ira/wip.git
14 years agos4-ntp_signd: Migrate to tsocket.
Andreas Schneider [Tue, 12 Jan 2010 16:30:04 +0000 (17:30 +0100)]
s4-ntp_signd: Migrate to tsocket.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
14 years agoFixed a problem with incorrect default SD owner/group.
Nadezhda Ivanova [Wed, 13 Jan 2010 13:16:38 +0000 (15:16 +0200)]
Fixed a problem with incorrect default SD owner/group.

14 years agos3: Fix a winbind segfault in "trusted_domains"
Volker Lendecke [Wed, 13 Jan 2010 11:20:26 +0000 (12:20 +0100)]
s3: Fix a winbind segfault in "trusted_domains"

We have to initialize domain->backend by calling "get_cache" before doing a
query

Thanks to Christian Ambach to find this :-)

14 years agoComparison tool for LDAP servers (using Ldb)
Zahari Zahariev [Wed, 13 Jan 2010 08:41:56 +0000 (10:41 +0200)]
Comparison tool for LDAP servers (using Ldb)

This tool is integrated with Samba4 Ldb. It provides a useful output
where you can find easy differences in objects or attributes within
naming context (Domain, Configuration or Schema).

Added functionality for two sets of credentials.

14 years agoMerge branch 'master' of git://git.samba.org/samba
Nadezhda Ivanova [Wed, 13 Jan 2010 10:02:31 +0000 (12:02 +0200)]
Merge branch 'master' of git://git.samba.org/samba

14 years agoFix bug #7034 - vfs_cap causes signal 11 (SIGSEGV)
SASAJIMA Toshihiro [Wed, 13 Jan 2010 05:43:23 +0000 (21:43 -0800)]
Fix bug #7034 - vfs_cap causes signal 11 (SIGSEGV)

14 years agoFix two uses of strncat -> strlcat. Ensure proper use of strncpy when setting socket...
Jeremy Allison [Wed, 13 Jan 2010 05:18:36 +0000 (21:18 -0800)]
Fix two uses of strncat -> strlcat. Ensure proper use of strncpy when setting socket name.

Jeremy.

14 years agoMake file access control decisions a lot easier to debug (at level 10).
Jeremy Allison [Wed, 13 Jan 2010 05:06:09 +0000 (21:06 -0800)]
Make file access control decisions a lot easier to debug (at level 10).

Jeremy.

14 years agoFix bug #7033 - SMBrmdir call always returns true, even on failure to delete a directory.
Jeremy Allison [Wed, 13 Jan 2010 00:55:31 +0000 (16:55 -0800)]
Fix bug #7033 - SMBrmdir call always returns true, even on failure to delete a directory.

Argh. Missed the second (and essential) part of the fix for the above :-(.

Jeremy

14 years agoFix bug #7033 - SMBrmdir call always returns true, even on failure to delete a directory.
Jeremy Allison [Wed, 13 Jan 2010 00:50:40 +0000 (16:50 -0800)]
Fix bug #7033 - SMBrmdir call always returns true, even on failure to delete a directory.

There is a codepath missing to propagate back error returns from the rmdir
POSIX call inside close_directory when delete on close is set. This means doing
an rmdir on a Windows command line will always report success, even when the
directory was not deleted. This fix adds that codepath back into Samba.

Jeremy.

14 years agoFix bug #6876 - Delete of an object whose parent folder does not have delete rights...
Jeremy Allison [Wed, 13 Jan 2010 00:04:44 +0000 (16:04 -0800)]
Fix bug #6876 - Delete of an object whose parent folder does not have delete rights fails even if the delete right is set on the object.

Final fix for the vfs_acl_xattr and vfs_acl_tdb code.
Ensure we can delete a file even if the underlying POSIX
permissions don't allow it, if the Windows permissions do.

Jeremy.

14 years agoRemove the global char *LastDir.
Jeremy Allison [Tue, 12 Jan 2010 22:48:23 +0000 (14:48 -0800)]
Remove the global char *LastDir.

This is no longer used for anything.

