ira/wip.git
10 years agos4:lib/registry/ldb.c - fix trailing whitespaces
Matthias Dieter Wallnöfer [Sat, 6 Mar 2010 18:39:45 +0000 (19:39 +0100)]
s4:lib/registry/ldb.c - fix trailing whitespaces

10 years agos4:registry library - fix up "reg_ldb_set_value"
Matthias Dieter Wallnöfer [Sat, 6 Mar 2010 18:36:01 +0000 (19:36 +0100)]
s4:registry library - fix up "reg_ldb_set_value"

The previous logic was wrong since it tried to add empty data in some cases
which always ended in an error. This problem should be fixed with the new logic.

10 years agos4:registry library - make "reg_ldb_pack/unpack_value" more robust
Matthias Dieter Wallnöfer [Sat, 6 Mar 2010 18:20:48 +0000 (19:20 +0100)]
s4:registry library - make "reg_ldb_pack/unpack_value" more robust

This to prevent segmentation faults, wrong server inputs ecc.

10 years agos4:provision.reg - call us Windows 2008 from the current version point of view
Matthias Dieter Wallnöfer [Sat, 6 Mar 2010 17:51:41 +0000 (18:51 +0100)]
s4:provision.reg - call us Windows 2008 from the current version point of view

10 years agos4:libregistry - change counters to be "unsigned"
Matthias Dieter Wallnöfer [Sat, 7 Nov 2009 20:07:20 +0000 (21:07 +0100)]
s4:libregistry - change counters to be "unsigned"

Also the s4 registry library has to have "unsigned" counters like the Windows
one.

10 years agos3: Fix the build of net_afs.c with --fake-kaserver=yes, bug 7216
Volker Lendecke [Sat, 6 Mar 2010 11:57:35 +0000 (12:57 +0100)]
s3: Fix the build of net_afs.c with --fake-kaserver=yes, bug 7216

Thanks to Geza Gemes <geza@kzsdabas.hu> for filing this bug

10 years agoLDB:tools - change counters to be unsigned
Matthias Dieter Wallnöfer [Fri, 6 Nov 2009 17:35:17 +0000 (18:35 +0100)]
LDB:tools - change counters to be unsigned

In most cases we do count LDB objects which are enumerated within the "unsigned"
type. Therefore no need to use "signed" counters.

10 years agos4:lsa RPC - fix up "gendb_*" result codes
Matthias Dieter Wallnöfer [Sat, 6 Mar 2010 10:19:04 +0000 (11:19 +0100)]
s4:lsa RPC - fix up "gendb_*" result codes

Make the resultcodes consistent: that means:
result < 0  -> NT_STATUS_INTERNAL_DB_CORRUPTION since our DB had a critical
               error
result >= 0 -> depends on the function usage. I tried to let the logic always as
               it was before.

10 years agos4:lsa RPC - Change some counters to be "unsigned" where needed
Matthias Dieter Wallnöfer [Sat, 21 Nov 2009 18:26:02 +0000 (19:26 +0100)]
s4:lsa RPC - Change some counters to be "unsigned" where needed

The "count" size specifiers I typed "uint32_t" since they're often returned as
an "uint32_t" (consider the IDL file). LDB counters need to be "signed" if they
count till a limit of a "gendb*" call or "unsigned" if they count directly the
number of objects.

10 years agos4:samr RPC - Change some counters to be "unsigned" where needed
Matthias Dieter Wallnöfer [Sat, 21 Nov 2009 18:26:02 +0000 (19:26 +0100)]
s4:samr RPC - Change some counters to be "unsigned" where needed

The "count" size specifiers I typed "uint32_t" since they're often returned as
an "uint32_t" (consider the IDL file). LDB counters need to be "signed" if they
count till a limit of a "gendb*" call or "unsigned" if they count directly the
number of objects.

