Commit Graph

27070 Commits

Author SHA1 Message Date
Love Hornquist Astrand
28563373a8 more documentation about pkinit 2011-10-14 14:49:00 +02:00
Love Hornquist Astrand
1b88a3b6c7 Only does implicit matching for first component, reported by Harry Coin 2011-10-14 00:33:57 +02:00
Love Hornquist Astrand
d6474982e5 document kdc options 2011-10-12 15:37:24 +02:00
Love Hornquist Astrand
c2be6a8580 we have @subsection Configure the KDC, let remove the XXX 2011-10-12 15:29:59 +02:00
Love Hornquist Astrand
f574312ce1 remove kaserver ref 2011-10-12 12:41:00 +02:00
Love Hornquist Astrand
a061e7b22f remove kaserver ref 2011-10-12 12:40:59 +02:00
Love Hornquist Astrand
8192b9ed35 remove refernces to kerberos 4 and kaserver 2011-10-12 12:40:59 +02:00
Nicolas Williams
4c6976a6bd Fix check-des
The previous fix was incomplete.  But it also finally uncovered an
    old check-des problem that I'd had once and which may have gotten
    papered over by changing the default of one of the *strongest* KDC
    parameters.  The old problem is that we were passing the wrong
    enctype to _kdc_encode_reply(): we were passing the session key
    enctype where the ticket enc-part key's enctype was expected.

    The whole enctype being passed in is superfluous anyways.  Let's
    clean that up next.
2011-10-12 01:17:54 -05:00
Nicolas Williams
12cd2c9cbd Fix TGS ticket enc-part key selection
When I added support for configuring how the KDC selects session,
    reply, and ticket enc-part keys I accidentally had the KDC use the
    session key selection algorithm for selecting the ticket enc-part
    key.  This becomes a problem when using a Heimdal KDC with an MIT
    KDB as the HDB backend and when the krbtgt keys are not in
    strongest-to-weakest order, in which case forwardable tickets minted
    by the Heimdal KDC will not be accepted by MIT KDCs with the same
    KDB.
2011-10-11 23:57:58 -05:00
Love Hornquist Astrand
8aceafc430 moved to lib/gssapi/oid.txt 2011-10-11 20:28:29 +02:00
Luke Howard
f48061bda7 check localname attr authenticated 2011-10-08 12:23:25 +11:00
Luke Howard
07777511d1 implement gss_localname 2011-10-08 12:15:09 +11:00
Nicolas Williams
1b03abb250 This should be the final fix for enctype 0 issues (tested)
But how to build an MIT KDB with enctype 0 keys for testing in
    Heimdal?  Hmmm...
2011-10-06 00:55:54 -05:00
Nicolas Williams
e15cabe10a Fix for enctype 0 / length 0 keys in MIT HDB backend was incomplete 2011-10-05 17:50:26 -05:00
Nicolas Williams
3d6f86af27 Fix segfault in hdb-mitdb when princ have salt 2011-10-02 23:08:37 -05:00
Love Hornquist Astrand
ee7340860a remove classic input 2011-10-02 17:23:26 +02:00
Love Hornquist Astrand
391bce7ca2 lets try classic input() too 2011-10-02 13:46:28 +02:00
Love Hornquist Astrand
49b471f1a2 Heimdal 1.5.1 2011-10-02 13:26:37 +02:00
Love Hornquist Astrand
923f3a434f free krb5_storage when dne 2011-09-30 12:44:34 +02:00
Love Hornquist Astrand
3c0e9b8fd1 dont use bool since its c99 2011-09-30 11:49:01 +02:00
Love Hornquist Astrand
b76f6f1e93 add more people 2011-09-29 13:58:23 +02:00
Love Hornquist Astrand
a1738def66 solaris want's _XOPEN_SOURCE set to 600, From Ragnar Sundblad 2011-09-29 13:57:24 +02:00
Love Hornquist Astrand
240b298e61 Simplify base/section rules, patch from David R Boldt 2011-09-29 13:52:06 +02:00
Nicolas Williams
1de3db5678 Make getpwnam_r() check work on Ubuntu
NULL is not defined w/o including stddefs.h, so the test program
    fails to compile.  A simple s/NULL/0/ fixes it.
2011-09-27 17:32:11 -05:00
Nicolas Williams
4ae39ce0b3 Added one more export to libheimbase -- needed by tests 2011-09-27 16:29:11 -05:00
Nicolas Williams
cdceea1b3e Revert "A simple (but wrong?) fix for Solaris build issues"
This reverts commit 6a7f2a5b2e.
2011-09-27 16:05:34 -05:00
Nicolas Williams
e927695a5d Better fix for the Solaris getpwnam_r() build issue 2011-09-27 16:05:03 -05:00
Jeffrey Altman
cbf126bede Permit NULL context to krb5_get_error_message()
Application developers have a tendency to use krb5_get_error_message()
as a drop in replacement for error_message() and under various
circumstances they pass in a NULL context.  This method works fine
for MIT's implementation which ignores the context but in Heimdal
passing in a NULL context would dump core.

