The version 0.4.0 fixed the issues of multi-digit version numbers
which we hit with releaseing libssh ABI version 4_10 with last
release.
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Pavol Žáčik <pzacik@redhat.com>
When we use empty configuration file, some stuff go south in c10s
and for example fips mode detection does not work anymore.
Providing minimal configuration file avoids the issues of loading
the provider too early, while keeping fips mode activation working
and tests happy.
It also configures the pkcs11-provider to assume the token provides
FIPS approved crypto so the tests can work.
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Pavol Žáčik <pzacik@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
The maximal lenght of unix domain socket path is 108 characters. When
the build directory (and UID wrapper home directories) are too deep
in the filesystem, OpenSSH will fail to create the socket file,
which is failing this test.
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Pavol Žáčik <pzacik@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
without explicitly setting the algorithms, they might be set by
some other configuration file, for example crypto policies pulled
from `/etc/libssh/libssh_server.config` during RPM build.
Log also the generated configuration file and change the other case
to use standard logging mechanism instead of fprintf.
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Pavol Žáčik <pzacik@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
This linking worked only in CI and local builds, but not during
the build in RPM as it fails on missing symbols that were defined
only in the main library. This is solved as with the other digest
dependencies in external crypto by removing the intermediate
function. We are already linking the md_*.o objects.
The error was like this
sh: symbol lookup error: /path/libssh/libssh-0.12.0-build/libssh-0.12.0/redhat-linux-build/lib/libsntrup761_override.so: undefined symbol: crypto_hash_sha512
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Pavol Žáčik <pzacik@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
Coverity thought that the best_nlines could underflow, but the best_size is
initialized to 0 before calling this function so its moot. Adjusting the code
to be better understandable to static analyzers by initializing the variable
inside of the function.
Thanks coverity!
CID 1548873
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
... if underlying functions read or write more than expected.
This should never happen, but static analysis tools are inventive.
Thanks coverity!
CID 1548868
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
The `first` is intialized to -1 and if we reach this without setting this, we
needlessly call close(-1). It should be no-op, but better be safe.
Thanks coverity!
CID 1644001
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
Checking the session->gssapi is resulting in the very same results, but this
approach is more direct and makes static analysis tools more happy.
Thanks coverity!
CID 1644000
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
... file, even if it was already set before. The options
level handles what was already set.
The proxyJump implementation sets the username from the proxyjump, which
is setting it to NULL, effectively writing the current username to the
new session, which was not possible to override due to the following check.
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
We want to make sure it suceeds because it could fail if
the client tries to send a hostbound public key authentication
request.
Signed-off-by: Pavol Žáčik <pzacik@redhat.com>
Reviewed-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
If there is no host key (e.g., because we are doing
gssapi-keyex with "null" host key algorithm), it does not
make sense to use host bound authentication.
Signed-off-by: Pavol Žáčik <pzacik@redhat.com>
Reviewed-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
All gssapi-keyex tests have to be disabled in Centos Stream 8
because the KEX is not allowed in FIPS. In Centos Stream 9,
only tests against OpenSSH have to be disabled because
OpenSSH only enables gssapi-keyex since Centos Stream 10.
Signed-off-by: Pavol Žáčik <pzacik@redhat.com>
Reviewed-by: Jakub Jelen <jjelen@redhat.com>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
