from when the
SSLeay support was removed.
Closes #3269 Reviewed-by: Daniel Gustafsson <daniel@yesql.se> Reviewed-by: Daniel Stenberg <daniel@haxx.se>
The partial struct initialization in 397664a065abffb7c3445ca9 caused
a warning on uninitialized MODULEENTRY32 struct members:
/src/tool_doswin.c:681:3: warning: missing initializer for field
'th32ModuleID' of 'MODULEENTRY32 {aka struct tagMODULEENTRY32}'
[-Wmissing-field-initializers]
This is sort of a bogus warning as the remaining members will be set
to zero by the compiler, as all omitted members are. Nevertheless,
remove the warning by omitting all members and setting the dwSize
members explicitly.
Closes #3254 Reviewed-by: Marcel Raad <Marcel.Raad@teamviewer.com> Reviewed-by: Jay Satiro <raysatiro@yahoo.com>
Commit 709cf76f6bb7dbac deprecated USE_SSLEAY, as curl since long isn't
compatible with the SSLeay library. This removes the few leftovers that
were omitted in the less frequently used platform targets.
Closes #3270 Reviewed-by: Daniel Stenberg <daniel@haxx.se>
Peter Wu [Fri, 16 Nov 2018 16:57:08 +0000 (17:57 +0100)]
openssl: do not log excess "TLS app data" lines for TLS 1.3
The SSL_CTX_set_msg_callback callback is not just called for the
Handshake or Alert protocols, but also for the raw record header
(SSL3_RT_HEADER) and the decrypted inner record type
(SSL3_RT_INNER_CONTENT_TYPE). Be sure to ignore the latter to avoid
excess debug spam when using `curl -v` against a TLSv1.3-enabled server:
* TLSv1.3 (IN), TLS app data, [no content] (0):
(Following this message, another callback for the decrypted
handshake/alert messages will be be present anyway.)
tool_cb_wrt: Silence function cast compiler warning
Commit 5bfaa86ceb3c2a9ac474a928e748c4a86a703b33 introduced a new
compiler warning on Windows cross compilation with GCC. See below
for an example of the warning from the autobuild logs (whitespace
edited to fit):
/src/tool_cb_wrt.c:175:9: warning: cast from function call of type
'intptr_t {aka long long int}' to non-matching type 'void *'
[-Wbad-function-cast]
(HANDLE) _get_osfhandle(fileno(outs->stream)),
^
Store the return value from _get_osfhandle() in an intermediate
variable and cast the variable in WriteConsoleW() rather than the
function call directly to avoid a compiler warning.
In passing, also add inspection of the MultiByteToWideChar() return
value and return failure in case an error is reported.
Closes #3263 Reviewed-by: Marcel Raad <Marcel.Raad@teamviewer.com> Reviewed-by: Viktor Szakats <commit@vszakats.net>
Jérémy Rocher [Fri, 9 Nov 2018 13:05:26 +0000 (14:05 +0100)]
openssl: support BoringSSL TLS renegotiation
As per BoringSSL porting documentation [1], BoringSSL rejects peer
renegotiations by default.
curl fails when trying to authenticate to server through client
certificate if it is requested by server after the initial TLS
handshake.
Enable renegotiation by default with BoringSSL to get same behavior as
with OpenSSL. This is done by calling SSL_set_renegotiate_mode [2]
which was introduced in commit 1d5ef3bb1eb9 [3].
Daniel Stenberg [Fri, 9 Nov 2018 12:05:26 +0000 (13:05 +0100)]
HISTORY: add some milestones
Added a few of the more notable milestones in curl history that were
missing. Primarily more recent ones but I also noted some older that
could be worth mentioning.
Daniel Stenberg [Mon, 5 Nov 2018 10:57:29 +0000 (11:57 +0100)]
curl: fix --local-port integer overflow
The tool's local port command line range parser didn't check for integer
overflows and could pass "weird" data to libcurl for this option.
libcurl however, has a strict range check for the values so it rejects
anything outside of the accepted range.
Daniel Stenberg [Sun, 4 Nov 2018 22:30:48 +0000 (23:30 +0100)]
url: a short host name + port is not a scheme
The function identifying a leading "scheme" part of the URL considered a
few letters ending with a colon to be a scheme, making something like
"short:80" to become an unknown scheme instead of a short host name and
a port number.