10 years agoFix for bug #7189 - Open txt files with notepad on samba shares creates problem.
Jeremy Allison [Fri, 5 Mar 2010 23:10:30 +0000 (15:10 -0800)]
Fix for bug #7189 - Open txt files with notepad on samba shares creates problem.

Ensure we don't use any of the create_options for Samba private
use. Add a new parameter to the VFS_CREATE call (private_flags)
which is only used internally. Renumber NTCREATEX_OPTIONS_PRIVATE_DENY_DOS
and NTCREATEX_OPTIONS_PRIVATE_DENY_FCB to match the S4 code).
Rev. the VFS interface to version 28.

Jeremy.

10 years agos4/tort: fix annoying "union spoolss_PrinterInfo" undefined warnings
Kamen Mazdrashki [Thu, 4 Mar 2010 16:00:15 +0000 (18:00 +0200)]
s4/tort: fix annoying "union spoolss_PrinterInfo" undefined warnings

Signed-off-by: Günther Deschner <gd@samba.org>
10 years agos3: re-run make samba3-idl.
Günther Deschner [Fri, 5 Mar 2010 21:03:26 +0000 (22:03 +0100)]
s3: re-run make samba3-idl.

Guenther

10 years agowinreg: fix winreg_QueryValue IDL.
Günther Deschner [Fri, 5 Mar 2010 20:56:50 +0000 (21:56 +0100)]
winreg: fix winreg_QueryValue IDL.

Note that before this change pidl generated code that just dereferenced size_is
and length_is values from unique pointers without checking whether these
pointers were actually NULL.

With this change, pidl now throws a warning like:

warning: Got pointer for `data_size', expected fully derefenced variable

which is not correct, probably because pidl does not evaluate the C expression.

Guenther

10 years agos3-rpcclient: Leave setprinterdata directly if not enough args were passed.
Andreas Schneider [Fri, 5 Mar 2010 10:15:35 +0000 (11:15 +0100)]
s3-rpcclient: Leave setprinterdata directly if not enough args were passed.

If we don't pass enough arguments we don't have to close the connection.

Signed-off-by: Günther Deschner <gd@samba.org>
10 years agos3-rpc_server: Document rpc_pipe_open_internal.
Andreas Schneider [Fri, 5 Mar 2010 14:54:31 +0000 (15:54 +0100)]
s3-rpc_server: Document rpc_pipe_open_internal.

Signed-off-by: Günther Deschner <gd@samba.org>
10 years agos4-smbtorture: fix RPC-SPOOLSS-WIN after PrinterData IDL changes.
Günther Deschner [Fri, 5 Mar 2010 19:32:16 +0000 (20:32 +0100)]
s4-smbtorture: fix RPC-SPOOLSS-WIN after PrinterData IDL changes.

Guenther

10 years agos4:drsuapi RPC - Change also here counters to "unsigned"
Matthias Dieter Wallnöfer [Sat, 21 Nov 2009 17:57:24 +0000 (18:57 +0100)]
s4:drsuapi RPC - Change also here counters to "unsigned"

No need to have "signed" counters at those places.

10 years agos4:dsdb/util - Change also here counters to "unsigned"
Matthias Dieter Wallnöfer [Fri, 6 Nov 2009 19:14:41 +0000 (20:14 +0100)]
s4:dsdb/util - Change also here counters to "unsigned"

No need to have "signed" counters at those places.

10 years agos4:repl - change also here the counter variables to "unsigned"
Matthias Dieter Wallnöfer [Fri, 6 Nov 2009 19:14:41 +0000 (20:14 +0100)]
s4:repl - change also here the counter variables to "unsigned"

I changed also some "uint32_t" to "unsigned" since the LDB interface doesn't
specify the bitlength of the unsigned type.

10 years agos4:wrepl_server - change also here the counter variables to "unsigned"
Matthias Dieter Wallnöfer [Sat, 7 Nov 2009 20:15:17 +0000 (21:15 +0100)]
s4:wrepl_server - change also here the counter variables to "unsigned"

I changed also some "uint32_t" to "unsigned" since the LDB interface doesn't
specify the bitlength of the unsigned type.

