Commit graph

67 commits

Author SHA1 Message Date
Kurt Roeckx
b6ece4c1fc Keep old method in case of an unsupported protocol
When we're configured with no-ssl3 and we receive an SSL v3 Client Hello, we set
the method to NULL.  We didn't used to do that, and it breaks things.  This is a
regression introduced in 62f45cc27d.  Keep the old
method since the code is not able to deal with a NULL method at this time.

CVE-2014-3569, PR#3571

Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit 392fa7a952)
2014-10-21 21:04:09 +02:00
Geoff Thorpe
82180dcc0a Fix no-ssl3 configuration option
CVE-2014-3568

Reviewed-by: Emilia Kasper <emilia@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>
2014-10-15 08:54:26 -04:00
Bodo Moeller
a46c705214 Support TLS_FALLBACK_SCSV.
Reviewed-by: Rich Salz <rsalz@openssl.org>
2014-10-15 04:04:55 +02:00
David Benjamin
40a2200d89 Fix protocol downgrade bug in case of fragmented packets
CVE-2014-3511

Reviewed-by: Emilia Käsper <emilia@openssl.org>
Reviewed-by: Bodo Möller <bodo@openssl.org>
2014-08-06 20:41:24 +01:00
Dr. Stephen Henson
ccf6a19e2d Add three Suite B modes to TLS code, supporting RFC6460.
(backport from HEAD)
2012-12-26 16:17:40 +00:00
Dr. Stephen Henson
7043fa702f add FIPS support to ssl: doesn't do anything on this branch yet as there is no FIPS compilation support 2011-05-19 18:22:16 +00:00
Dr. Stephen Henson
9472baae0d Backport TLS v1.2 support from HEAD.
This includes TLS v1.2 server and client support but at present
client certificate support is not implemented.
2011-05-11 13:37:52 +00:00
Dr. Stephen Henson
b4b15f68c0 Backport TLS v1.1 support from HEAD, ssl/ changes 2010-06-27 14:22:11 +00:00
Dr. Stephen Henson
45d6a15ae9 PR: 2171
Submitted by: Tomas Mraz <tmraz@redhat.com>

Since SSLv2 doesn't support renegotiation at all don't reject it if
legacy renegotiation isn't enabled.

Also can now use SSL2 compatible client hello because RFC5746 supports it.
2010-02-16 14:20:40 +00:00
Dr. Stephen Henson
41c0f68630 Fix version handling so it can cope with a major version >3.
Although it will be many years before TLS v2.0 or later appears old versions
of servers have a habit of hanging around for a considerable time so best
if we handle this properly now.
2010-01-13 19:08:29 +00:00
Dr. Stephen Henson
3e8e12a6b6 Servers can't end up talking SSLv2 with legacy renegotiation disabled 2009-11-18 15:09:35 +00:00
Ben Laurie
8671b89860 Memory saving patch. 2008-06-03 02:48:34 +00:00
Bodo Möller
1e24b3a09e fix support for receiving fragmented handshake messages 2006-11-29 14:45:50 +00:00
Bodo Möller
f1fd4544a3 Various changes in the new TLS extension code, including the following:
- fix indentation
 - rename some functions and macros
 - fix up confusion between SSL_ERROR_... and SSL_AD_... values
2006-01-03 03:27:19 +00:00
Bodo Möller
ed3883d21b Support TLS extensions (specifically, HostName)
Submitted by: Peter Sylvester
2006-01-02 23:14:37 +00:00
Dr. Stephen Henson
7bbcb2f690 Avoid warnings on VC++ 2005. 2005-12-05 17:21:22 +00:00
Bodo Möller
72dce7685e Add fixes for CAN-2005-2969.
(This were in 0.9.7-stable and 0.9.8-stable, but not in HEAD so far.)
2005-10-26 19:40:45 +00:00
Nils Larsch
4ebb342fcd Let the TLSv1_method() etc. functions return a const SSL_METHOD
pointer and make the SSL_METHOD parameter in SSL_CTX_new,
SSL_CTX_set_ssl_version and SSL_set_ssl_method const.
2005-08-14 21:48:33 +00:00
Dr. Stephen Henson
f3b656b246 Initialize SSL_METHOD structures at compile time. This removes the need
for locking code. The CRYPTO_LOCK_SSL_METHOD lock is now no longer used.
2005-08-05 23:56:11 +00:00
Ben Laurie
54a656ef08 Security fixes brought forward from 0.9.7. 2002-11-13 15:43:43 +00:00
Bodo Möller
b8565a9af9 really fix race conditions
Submitted by: "Patrick McCormick" <patrick@tellme.com>