14 years agos3: Lift the version of the scannedonly VFS module
Volker Lendecke [Tue, 12 Jan 2010 21:58:51 +0000 (22:58 +0100)]
s3: Lift the version of the scannedonly VFS module

14 years agos3: Add the "scannedonly" vfs virus scanner interface module
Olivier Sessink [Mon, 11 Jan 2010 20:53:37 +0000 (21:53 +0100)]
s3: Add the "scannedonly" vfs virus scanner interface module

14 years agoStrip trailing spaces
Simo Sorce [Tue, 12 Jan 2010 16:05:16 +0000 (11:05 -0500)]
Strip trailing spaces

14 years agodocs: fix xml tag in the pdbedit manpage
Michael Adam [Tue, 12 Jan 2010 11:29:14 +0000 (12:29 +0100)]
docs: fix xml tag in the pdbedit manpage

Michael

14 years agos3:auth: add comment to nulling out stolen sampass
Michael Adam [Tue, 12 Jan 2010 11:25:32 +0000 (12:25 +0100)]
s3:auth: add comment to nulling out stolen sampass

Adding this comment makes me think, I could also
have changed make_server_info_sam() talloc_move
instead of talloc_steal, but that would have
changed the signature... Well the comment is a
first step. :-)

Michael

14 years agos3: Remove a pointless if-statement
Volker Lendecke [Tue, 12 Jan 2010 12:53:01 +0000 (13:53 +0100)]
s3: Remove a pointless if-statement

14 years agos4-selftest: RPC-SAMR-PASSWORDS-BADPWDCOUNT fails against s4.
Günther Deschner [Tue, 12 Jan 2010 11:34:55 +0000 (12:34 +0100)]
s4-selftest: RPC-SAMR-PASSWORDS-BADPWDCOUNT fails against s4.

Seems like account lockout is not implemented at all yet.

Guenther

14 years agotestprogs: remove unused test_EachDriver from spoolss test.
Günther Deschner [Mon, 11 Jan 2010 14:40:37 +0000 (15:40 +0100)]
testprogs: remove unused test_EachDriver from spoolss test.

Guenther

14 years agotestprogs: add EnumPrinterKey test to spoolss test.
Günther Deschner [Mon, 11 Jan 2010 14:40:16 +0000 (15:40 +0100)]
testprogs: add EnumPrinterKey test to spoolss test.

Guenther

14 years agos4-smbtorture: fix GetAliasMembership test in RPC-SAMR.
Günther Deschner [Mon, 11 Jan 2010 11:40:01 +0000 (12:40 +0100)]
s4-smbtorture: fix GetAliasMembership test in RPC-SAMR.

Guenther

14 years agos4-smbtorture: add RPC-SAMR-PASSWORDS-BADPWDCOUNT torture test.
Günther Deschner [Mon, 11 Jan 2010 20:18:51 +0000 (21:18 +0100)]
s4-smbtorture: add RPC-SAMR-PASSWORDS-BADPWDCOUNT torture test.

This test checks the behavior (since w2k3 sp1) of the badPwdCount samr attribute
in relation to password history and successfull and unsucessful netlogon
samlogons.

Michael, please check. This should help verifiying Bug #4347.

Guenther

14 years agos4-smbtorture: allow test_SamLogon to test interactive samlogon in RPC-SAMR family...
Günther Deschner [Tue, 12 Jan 2010 10:48:23 +0000 (11:48 +0100)]
s4-smbtorture: allow test_SamLogon to test interactive samlogon in RPC-SAMR family of tests.

Guenther

14 years agos3-lanman: Allow a level2 descriptor for a level1 NetShareGetInfo
Giovanni Bajo [Mon, 11 Jan 2010 21:18:40 +0000 (22:18 +0100)]
s3-lanman: Allow a level2 descriptor for a level1 NetShareGetInfo

Windows seems to allow this

http://lists.samba.org/archive/samba-technical/2009-November/068116.html

has a dump of this.

