gss_init_sec_context() with input_cred_handle != GSS_C_NO_CREDENTIAL
should NOT proceed if there is no element in the given credential for
the requested mechanism.
It isn't possible to use non-default credentials to init a context with
a mechanism for which the credential doesn't have an element. That's
exactly what was happening here. The code was left commented out to
describe how one might use SPNEGO with delegated credentials:
store them with gss_store_cred(), then acquire them back.
Bugs exposed by 61720a0:
- test_context --client-name=... --mech-type=ntlm ... fails;
- gss_acquire_cred() with desired_mech=NTLM and
desired_name==GSS_C_NO_NAME fails;
- gss_init_sec_context() with non-default cred handle calls the
mechanism even when the given cred handle has no element for the
requencet mechanism.
tests/gss/check-ntlm works by accident: gss_acquire_cred() with
desired_mechs==GSS_C_NO_OID_SET succeeds mostly because there are
Kerberos credentials available, and then the subsequent
gss_init_sec_context() call works because of the third bug described
above.
Solaris'/Illumos' gss_acquire_cred_with_password() does not have
side-effects. MIT and Heimdal have differed, but it's now agreed that
the Solaris/Illumos behavior is correct.
To make a credential obained with gss_acquire_cred_with_password()
available to other processes, use gss_store_cred().
In at least two instances the krb5 cred handle expiration time was misused
as a remaining lifetime. This is not surprising since the field name is
wrong ("lifetime" not "expiration"). This commit fixes the code, the next
commit will rename the field and change its type from OM_uint32 to time_t.
gss_add_cred() with GSS_C_NO_CREDENTIAL as the input_cred_handle should
act like gss_acquire_cred() with desired_mechs containing just the
desired_mech.
Modify the NTMakefile rules for tests so that a failed test does
not prevent subsequent tests from being executed.
Change-Id: I9595ad4a1527feae7c402241bf06ab21a0b76d4a
In gsskrb5_accept_delegated_token() it is wrong to store the delegated
credentials in the default ccache by default. When the caller does not
provide a target credential handle, we just do nothing and return success.
Test the return value of gsskrb5_accept_delegated_token() against
GSS_S_COMPLETE, rather than 0.
Delegated or other explicit credentials were mishandled, the code only
worked correctly when processing default credentials. In particular
this caused root's default credential cache to be accessed when accepting
delegated credentials in SSH:
ssh_gssapi_accept_ctx() ->
ssh_gssapi_getclient() ->
gss_inquire_cred_by_mech()
When /tmp/krb5cc_0 contained expired tickets, cascaded credentials
stopped working for non-root users!
mechanism credentials created by dynamically loaded mechanisms do not work
because the gm_mech_oid field is unset for such mechanisms (instead, only
gm_mech.gm_mech_oid is).
The first enctype RFC3961 prf output length's bytes are correct because
the little- and big-endian representations of unsigned zero are the
same. The second block of output was wrong because the counter was not
being encoded as big-endian.
This change could break applications. But those applications would not
have been interoperating with other implementations anyways (in
particular: MIT's).
Instead of locally defining prototypes for private functions
_krb5_crc_update and _krb5_crc_init_table simply include
krb5-private.h.
Change-Id: Ia7931f8df2e68eb038d112797edfd456ffcdd23a
To stop the errors when building concurrently, we make a number of
changes:
1. stop including generated files in *_SOURCES,
2. make *-protos.h and *-private.h depend on the *_SOURCES,
3. make all objects depend on *-{protos,private}.h,
4. in a few places change dir/header.h to $(srcdir)/dir/header.h,
This appears to work for me with make -j16 on a 4-way box.
__gss_krb5_mechanism_oid_desc is now defined in gssapi/gssapi_oid.h,
so remove the definition in gssapi/gssapi_krb5.h in favor of including
that header.
Signed-off-by: Love Hörnquist Åstrand <lha@h5l.org>
We turn on a few extra warnings and fix the fallout that occurs
when building with --enable-developer. Note that we get different
warnings on different machines and so this will be a work in
progress. So far, we have built on NetBSD/amd64 5.99.64 (which
uses gcc 4.5.3) and Ubuntu 10.04.3 LTS (which uses gcc 4.4.3).
Notably, we fixed
1. a lot of missing structure initialisers,
2. unchecked return values for functions that glibc
marks as __attribute__((warn-unused-result)),
3. made minor modifications to slc and asn1_compile
which can generate code which generates warnings,
and
4. a few stragglers here and there.
We turned off the extended warnings for many programs in appl/ as
they are nearing the end of their useful lifetime, e.g. rsh, rcp,
popper, ftp and telnet.
Interestingly, glibc's strncmp() macro needed to be worked around
whereas the function calls did not.
We have not yet tried this on 32 bit platforms, so there will be
a few more warnings when we do.