PR: 262
PR: 291
2002-09-25 15:38:57 +00:00
Bodo Möller
e78f137899 really fix race condition
PR: 262
2002-09-23 14:25:07 +00:00
Lutz Jänicke
7b63c0fa8c Reorder inclusion of header files:
des_old.h redefines crypt:
#define crypt(b,s)\
        DES_crypt((b),(s))

This scheme leads to failure, if header files with the OS's true definition
of crypt() are processed _after_ des_old.h was processed. This is e.g. the
case on HP-UX with unistd.h.
As evp.h now again includes des.h (which includes des_old.h), this problem
only came up after this modification.
Solution: move header files (indirectly) including e_os.h before the header
files (indirectly) including evp.h.
Submitted by:
Reviewed by:
PR:
2002-07-10 07:01:54 +00:00
Bodo Möller
2fb3f002d0 fix length field we create when converting SSL 2.0 format into SSL 3.0/TLS 1.0 format
(the bug was introduced with message callback support)
2002-04-14 13:05:15 +00:00
Ben Laurie
45d87a1ffe Prototype info function. 2002-01-12 15:56:13 +00:00
Bodo Möller
cf82191d77 Implement msg_callback for SSL 2.0.
Important SSL 2.0 bugfixes (bugs found while implementing msg_callback).
2001-11-10 01:16:28 +00:00
Bodo Möller
ba1c602281 Assume TLS 1.0 when ClientHello fragment is too short. 2001-10-25 06:09:51 +00:00
Bodo Möller
979689aa5c Fix SSL handshake functions and SSL_clear() such that SSL_clear()
never resets s->method to s->ctx->method when called from within
one of the SSL handshake functions.
2001-10-24 19:03:22 +00:00
Bodo Möller
a661b65357 New functions SSL[_CTX]_set_msg_callback().
New macros SSL[_CTX]_set_msg_callback_arg().

Message callback imlementation for SSL 3.0/TLS 1.0 (no SSL 2.0 yet).

New '-msg' option for 'openssl s_client' and 'openssl s_server'
that enable a message callback that displays all protocol messages.


In ssl3_get_client_hello (ssl/s3_srvr.c), generate a fatal alert if
client_version is smaller than the protocol version in use.
Also change ssl23_get_client_hello (ssl/s23_srvr.c) to select TLS 1.0
if the client demanded SSL 3.0 but only TLS 1.0 is enabled; then the
client will at least see that alert.

Fix SSL[_CTX]_ctrl prototype (void * instead of char * for generic
pointer).

Add/update some OpenSSL copyright notices.
2001-10-20 17:56:36 +00:00
Bodo Möller
9ba3ec9176 The message header for fake SSL 3.0/TLS 1.0 client hellos created from
SSL 2.0 client hellos added with the previous commit was totally wrong --
it must start with the message type, not the protocol version.
(Not that this particular header is actually used anywhere ...)
2001-10-16 00:56:04 +00:00
Bodo Möller
48948d53b6 Change ssl3_get_message and the functions using it so that complete
'Handshake' protocol structures are kept in memory, including
'msg_type' and 'length'.