14 years agoFix comment/debug messages
Simo Sorce [Mon, 11 Jan 2010 16:48:12 +0000 (11:48 -0500)]
Fix comment/debug messages

14 years agotestprogs: fix driver levels in spoolss test.
Günther Deschner [Mon, 11 Jan 2010 11:38:22 +0000 (12:38 +0100)]
testprogs: fix driver levels in spoolss test.

Guenther

14 years agotestprogs: test each printer, do not abort after first failure.
Günther Deschner [Mon, 11 Jan 2010 11:37:45 +0000 (12:37 +0100)]
testprogs: test each printer, do not abort after first failure.

Guenther

14 years agotestprogs: fix EnumPrintProcessorDatatypes spoolss test.
Günther Deschner [Mon, 11 Jan 2010 11:34:56 +0000 (12:34 +0100)]
testprogs: fix EnumPrintProcessorDatatypes spoolss test.

Guenther

14 years agotestprogs: add more error codes to spoolss test.
Günther Deschner [Mon, 11 Jan 2010 11:33:22 +0000 (12:33 +0100)]
testprogs: add more error codes to spoolss test.

Guenther

14 years agos3: Add a zfsacl:denymissingspecial parameter
Volker Lendecke [Mon, 11 Jan 2010 11:10:47 +0000 (12:10 +0100)]
s3: Add a zfsacl:denymissingspecial parameter

When setting an ACL without any of the user/group/other entries, ZFS
automatically creates them. This can at times confuse users a lot. This
parameter denies setting such an acl, users explicitly have to for example set
an ACE with everyone allowing nothing. Users need to be educated about this,
but this helps avoid a lot of confusion.

14 years agotsocket: Added doxygen config file.
Andreas Schneider [Mon, 21 Dec 2009 11:44:00 +0000 (12:44 +0100)]
tsocket: Added doxygen config file.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
14 years agotsocket: Added complete doxygen documentation.
Andreas Schneider [Wed, 16 Dec 2009 12:26:48 +0000 (13:26 +0100)]
tsocket: Added complete doxygen documentation.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
14 years agoMerge remote branch 'origin/master' into alpha11release
Andrew Bartlett [Mon, 11 Jan 2010 06:10:32 +0000 (17:10 +1100)]
Merge remote branch 'origin/master' into alpha11release

14 years agoand we move on towards Samba4 alpha12!
Andrew Bartlett [Mon, 11 Jan 2010 06:05:06 +0000 (17:05 +1100)]
and we move on towards Samba4 alpha12!

14 years agomore WHATSNEW4 samba-4.0.0alpha11
Andrew Bartlett [Mon, 11 Jan 2010 05:35:28 +0000 (16:35 +1100)]
more WHATSNEW4

14 years agoThis is Samba4 alpha11!
Andrew Bartlett [Mon, 11 Jan 2010 03:57:55 +0000 (14:57 +1100)]
This is Samba4 alpha11!

14 years agorelease notes for Samba4 alpha11 (to be released this week)
Andrew Bartlett [Mon, 11 Jan 2010 01:00:34 +0000 (12:00 +1100)]
release notes for Samba4 alpha11 (to be released this week)

14 years agos4/torture: Parameterize output in LOCK tests based off server support
Steven Danneman [Mon, 11 Jan 2010 00:06:57 +0000 (16:06 -0800)]
s4/torture: Parameterize output in LOCK tests based off server support

Two new torture parameters:

* smbexit_pdu_support: if the Server supports the Exit command

* range_not_locked_on_file_close: whether the server returns the
  NT_STATUS_RANGE_NOT_LOCKED error when a file is closed which has a
  pending lock request.  Windows returns this error, though per the
  spec, this error should only be returned to an unlock request.

14 years agoRevert "s4:provision_users.ldif - Import all essential groups for Windows Server...
Andrew Tridgell [Sun, 10 Jan 2010 23:08:30 +0000 (10:08 +1100)]
Revert "s4:provision_users.ldif - Import all essential groups for Windows Server 2008 mode"

This reverts commit 5c174c68ccba7506147feab1d09ad676792139b3.

This series of commits broke 'make test'.

Matthias, please make sure you run a _full_ make test before every
push.

