This fixes a segfault if the _kdc_db_fetch function does not find
the entry in the database (the entry pointer will be NULL if entry
is not found).
Signed-off-by: Samuel Cabrero <scabrero@zentyal.com>
Signed-off-by: Love Hörnquist Åstrand <lha@h5l.org>
RFC6112 provides a method of computing a session key when the PKINIT DH
is used, and mandates it for anonymous pkinit. The session key is computed
using KRB-FX-CF2 from the reply key and a random key chosen by the kdc.
The random key is provided to the client, which is supposed to verify
that the session key was computed this way.
rfc6112 requires kdcs implementing anonymous PKINIT to include an
empty PKINIT-KX padata in PREAUTH_REQUIRED messages.
Including this improves compatibility with MIT kerberos.
This handles referrals for SPNs of the form
E3514235-4B06-11D1-AB04-00C04FC2DCD2/NTDSGUID/REALM, which are
used during DRS replication when we don't know the dnsHostName of the
target DC (which we don't know until the first replication from that
DC completes).
We use the 3rd part of the SPN directly as the realm name in the
referral.
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
kx509 fails to create certs for principals with slashes in them. For
example:
client% kinit foo/admin
foo/admin@EXAMPLE.COM's Password:
client% kx509
Timed out waiting on KCA
The KCA reports: "Principal is not a user." However, there is a use
case set out in this post:
https://thestaticvoid.com/post/2012/10/25/protecting-puppet-with-kerberos/
that would create a kx509 cert for a host principal for authenticating
against a secure HTTP service. This commit modifies the certificate
creation code to allow principals with slashes in them.
The TGS was incorrectly using authtime to compute renew_till for new
tickets, which was in turn leading to endtime potentially being equal to
starttime, which caused the TGS to return KRB5KDC_ERR_NEVER_VALID.
This happens when the TGT renewal lifetime is longer than the max renew
lifetime of any other services, after that much time (target services'
max renew life) passes. The TGT is still good but TGS-REQs fail.
The default heimdal KDC chokes when trying to encrypt a ticket with a weak
server key that has a different type than the session key. The problem
happens in the krb5_crypto_init function called from the _kdc_encode_reply
function.
The existing work-around of the problem temporarily enabled the weak
enctype in case it was disabled but the principal was on the (hard-coded)
exception list.
Unfortunately the code used the keytype of the key encoded in the ticked
(the session key) instead of the keytype of the key used to encrypt the ticket
(the serverkey) thus enabling the incorrect encryption type if those two
are different, for instance des-cbc-md5 and des-cbc-crc.
Change-Id: Ia55dc344e3e5fc9ec1eb93c9e8ebb0a58c673d57
if the query is "preauth" and the caller is seeking a Key, search
try to find a Key that has the default salt but do not exclude keys
that have a non-default salt.
Move the assignment of 'ret' and 'enctype' before the preauth
default salt test. If the only key of the given type is the non-default
salt key, it should be used.
If the caller is not seeking a Key, do not bother with the preauth
test at all since the Key itself doesn't matter and we are simply
seeking an enctype.
Change-Id: I7cd37c579c0bfdd88bccfbc9eb5e5f55cd1910cb
As part of the keytype validity checks within _kdc_get_preferred_key
_kdc_is_weak_exception must be used to permit the afs/* principals
to have only DES in the key list.
Change-Id: I70801ce9b8c4d3f057542541ce11e06d195efd52
If _kdc_find_etype() is being called with 'ret_key' != NULL, the
caller is attempting to find an actual principal key. If 'ret_key'
is NULL then it is seeking a session key type. Only return an enctype
that is not in the principal key list unless 'ret_key' is NULL.
As part of this change remove 'clientbest' and the associated
logic as it is both unnecessary and can produce an enctype for
which the key cannot be returned.
Change-Id: Iba319e95fc1eac139f00b0cce20e1249482d2c6f
The 'use_strongest_session_key' block and its alternate should
have similar behavior except for the order in which the enctype
lists are processed. This patchset attempts to consolidate the
exit processing and ensure that the inner loop enctype and key
validation is the same.
Bugs fixed:
1. In the 'use_strongest_session_key' case, the _kdc_is_weak_exception()
test was applied during the client enctype loop which is only
processed for acceptable enctypes. This test is moved to the
local supported enctypes loop so as not to filter out weak keys
when the service principal has an explicit exception.
2. In the 'use_strongest_session_key' case, the possibility of an
enctype having keys with more than one salt was excluded.
3. In the 'use_strongest_session_key' case, the 'key' variable was
not reset to NULL within each loop of the client enctype list.
4. In the '!use_strongest_session_key' case, the default salt test
and is_preauth was inconsistent with the 'use_strongest_session_key'
block.
With this consolidation, if no enctype is selected and the service
principal is permitted to use 1DES, then 1DES is selected. It doesn't
matter whether 'use_strongest_session_key' is in use or not.
Change-Id: Ib57264fc8bc23df64c70d39b4f6de48beeb54739
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.