10 years agos3: Make "smbcontrol xx debuglevel" print the correct cluster pid
Volker Lendecke [Fri, 5 Mar 2010 16:06:08 +0000 (17:06 +0100)]
s3: Make "smbcontrol xx debuglevel" print the correct cluster pid

10 years agos4-spoolss: fix dcesrv_spoolss_GetPrinterData build.
Günther Deschner [Fri, 5 Mar 2010 15:34:17 +0000 (16:34 +0100)]
s4-spoolss: fix dcesrv_spoolss_GetPrinterData build.

really not my day today...

Guenther

10 years agos4-smbtorture: fix test_SetPrinterDataEx_matrix() call. sorry...
Günther Deschner [Fri, 5 Mar 2010 15:00:14 +0000 (16:00 +0100)]
s4-smbtorture: fix test_SetPrinterDataEx_matrix() call. sorry...

Guenther

10 years agos3: Remove the unused parameter "persistent" from fetch_locked_internal
Volker Lendecke [Fri, 5 Mar 2010 14:30:22 +0000 (15:30 +0100)]
s3: Remove the unused parameter "persistent" from fetch_locked_internal

10 years agos3: db->persistent==true was handled earlier, make this more obvious
Volker Lendecke [Fri, 5 Mar 2010 14:28:39 +0000 (15:28 +0100)]
s3: db->persistent==true was handled earlier, make this more obvious

10 years agos4-smbtorture: when checking PrinterData matrix, include winreg cross reference checks.
Günther Deschner [Fri, 5 Mar 2010 13:42:10 +0000 (14:42 +0100)]
s4-smbtorture: when checking PrinterData matrix, include winreg cross reference checks.

Guenther

10 years agos4-smbtorture: hand down printername to test_SetPrinterDataEx_matrix.
Günther Deschner [Fri, 5 Mar 2010 13:03:03 +0000 (14:03 +0100)]
s4-smbtorture: hand down printername to test_SetPrinterDataEx_matrix.

Guenther

10 years agos4-smbtorture: add more paranoid checks for REG_DWORD SetPrinterDataEx tests.
Günther Deschner [Thu, 4 Mar 2010 23:56:52 +0000 (00:56 +0100)]
s4-smbtorture: add more paranoid checks for REG_DWORD SetPrinterDataEx tests.

Guenther

10 years agos4-smbtorture: remove check for selfmade parsing artefact.
Günther Deschner [Thu, 4 Mar 2010 22:13:10 +0000 (23:13 +0100)]
s4-smbtorture: remove check for selfmade parsing artefact.

We were checking against a phenomenon that we created ourselves at the pidl
layer...

Guenther

10 years agos3-libads: fix get_remote_printer_publishing_data after spoolss_EnumPrinterDataEx...
Günther Deschner [Thu, 4 Mar 2010 16:58:49 +0000 (17:58 +0100)]
s3-libads: fix get_remote_printer_publishing_data after spoolss_EnumPrinterDataEx IDL change.

Guenther

10 years agos3-spoolss: fix _spoolss_EnumPrinterDataEx after idl
Günther Deschner [Thu, 4 Mar 2010 15:47:33 +0000 (16:47 +0100)]
s3-spoolss: fix _spoolss_EnumPrinterDataEx after idl

Guenther

10 years agos3-rpcclient: fix rpcclient after spoolss_EnumPrinterDataEx IDL change.
Günther Deschner [Thu, 4 Mar 2010 15:47:10 +0000 (16:47 +0100)]
s3-rpcclient: fix rpcclient after spoolss_EnumPrinterDataEx IDL change.

Guenther

10 years agos3-net: fix net after spoolss_EnumPrinterDataEx IDL change.
Günther Deschner [Thu, 4 Mar 2010 15:46:53 +0000 (16:46 +0100)]
s3-net: fix net after spoolss_EnumPrinterDataEx IDL change.