(This is in preparation of future support for callbacks that get to
peek at handshake messages and the like.)
2001-10-15 19:49:25 +00:00
Lutz Jänicke
d92f0bb6e9 Remove SSL_OP_NON_EXPORT_FIRST:
It did not work, it was deactivated by #if 0/#endif anyway _and_ we now have
the working SSL_OP_CIPHER_SERVER_PREFERENCE.
2001-08-01 10:06:32 +00:00
Richard Levitte
bc36ee6227 Use new-style system-id macros everywhere possible. I hope I haven't
missed any.

This compiles and runs on Linux, and external applications have no
problems with it.  The definite test will be to build this on VMS.
2001-02-20 08:13:47 +00:00
Bodo Möller
f1192b7f2e Avoid protocol rollback. 2000-09-22 21:39:33 +00:00
Bodo Möller
aa826d88e1 Document rollback issues. 2000-07-29 19:27:20 +00:00
Bodo Möller
37569e64e8 Fix SSL 2.0 rollback checking: The previous implementation of the
test was never triggered due to an off-by-one error.

In s23_clnt.c, don't use special rollback-attack detection padding
(RSA_SSLV23_PADDING) if SSL 2.0 is the only protocol enabled in the
client; similarly, in s23_srvr.c, don't do the rollback check if
SSL 2.0 is the only protocol enabled in the server.
2000-07-29 18:50:41 +00:00
Richard Levitte
26a3a48d65 There have been a number of complaints from a number of sources that names
like Malloc, Realloc and especially Free conflict with already existing names
on some operating systems or other packages.  That is reason enough to change
the names of the OpenSSL memory allocation macros to something that has a
better chance of being unique, like prepending them with OPENSSL_.

This change includes all the name changes needed throughout all C files.
2000-06-01 22:19:21 +00:00
Ulf Möller
4c5fac4ac4 Fix NO_RSA (misplaced #endif). 2000-02-24 04:41:03 +00:00
Dr. Stephen Henson
72b60351f1 Change EVP_MD_CTX_type so it is more logical and add EVP_MD_CTX_md for
the old functionality.

Various warning fixes.

Initial EVP symmetric cipher docs.
2000-02-22 02:59:26 +00:00
Bodo Möller
dab6f09573 Workaround for irrelevant problem. 2000-02-20 23:40:01 +00:00
Bodo Möller
b35e9050f2 Tolerate fragmentation and interleaving in the SSL 3/TLS record layer. 2000-02-20 23:04:06 +00:00
Ulf Möller
657e60fa00 ispell (and minor modifications) 2000-02-03 23:23:24 +00:00
Bodo Möller
cef80e8c14 SSL_R_UNSUPPORTED_PROTOCOL (as in s23_clnt.c) for SSL 2 when
NO_SSL2 is defined, not SSL_R_UNKNOWN_PROTOCOL.
2000-01-16 21:29:57 +00:00
Ulf Möller
aa82db4fb4 Add missing #ifndefs that caused missing symbols when building libssl
as a shared library without RSA.  Use #ifndef NO_SSL2 instead of
NO_RSA in ssl/s2*.c.

Submitted by: Kris Kennaway <kris@hub.freebsd.org>
Modified by Ulf Möller
2000-01-16 21:10:00 +00:00
Ulf Möller
eb952088f0 Precautions against using the PRNG uninitialized: RAND_bytes() now
returns int (1 = ok, 0 = not seeded). New function RAND_add() is the
same as RAND_seed() but takes an estimate of the entropy as an additional
argument.
2000-01-13 20:59:17 +00:00
Bodo Möller
c51ae173a6 Clean up some of the SSL server code. 2000-01-11 01:07:26 +00:00
Bodo Möller
ac7da00048 Set s->version correctly for "natural" SSL 3.0 client hello 1999-09-13 13:02:07 +00:00
Bodo Möller
cb0369d885 Repair another bug in s23_get_client_hello:
tls1 did not survive to restarts, so get rid of it.
1999-09-10 16:41:01 +00:00
Bodo Möller
ba3a6e7262 use explicit constant 11 just once 1999-09-03 22:37:38 +00:00
Bodo Möller
f70df1b887 Make previous bugfix actually work 1999-09-03 16:49:11 +00:00