SNI needs to be consistent before we accept early_data. However a
server may choose to not acknowledge SNI. In that case we have to
expect that a client may send it anyway. We change the consistency
checks so that not acknowledging is treated more a like a "wild card",
accepting any SNI as being consistent.
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/4738)
Test for the bug where early_data is not accepted by the server when it
does not have an SNI callback set up, but the client sent a servername in
the initial ClientHello establishing the session.
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/4519)
It is otherwise unclear what all the magic numbers mean.
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/4349)
"Early callback" is a little ambiguous now that early data exists.
Perhaps "ClientHello callback"?
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/4349)
A condition was removed by commit 1053a6e2281d; presumably it was an
unintended change. Restore the previous behavior so the get_session_cb
won't be called with zero-length session ID.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/4236)
Force non-empty padding extension.
When enabled, force the padding extension to be at least 1 byte long.
WebSphere application server cannot handle having an empty
extension (e.g. EMS/EtM) as the last extension in a client hello.
This moves the SigAlgs extension last for TLSv1.2 to avoid this
issue.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3921)
Apart from ssltest_old.c, the test suite relied on e_os.h for the
OSSL_NELEM macro and nothing else.
The ssltest_old.c also requires EXIT and some socket macros.
Create a new header to define the OSSL_NELEM macro and use that instead.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4186)
When we are using the internal cache we have to make a copy of the
session before removing it from the parent context's cache, since
we want our copy to still be resumable. However, SSL_CTX_remove_session()
just detaches the session from the SSL_CTX; it does not free the session.
So, we must call SSL_SESSION_free() ourselves before overwriting the
variable that we dup'd from.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4126)
Removing the use of SETUP_TEST_FIXTURE reduces complxity in those tests that
used it.
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Andy Polyakov <appro@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4066)
In TLSv1.3 we can resume, but still get a new session. This adds a test to
make sure that is happening.
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/4068)
that needed test_main now works using the same infrastructure as tests that used
register_tests.
This meant:
* renaming register_tests to setup_tests and giving it a success/failure return.
* renaming the init_test function to setup_test_framework.
* renaming the finish_test function to pulldown_test_framework.
* adding a user provided global_init function that runs before the test frame
work is initialised. It returns a failure indication that stops the stest.
* adding helper functions that permit tests to access their command line args.
* spliting the BIO initialisation and finalisation out from the test setup and
teardown.
* hiding some of the now test internal functions.
* fix the comments in testutil.h
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3953)
Our test was using 32. The latest ticket nonce changes now validate this
value and so sslapitest was failing.
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3852)
There aren't any test vectors for this, so all we do is test that both
sides of the communication create the same result for different protocol
versions.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3782)
Make it clear that we are pausing one of the connections and then
restarting it again.
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3623)
This does things as per the recommendation in the TLSv1.3 spec. It also
means that the server will always choose its preferred ciphersuite.
Previously the server would only select ciphersuites compatible with the
session.
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3623)
Test that if a server selects a differenct ciphersuite with the same hash
in TLSv1.3 then this is accepted by the client.
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3623)
Call it from the early callback used for testing these functions, and verify
the expected contents of the ClientHello
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/2976)
- Mostly missing fall thru comments
- And uninitialized value used in sslapitest.c
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3440)
The previous commit fixed a bug which occurs when serverinfo is loaded
from memory (not from a file). This adds a test for loading serverinfo
from memory.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3382)
Make sure the server can write normal data after earlier writing early data.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3091)
Commit 9b5c865df introduced a synthetic delay between arrival of EoED and
CF. We actually want to delay the arrival of CF even further to demonstrate
that we can write early data even when "in init".
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3091)
This tests the bug fixed in the previous commit. We introduce a synthetic
delay between the server receiving EoED and CF and check that we can still
send early data.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3089)
We want to make sure that if we if are using SSL_MODE_AUTO_RETRY then
if SSL_read_early_data() hits EndOfEarlyData then it doesn't auto retry
and end up with normal data. The same issue could occur with read_ahead
which is what we use in this test.
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3077)
Fix some comments too
[skip ci]
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3069)
The end of early data is now indicated by a new handshake message rather
than an alert.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/2895)
Change the early data API so that the server must use
SSL_write_early_data() to write to an unauthenticated client.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/2737)
This is for consistency with the rest of the API where all the functions
are called *early_data*.
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/2737)