Guenther

10 years agos3: re-run make samba3-idl.
Günther Deschner [Thu, 4 Mar 2010 16:04:19 +0000 (17:04 +0100)]
s3: re-run make samba3-idl.

Guenther

10 years agospoolss: fix spoolss_EnumPrinterDataEx IDL.
Günther Deschner [Thu, 4 Mar 2010 15:40:31 +0000 (16:40 +0100)]
spoolss: fix spoolss_EnumPrinterDataEx IDL.

Guenther

10 years agos4-smbtorture: fix smbtorture after GetPrinterData{Ex} after IDL changes.
Günther Deschner [Thu, 4 Mar 2010 21:40:31 +0000 (22:40 +0100)]
s4-smbtorture: fix smbtorture after GetPrinterData{Ex} after IDL changes.

Guenther

10 years agos4-spoolss: fix spoolss_GetPrinterData implementation after IDL change.
Günther Deschner [Thu, 4 Mar 2010 17:21:50 +0000 (18:21 +0100)]
s4-spoolss: fix spoolss_GetPrinterData implementation after IDL change.

Guenther

10 years agos3-rpcclient: fix rpcclient after spoolss_GetPrinterData{Ex} IDL change.
Günther Deschner [Thu, 4 Mar 2010 14:46:28 +0000 (15:46 +0100)]
s3-rpcclient: fix rpcclient after spoolss_GetPrinterData{Ex} IDL change.

Guenther

10 years agos3-spoolss: fix _spoolss_GetPrinterDataEx after IDL change.
Günther Deschner [Thu, 4 Mar 2010 14:39:59 +0000 (15:39 +0100)]
s3-spoolss: fix _spoolss_GetPrinterDataEx after IDL change.

Guenther

10 years agos3: re-run make samba3-idl.
Günther Deschner [Thu, 4 Mar 2010 16:02:42 +0000 (17:02 +0100)]
s3: re-run make samba3-idl.

Guenther

10 years agospoolss: rollback GetPrinterData[Ex] IDL.
Günther Deschner [Thu, 4 Mar 2010 14:38:18 +0000 (15:38 +0100)]
spoolss: rollback GetPrinterData[Ex] IDL.

Guenther

10 years agos4-smbtorture: fix smbtorture after SetPrinterData{Ex} IDL changes.
Günther Deschner [Thu, 4 Mar 2010 21:22:45 +0000 (22:22 +0100)]
s4-smbtorture: fix smbtorture after SetPrinterData{Ex} IDL changes.

Guenther

10 years agos3-rpcclient: fix rpcclient after spoolss_SetPrinterData{Ex} IDL change.
Günther Deschner [Thu, 4 Mar 2010 14:34:49 +0000 (15:34 +0100)]
s3-rpcclient: fix rpcclient after spoolss_SetPrinterData{Ex} IDL change.

Guenther

10 years agos3-net: fix net after spoolss_SetPrinterData{Ex} IDL change.
Günther Deschner [Thu, 4 Mar 2010 14:34:22 +0000 (15:34 +0100)]
s3-net: fix net after spoolss_SetPrinterData{Ex} IDL change.

Guenther

10 years agos3-spoolss: fix _spoolss_SetPrinterData{Ex} after IDL change.
Günther Deschner [Thu, 4 Mar 2010 16:00:34 +0000 (17:00 +0100)]
s3-spoolss: fix _spoolss_SetPrinterData{Ex} after IDL change.

Guenther

10 years agos3: re-run make samba3-idl.
Günther Deschner [Thu, 4 Mar 2010 16:00:09 +0000 (17:00 +0100)]
s3: re-run make samba3-idl.

Guenther

10 years agospoolss: rollback SetPrinterData{Ex} IDL.
Günther Deschner [Wed, 3 Mar 2010 17:22:57 +0000 (18:22 +0100)]
spoolss: rollback SetPrinterData{Ex} IDL.

