Tobias Brunner [Fri, 14 Mar 2014 16:33:22 +0000 (17:33 +0100)]
openssl: Add default fallback when calculating fingerprints of RSA keys
We still try to calculate these directly as it can avoid a dependency on
the pkcs1 or other plugins. But for e.g. PGPv3 keys we need to delegate the
actual fingerprint calculation to the pgp plugin.
Tobias Brunner [Thu, 20 Mar 2014 17:49:03 +0000 (18:49 +0100)]
Merge branch 'travis-ci'
Adds a config file and build script for Travis CI. Makes the unit tests
buildable with Clang, and test vectors are now actually verified when
the unit tests are executed.
Also adds options to run only selected test suites and to increase the debug
level during unit tests.
The --enable/disable configure options have been reordered and grouped, and
an option to enable all the features has been added (plus an option to
select a specific printf-hook implementation).
Tobias Brunner [Tue, 18 Mar 2014 14:25:56 +0000 (15:25 +0100)]
travis: Use parallel build
Not sure if 4 jobs is optimal, but according to the docs each build host
has 1.5 virtual cores available (although "getconf _NPROCESSORS_ONLN"
returns 32, which is probably the number of real cores underneath), so
more jobs might not actually reduce the build time much more.
Tobias Brunner [Fri, 14 Mar 2014 08:56:23 +0000 (09:56 +0100)]
crypto-tester: Don't fail if key size is not supported
The Blowfish and Twofish implementations provided by the gcrypt plugin
only support specific key lengths, which we don't know when testing
against vectors (either during unit tests or during algorithm
registration). The on_create test with a specific key length will be
skipped anyway, so there is no point in treating this failure differently.
Tobias Brunner [Thu, 13 Mar 2014 15:03:05 +0000 (16:03 +0100)]
unit-tests: Actually verify registered algorithms against test vectors
Previously, the {ns}.crypto_test.on_add option had to be enabled to
actually test the algorithms, which we can't enforce for the tests in
the test_runner as the option is already read when the crypto factory
is initialized. Even so, we wouldn't want to do this for every unit
test, which would be the result of enabling that option.
Tobias Brunner [Fri, 28 Feb 2014 16:08:39 +0000 (17:08 +0100)]
unit-tests: Change how hashtable for testable functions is created
Because GCC does not adhere to the priorities defined for constructors
when building with --enable-monolithic (not sure if it was just luck
that it worked in non-monolithic mode - anyway, it's not very portable)
function registration would fail because the hashtable would not be
created yet.
Tobias Brunner [Thu, 20 Mar 2014 12:42:57 +0000 (13:42 +0100)]
plugin-feature: Hash only the actually used feature argument
Clang does not initialize padding in union members so hashing the
complete "arg" union could lead to different hashes if the hashed
plugin_feature_t does not have static storage duration.
A valgrind analysis of libstrongswan revealed an invalid read of 1 in
the function starts_with(). A more thorough analysis proved this to be
true and showed that with a specially crafted config file (e.g. a single
'#'-character not followed by a newline), the parser might even
interpret the random memory contents following the allocated buffer as
part of the configuration file.
The way the parser is designed, it must be able to skip an inserted
'\0' and continue parsing. Since it is not able to skip two '\0'
characters, the 'fix' of allocating two more bytes than the size of the
parsed file and setting them to '\0' seems to be a safe bet.
Signed-off-by: Thomas Egerer <thomas.egerer@secunet.com>
Tobias Brunner [Mon, 3 Mar 2014 16:14:26 +0000 (17:14 +0100)]
configure: Fix autoreconf with older autotools
Older autoconf versions (e.g. on CentOS 6.5) produce an empty else block
for the removed empty argument, which the shell then trips over when
executing ./configure.
Tobias Brunner [Thu, 27 Feb 2014 10:40:32 +0000 (11:40 +0100)]
conf: Fix out-of-tree build from distribution
It worked from the repository, where strongswan.conf.5.main is generated
in the build dir, but not from the distribution where it is located in
the source dir, so explicitly create it in the source dir.
Tobias Brunner [Fri, 21 Feb 2014 13:58:01 +0000 (14:58 +0100)]
settings: Use dirname(3) correctly
dirname(3) may return a pointer to a statically allocated buffer.
So freeing the returned value can result to undefined behavior. This was
noticed on FreeBSD where it caused very strange crashes.
It is also not thread-safe, which will be addressed later.
Martin Willi [Mon, 17 Feb 2014 08:44:14 +0000 (09:44 +0100)]
lookip: Disconnect asynchronously to avoid dead-locking watcher unregistration
While it really would be desirable to allow stream destruction during on_read()
callbacks, this does not work anymore since e49b2998. Until we have a proper
solution for this issue, use asynchronous disconnects for the only user doing
so.
Andreas Steffen [Fri, 14 Feb 2014 14:06:57 +0000 (15:06 +0100)]
Fixed a minor vulnerability in which a malformed ASN.1 length field could cause a crash of the charon daemon if the verbose debug level 3 (raw hex dump) for the asn subsystem is enabled.
Tobias Brunner [Wed, 12 Feb 2014 14:24:47 +0000 (15:24 +0100)]
Merge branch 'modular-load'
Introduces a new configuration file layout. strongswan.conf is now only
very simple and mainly includes the config snippets from the strongswan.d
and strongswan.d/charon directories (the latter containing snippets for
individual plugins).
Config snippets with commented defaults are generated for all currently
defined settings and are installed if they don't exist yet and the
respective plugin/component is enabled. Similarly, the strongswan.conf(5)
man page, which documents all these settings, is automatically generated
from the same source.
The config snippets are also installed in $prefix/share/strongswan so
existing files can be compared to the most current defaults.
As an alternative to the non-extensible charon.load option, the plugins
to load can now be determined via the respective charon.plugins.<name>.load
setting. This functionality is enabled by the new default strongswan.conf
file (via the charon.load_modular option) and the load setting in the
generated config snippets of all enabled plugins. The load setting
optionally takes a numeric priority value that allows reordering the
plugins (plugins with the same priority are ordered according to the
default plugin order).
Additionally, all settings that were formerly defined in library
specific "global" sections are now application specific. For instance,
instead of configuring libstrongswan.plugins.random.random and affecting
charon, charon-cmd, pki, basically every application using libstrongswan,
the option can now be set individually for each application (e.g.
pki.plugins.random.random to affect only pki). The old options are still
supported though, which actually allows to define defaults for all
applications in the libstrongswan section.
The libtls options are mapped to <app>.tls. The libimcv and libtnccs options
are mapped to <app>.imcv and <app>.tnc, respectively (while their plugin's
options are now under <app>.plugins together with all the others).