This patch set modifies krb5_get_error_message() in order to
permit the passing of a NULL context.  First, if the context
is NULL, an attempt will be made to allocate one locally for
the purpose of evaluating the error code.  Second, if a local
context cannot be allocated, fall back on calling error_message().
If error_message() fails to return a string, then generate an
"unknown error" response.

Only if all of the above fails is NULL returned.

Change-Id: If4baf7d6c428cf0baf11c044b8dfd5c2b3cdf7e4
2011-09-27 14:26:26 -04:00
Love Hornquist Astrand
03df77d491 link with heimbase 2011-09-27 19:40:10 +02:00
Nicolas Williams
6a7f2a5b2e A simple (but wrong?) fix for Solaris build issues 2011-09-27 01:59:44 -05:00
Nicolas Williams
c3936f72c3 Test for hdb-mitdb, including an MIT KDB file. 2011-09-26 17:08:33 -05:00
Love Hornquist Astrand
308be3a9e2 Don't export tests from library, reported in bug from Tom Payeire 2011-09-26 17:53:13 +02:00
Love Hornquist Astrand
f1df66341e quote to make solaris sh apply, patch from Tom Payeire 2011-09-26 17:50:45 +02:00
Love Hornquist Astrand
0595af118e document KRB5_CONFIG 2011-09-26 14:59:30 +02:00
Love Hornquist Astrand
e138bed4c7 export error functions 2011-09-26 09:29:14 +02:00
Love Hornquist Astrand
777b24fbb5 add krb5_is_enctype_weak 2011-09-26 08:47:37 +02:00
Jeffrey Altman
9a127beb26 Windows: set default ccache to registry
Add _krb5_set_default_cc_name_to_registry() function and
call use it on Windows to set the user's default credential cache.

Change-Id: Ib59ff218a098a841bc61846abf873736380b5c6c
2011-09-26 02:00:13 -04:00
Jeffrey Altman
132693df5a Add _krb5_store_string_to_reg_value() for Windows
The new _krb5_store_string_to_reg_value() function permits
the caller to create REG_SZ, REG_EXPAND_SZ, REG_DWORD, and
REG_MULTI_SZ entries in the registry.

Change-Id: Ib5740ad07209618d8ea4c0bf3c75615f27e98b4e
2011-09-26 01:58:29 -04:00
Jeffrey Altman
d6bff3199d add cache type column to kswitch output
Change-Id: I6b27ba6f7c853da56f81b45b312ef06e13f2dbb0
2011-09-26 01:56:14 -04:00
Jeffrey Altman
91a9a11b75 avoid use of vasnprintf in base/error.c
commit c94f299fc8 uses vasnprintf
which was previously removed from the libheimbase directory in
order to prevent a dependency on libroken.

Replace vasnprintf with snprintf and malloc to avoid the
dependency.

Change-Id: I029e7e6883406ca7311490a3dab4b65cad3ba70b
2011-09-26 01:53:25 -04:00
Jeffrey Altman
3854e64a4a include weak etypes in default etype list if allow_weak_crypto
commit 0ed83cebd3 removed the
weak enctypes from the default enctype list.  This is a change
in behavior from 1.5.x which permitted the use of weak enctypes
if "allow_weak_crypto" is set to true.  This patchset creates
two default enctype lists.  One with weak enctypes and the other
without.  The weak version is used if "allow_weak_crypto" is set
to true.

Change-Id: Ide5cce0645836249031350bfaf619d970635e579
2011-09-26 01:44:16 -04:00
Love Hornquist Astrand
00773b4119 test verify_mic is both directions 2011-09-25 20:38:17 +02:00
Love Hornquist Astrand
f910d2174b Disable java gssapi tests because of 7077640 and 7077646 2011-09-25 20:24:26 +02:00
Love Hornquist Astrand
d9dc7c9761 try read message in java code too 2011-09-25 19:40:02 +02:00
Love Hornquist Astrand
c94f299fc8 add heim_error_t 2011-09-25 19:03:23 +02:00
Love Hornquist Astrand
d3f85af92c move deprecated function to deprecated.c 2011-09-25 19:03:11 +02:00
Love Hornquist Astrand
0793246e99 use heim_error_t 2011-09-25 19:00:01 +02:00
Love Hornquist Astrand
7e36705ee3 use heim_error_t 2011-09-25 18:59:42 +02:00
Love Hornquist Astrand
9f46874cfb add heimbase.la 2011-09-25 18:59:10 +02:00
Love Hornquist Astrand
0b02f05a19 error_buf is gone 2011-09-25 17:49:06 +02:00