Guenther

10 years agos4:dcesrv_samr_AddAliasMember - wrap a long "DEBUG" statement
Matthias Dieter Wallnöfer [Fri, 5 Mar 2010 14:01:07 +0000 (15:01 +0100)]
s4:dcesrv_samr_AddAliasMember - wrap a long "DEBUG" statement

10 years agos4:libcli/wrepl: convert wrepl_associate_stop_send to tevent_req
Stefan Metzmacher [Wed, 3 Mar 2010 17:35:55 +0000 (18:35 +0100)]
s4:libcli/wrepl: convert wrepl_associate_stop_send to tevent_req

metze

10 years agos4:libcli/wrepl: convert wrepl_connect_send to tevent_req
Stefan Metzmacher [Wed, 3 Mar 2010 16:18:23 +0000 (17:18 +0100)]
s4:libcli/wrepl: convert wrepl_connect_send to tevent_req

metze

10 years agos4:libcli/wrepl: convert wrepl_associate_send to tevent_req
Stefan Metzmacher [Wed, 3 Mar 2010 17:38:16 +0000 (18:38 +0100)]
s4:libcli/wrepl: convert wrepl_associate_send to tevent_req

metze

10 years agos4:libcli/wrepl: convert wrepl_pull_table_send to tevent_req
Stefan Metzmacher [Wed, 3 Mar 2010 14:14:28 +0000 (15:14 +0100)]
s4:libcli/wrepl: convert wrepl_pull_table_send to tevent_req

metze

10 years agos4:libcli/wrepl: convert wrepl_pull_names_send to tevent_req
Stefan Metzmacher [Wed, 3 Mar 2010 14:12:05 +0000 (15:12 +0100)]
s4:libcli/wrepl: convert wrepl_pull_names_send to tevent_req

metze

10 years agos4:libcli/wrepl: use UTIL_TEVENT
Stefan Metzmacher [Wed, 3 Mar 2010 12:38:08 +0000 (13:38 +0100)]
s4:libcli/wrepl: use UTIL_TEVENT

metze

10 years agos4:libcli/wrepl: avoid neested named structures
Stefan Metzmacher [Wed, 3 Mar 2010 08:45:01 +0000 (09:45 +0100)]
s4:libcli/wrepl: avoid neested named structures

metze

10 years agos4:torture/nbt: avoid the usage of wrepl_pull_table_send()
Stefan Metzmacher [Wed, 3 Mar 2010 10:31:03 +0000 (11:31 +0100)]
s4:torture/nbt: avoid the usage of wrepl_pull_table_send()

metze

10 years agolibcli/auth: add a const to des_crypt112_16()
Stefan Metzmacher [Fri, 5 Mar 2010 07:22:36 +0000 (08:22 +0100)]
libcli/auth: add a const to des_crypt112_16()

metze

10 years agos3:winbindd: add DEBUG(10,...) for the end of each top level
Stefan Metzmacher [Fri, 5 Mar 2010 10:16:12 +0000 (11:16 +0100)]
s3:winbindd: add DEBUG(10,...) for the end of each top level

That will hopefully make debugging a bit easier (at least for me).

metze

10 years agos4:kcc - Change some counter variables to be unsigned
Matthias Dieter Wallnöfer [Fri, 6 Nov 2009 19:14:41 +0000 (20:14 +0100)]
s4:kcc - Change some counter variables to be unsigned

The upper limits are unsigned variables therefore also the counter variables
need to be like that.

10 years agos4:samdb_privilege.c - Change two counter variables to unsigned
Matthias Dieter Wallnöfer [Fri, 6 Nov 2009 19:14:41 +0000 (20:14 +0100)]
s4:samdb_privilege.c - Change two counter variables to unsigned

Also here in both cases the unsigned counter fits better than the signed one.