Extended test 1560 to verify.
Also fixed test203 to use file_pwd to make it get the correct path on
windows. Removed test 2070 since it was a duplicate of 203.
Jay Satiro [Thu, 1 Nov 2018 06:50:40 +0000 (02:50 -0400)]
tool: add undocumented option --dump-module-paths for win32
- Add an undocumented diagnostic option for Windows to show the full
paths of all loaded modules regardless of whether or not libcurl
initialization succeeds.
This is needed so that in the CI we can get a list of all DLL
dependencies after initialization (when they're most likely to have
finished loading) and then package them as artifacts so that a
functioning build can be downloaded. Also I imagine it may have some use
as a diagnostic for help requests.
Jay Satiro [Thu, 1 Nov 2018 06:53:22 +0000 (02:53 -0400)]
curl_multibyte: fix a malloc overcalculation
Prior to this change twice as many bytes as necessary were malloc'd when
converting wchar to UTF8. To allay confusion in the future I also
changed the variable name for the amount of bytes from len to bytes.
netrc: don't ignore the login name specified with "--user"
- for "--netrc", don't ignore the login/password specified with "--user",
only ignore the login/password in the URL.
This restores the netrc behaviour of curl 7.61.1 and earlier.
- fix the documentation of CURL_NETRC_REQUIRED
- improve the detection of login/password changes when reading .netrc
- don't read .netrc if both login and password are already set
The previous coding used a format string whose output depended on the
current locale of the environment running the test. Since the gist of
the test is to have a format string, with the actual formatting being
less important, switch to a more stable formatstring with decimals.
The internal buffer in infof() is limited to 2048 bytes of payload plus
an additional byte for NULL termination. Servers with very long error
messages can however cause truncation of the string, which currently
isn't very clear, and leads to badly formatted output.
This appends a "...\n" (or just "..." in case the format didn't with a
newline char) marker to the end of the string to clearly show
that it has been truncated.
Also include a unittest covering infof() to try and catch any bugs
introduced in this quite important function.
Closes #3216 Reviewed-by: Daniel Stenberg <daniel@haxx.se> Reviewed-by: Marcel Raad <Marcel.Raad@teamviewer.com>
Daniel Stenberg [Fri, 2 Nov 2018 22:46:01 +0000 (23:46 +0100)]
url: a short host name + port is not a scheme
The function identifying a leading "scheme" part of the URL considered a few
letters ending with a colon to be a scheme, making something like "short:80"
to become an unknown scheme instead of a short host name and a port number.
Daniel Stenberg [Thu, 1 Nov 2018 22:45:57 +0000 (23:45 +0100)]
Curl_follow: accept non-supported schemes for "fake" redirects
When not actually following the redirect and the target URL is only
stored for later retrieval, curl always accepted "non-supported"
schemes. This was a regression from 46e164069d1a5230.
Commit efc696a2e09225bfeab4 renamed persistant.c to persistent.c to
fix the typo in the name, but missed to update the OpenVMS package
files which still looked for the old name.
Closes #3217 Reviewed-by: Daniel Stenberg <daniel@haxx.se> Reviewed-by: Viktor Szakats <commit@vszakats.net>
MesaLink support was added in commit 57348eb97d1b8fc3742e02c but the
backend was never added to the curl_sslbackend enum in curl/curl.h.
This adds the new backend to the enum and updates the relevant docs.
Closes #3195 Reviewed-by: Daniel Stenberg <daniel@haxx.se>
randomswdev [Sat, 27 Oct 2018 13:28:59 +0000 (15:28 +0200)]
system.h: use proper setting with Sun C++ as well
system.h selects the proper Sun settings when __SUNPRO_C is defined. The
Sun compiler does not define it when compiling C++ files. I'm adding a
check also on __SUNPRO_CC to allow curl to work properly also when used
in a C++ project on Sun Solaris.
Ensure to clear the session object in case the libssh2 initialization
fails.
It could be argued that the libssh2 error function should be called to
get a proper error message in this case. But since the only error path
in libssh2_knownhost_init() is memory a allocation failure it's safest
to avoid since the libssh2 error handling allocates memory.
Closes #3179 Reviewed-by: Daniel Stenberg <daniel@haxx.se>