10 years agos4:cracknames - Change two counter variables to unsigned
Matthias Dieter Wallnöfer [Fri, 6 Nov 2009 19:14:41 +0000 (20:14 +0100)]
s4:cracknames - Change two counter variables to unsigned

In both cases the unsigned counter fits better:
- in the first one since we are counting LDB objects starting from 0
- in the second since we are counting an array starting from 0

10 years agos4-pvfs_sys: build on systems without O_NOFOLLOW or O_DIRECTORY
Andrew Tridgell [Fri, 5 Mar 2010 12:07:26 +0000 (23:07 +1100)]
s4-pvfs_sys: build on systems without O_NOFOLLOW or O_DIRECTORY

10 years agos4-pvfs_sys: talloc_free should be before errno restore
Andrew Tridgell [Fri, 5 Mar 2010 11:58:18 +0000 (22:58 +1100)]
s4-pvfs_sys: talloc_free should be before errno restore

talloc can potentially change the errno

10 years agos4-pvfs: use pvfs_sys_fchmod()
Andrew Tridgell [Fri, 5 Mar 2010 11:48:09 +0000 (22:48 +1100)]
s4-pvfs: use pvfs_sys_fchmod()

10 years agos4-pvfs: set default for perm override based on system features
Andrew Tridgell [Fri, 5 Mar 2010 11:47:50 +0000 (22:47 +1100)]
s4-pvfs: set default for perm override based on system features

If the system has O_NOFOLLOW and O_DIRECTORY then we allow for
overrides by default. If not, then we disable by default, as we will
be more vulnerable to symlink attacks

10 years agos4-pvfs: use O_FOLLOW one level at a time for security overrides
Andrew Tridgell [Fri, 5 Mar 2010 11:46:44 +0000 (22:46 +1100)]
s4-pvfs: use O_FOLLOW one level at a time for security overrides

To prevent symlink attacks we need to use O_NOFOLLOW one level at a
time when processing a root security override

10 years agoreplace: added get_current_dir_name()
Andrew Tridgell [Fri, 5 Mar 2010 11:45:27 +0000 (22:45 +1100)]
replace: added get_current_dir_name()

10 years agos4-pvfs: use pvfs_sys_*() functions to wrap posix calls
Andrew Tridgell [Fri, 5 Mar 2010 06:52:35 +0000 (17:52 +1100)]
s4-pvfs: use pvfs_sys_*() functions to wrap posix calls

This allows for root override, which fixes many problems with
mismatches between NT ACL permissions and unix permissions.

Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>

10 years agos4-pvfs: new pvfs_sys module
Andrew Tridgell [Fri, 5 Mar 2010 06:51:12 +0000 (17:51 +1100)]
s4-pvfs: new pvfs_sys module

The pvfs_sys_*() calls provide wrapper functions for posix file
functions which use root privileges to override EACCES failures if
PVFS_FLAG_PERM_OVERRIDE is set

Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>

10 years agos4-pvfs: added new pvfs flag PVFS_FLAG_PERM_OVERRIDE
Andrew Tridgell [Fri, 5 Mar 2010 06:50:18 +0000 (17:50 +1100)]
s4-pvfs: added new pvfs flag PVFS_FLAG_PERM_OVERRIDE

This flag indicates that we should use root privileges to override
unix permissions when the NT ACLs indicate that access should be
granted

Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>

10 years agos4-privs: add root_privileges_original_uid()
Andrew Tridgell [Fri, 5 Mar 2010 06:49:11 +0000 (17:49 +1100)]
s4-privs: add root_privileges_original_uid()

This can be used to get the uid we changed away from when we gained
root privileges

Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>

10 years agos3-selftest: enable RPC-WINREG against Samba 3.
Günther Deschner [Fri, 5 Mar 2010 10:06:46 +0000 (11:06 +0100)]
s3-selftest: enable RPC-WINREG against Samba 3.

Guenther

10 years agos4-smbtorture: skip NotifyChangeKeyValue test against s3 for now.
Günther Deschner [Fri, 5 Mar 2010 10:06:18 +0000 (11:06 +0100)]
s4-smbtorture: skip NotifyChangeKeyValue test against s3 for now.

Guenther

10 years agos4:torture/rpc/samr.c - add some decision possibility constants to some switch
Matthias Dieter Wallnöfer [Fri, 5 Mar 2010 07:50:08 +0000 (08:50 +0100)]
s4:torture/rpc/samr.c - add some decision possibility constants to some switch

At the moment nothing is done when the enumeration variable is set to one of
those constants as before. This is only to quite nasty warnings.

10 years agos4:torture/rpc/samr.c - make some argument of function "test_SamLogon_with_creds...
Matthias Dieter Wallnöfer [Fri, 5 Mar 2010 07:49:25 +0000 (08:49 +0100)]
s4:torture/rpc/samr.c - make some argument of function "test_SamLogon_with_creds" constant

This to quiet warnings.

10 years agos4:torture/winbind/struct_based.c - fix up (un)signedness of a function argument
Matthias Dieter Wallnöfer [Fri, 5 Mar 2010 07:47:48 +0000 (08:47 +0100)]
s4:torture/winbind/struct_based.c - fix up (un)signedness of a function argument

Otherwise always a warning is generated.

10 years agos4-pvfs: log more error conditions in NTVFS backend
Andrew Tridgell [Fri, 5 Mar 2010 03:59:08 +0000 (14:59 +1100)]
s4-pvfs: log more error conditions in NTVFS backend

This should make is easier to track down some bug reports

Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>

10 years agos4-pvfs: move the private ntcreatex flags to private_flags
Andrew Tridgell [Fri, 5 Mar 2010 03:57:50 +0000 (14:57 +1100)]
s4-pvfs: move the private ntcreatex flags to private_flags

Re-using two of the create_options bits was bound to eventually
cause problems, and indeed, Windows7 now uses one of those bits
when opening text files.

Fixes bug 7189

10 years agos4-rpc: don't use s->credentials after it is freed
Andrew Tridgell [Fri, 5 Mar 2010 02:49:49 +0000 (13:49 +1100)]
s4-rpc: don't use s->credentials after it is freed

10 years agos4-torture: fixed commas separating C statements
Andrew Tridgell [Fri, 5 Mar 2010 02:02:16 +0000 (13:02 +1100)]
s4-torture: fixed commas separating C statements

10 years agos3: Fix unnecessary traversing winbindd_cache.tdb in SIGHUP handler.
Bo Yang [Sat, 6 Mar 2010 12:58:23 +0000 (20:58 +0800)]
s3: Fix unnecessary traversing winbindd_cache.tdb in SIGHUP handler.

Signed-off-by: Bo Yang <boyang@samba.org>
10 years agos4-python: only install external python libs that are missing
Andrew Tridgell [Fri, 5 Mar 2010 00:53:19 +0000 (11:53 +1100)]
s4-python: only install external python libs that are missing

10 years agos4-python: import a copy of the python dns library
Andrew Tridgell [Fri, 5 Mar 2010 00:45:40 +0000 (11:45 +1100)]
s4-python: import a copy of the python dns library

This library is not installed on enough systems for us to rely
on it being available. We use the system copy if possible, and
fallback to this local copy

Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>

10 years agos4-dns: use samba.external to pull in the dns.resolver library
Andrew Tridgell [Fri, 5 Mar 2010 00:45:10 +0000 (11:45 +1100)]
s4-dns: use samba.external to pull in the dns.resolver library

10 years agos4-python: allow us to have samba copies of python libraries we depend on
Andrew Tridgell [Fri, 5 Mar 2010 00:44:45 +0000 (11:44 +1100)]
s4-python: allow us to have samba copies of python libraries we depend on

For python libraries like dns.resolver it is useful to be able to install
a copy of the library with Samba. This set of functions allows us to do that
while using the locally installed version if it is available

Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>

10 years agos4-dns-ex: use autoclose on the dns child pipe
Andrew Tridgell [Thu, 4 Mar 2010 23:43:34 +0000 (10:43 +1100)]
s4-dns-ex: use autoclose on the dns child pipe

I'm hoping this will fix an occasional segfault I've noticed where
epoll still calls events on a closed fde

Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>

10 years agos4-messaging: use auto-close on the socket
Andrew Tridgell [Thu, 4 Mar 2010 23:42:41 +0000 (10:42 +1100)]
s4-messaging: use auto-close on the socket

10 years agosrv_pipe.c doesn't reference current_user anymore. Remove it.
Jeremy Allison [Thu, 4 Mar 2010 22:59:29 +0000 (14:59 -0800)]
srv_pipe.c doesn't reference current_user anymore. Remove it.

Jeremy.

10 years agos4:auth/sam.c - change base context for the "tmp_ctx" context in "authsam_expand_nest...
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 21:18:10 +0000 (22:18 +0100)]
s4:auth/sam.c - change base context for the "tmp_ctx" context in "authsam_expand_nested_groups"

Better use the "res_sids_ctx" as base context for the "tmp_ctx" and not the
long-living "sam_ctx"/"ldb" context to prevent memory leaks.

10 years agos4:ldap.py - give the "primaryGroupToken" test a better name
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 16:54:45 +0000 (17:54 +0100)]
s4:ldap.py - give the "primaryGroupToken" test a better name

It tests also some other constructed attributes in a basic way.

10 years agos4:ldap.py - add test for "tokenGroups"
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 16:51:09 +0000 (17:51 +0100)]
s4:ldap.py - add test for "tokenGroups"

10 years agos4:operational LDB - don't accidentally "ate" search helper attributes if we need...
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 15:28:03 +0000 (16:28 +0100)]
s4:operational LDB - don't accidentally "ate" search helper attributes if we need them for more constructed attributes

With this patch we delete the helper attributes at the end where all constructed
attributes have already been computed.

10 years agos4:operational LDB module - make the counters unsigned
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 14:44:10 +0000 (15:44 +0100)]
s4:operational LDB module - make the counters unsigned

No need to have signed counters here.

10 years agos4:operational LDB - implement the "tokenGroups" constructed attribute
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 12:58:16 +0000 (13:58 +0100)]
s4:operational LDB - implement the "tokenGroups" constructed attribute

It contains the transitive SID closure (expand member/memberOf attributes) of a
certain SAM object. The "tokenGroups" attribute never contains the SID of the
object itself.

References: http://msdn.microsoft.com/en-us/library/ms680275(VS.85).aspx,
http://support.microsoft.com/kb/301916,
MS-ADTS 3.1.1.4.5.19.

10 years agos4:sam.c - make "authsam_expand_nested_groups" public
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 12:49:18 +0000 (13:49 +0100)]
s4:sam.c - make "authsam_expand_nested_groups" public

This is needed by the "tokenGroups" work in the operational LDB module.

10 years agos4:sam.c - cosmetic indentation fix
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 12:48:17 +0000 (13:48 +0100)]
s4:sam.c - cosmetic indentation fix

10 years agos4:sam.c - change variable types to unsigned in "sids_contains_sid"
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 12:46:34 +0000 (13:46 +0100)]
s4:sam.c - change variable types to unsigned in "sids_contains_sid"

Should also be unsigned - no need for a signed "i" and "num_sids" here.

10 years agos4:operational LDB module - use right memory context int "construct_primary_group_token"
Matthias Dieter Wallnöfer [Thu, 4 Mar 2010 13:28:04 +0000 (14:28 +0100)]
s4:operational LDB module - use right memory context int "construct_primary_group_token"

Use the "msg" as temporary context and not "ldb" which lives much longer.

10 years agoRevert "s3:configure: add --enable-as-needed"
Karolin Seeger [Thu, 4 Mar 2010 15:18:44 +0000 (16:18 +0100)]
Revert "s3:configure: add --enable-as-needed"

This reverts commit 22d316926b9589608d332143c1fa134229b75b3c.

Please see bug #7209 for details.