14 years agoRevert "s4:provision_users.ldif - Remove foreign security principal S-1-5-17 for...
Andrew Tridgell [Sun, 10 Jan 2010 23:07:53 +0000 (10:07 +1100)]
Revert "s4:provision_users.ldif - Remove foreign security principal S-1-5-17 for now"

This reverts commit 61dfd3dc1dce2c0dd6693de80930af312ad3e39f.

This series of commits broke 'make test'.

Matthias, please make sure you run a _full_ make test before every
push.

14 years agoRevert "s4:provision_users.ldif - Fix memberships regarding the denied password RODC...
Andrew Tridgell [Sun, 10 Jan 2010 23:06:58 +0000 (10:06 +1100)]
Revert "s4:provision_users.ldif - Fix memberships regarding the denied password RODC replication group"

This reverts commit 9ee895fcf6327b1c2f5ee09fa565bd62974e9c58.

This series of commits broke 'make test'.

Matthias, please make sure you run a _full_ make test before every
push.

14 years agoRevert "s4:provision_users.ldif - Add objects for IIS"
Andrew Tridgell [Sun, 10 Jan 2010 23:05:50 +0000 (10:05 +1100)]
Revert "s4:provision_users.ldif - Add objects for IIS"

This reverts commit 91e210028790397996659116446e6add452707f6.

This series of commits broke 'make test'.

Matthias, please make sure you run a _full_ make test before every
push.

14 years agos4-selftest: when a command fails show both normal and expanded command
Andrew Tridgell [Sun, 10 Jan 2010 22:36:48 +0000 (09:36 +1100)]
s4-selftest: when a command fails show both normal and expanded command

It is sometimes hard to tell which varient of something like
$SMB_CONF_PATH or $USERNAME is being used in a test. By giving both
the expanded command ($command with environment variables expanded)
and non-expanded command it is easier to reproduce bugs outside the
test environment.

14 years agos4-test: fixed make test without having done make install
Andrew Tridgell [Sun, 10 Jan 2010 22:29:29 +0000 (09:29 +1100)]
s4-test: fixed make test without having done make install

client.conf didn't specify "setup directory"

14 years agos4:upgradeprovision - fix up the script regarding linked attributes
Matthias Dieter Wallnöfer [Sun, 10 Jan 2010 20:34:05 +0000 (21:34 +0100)]
s4:upgradeprovision - fix up the script regarding linked attributes

We have to try to add new objects until between two iterations we didn't make
any progress. Either we are then done (no objects remaining) or we are
incapable to do this fully automatically.

The latter can happen if important system objects (builtin groups, users...)
moved (e.g. consider one of my recent comments). Then the new object can't be
added if it contains the same "sAMAccountName" attribute as the old one. We
have to let the user delete the old one (also to give him a chance to backup
personal changes - if needed) and only then the script is capable to add the
new one onto the right place. Make this clear with an exhaustive error output.

I personally don't see a good way how to do this better for now so I would leave
this as a manual step.

14 years agos4:upgradeprovision - Reformat comments
Matthias Dieter Wallnöfer [Sun, 10 Jan 2010 19:08:50 +0000 (20:08 +0100)]
s4:upgradeprovision - Reformat comments

Make them break at line 80 (better readability).

14 years agos4:repl_meta_data - Transform a "1" into a "true" on a boolean variable
Matthias Dieter Wallnöfer [Sun, 10 Jan 2010 18:49:40 +0000 (19:49 +0100)]
s4:repl_meta_data - Transform a "1" into a "true" on a boolean variable

14 years agos4:provision_users.ldif - Add objects for IIS
Matthias Dieter Wallnöfer [Sun, 10 Jan 2010 13:20:09 +0000 (14:20 +0100)]
s4:provision_users.ldif - Add objects for IIS

Some WSPP locations point out that they're defacto-standards for Windows Server deployments starting with 2008. So we should add them to s4 too.

14 years agos4:provision_self_join_modify.ldif - Point out that account "dns" is s4 specific
Matthias Dieter Wallnöfer [Sun, 10 Jan 2010 14:43:07 +0000 (15:43 +0100)]
s4:provision_self_join_modify.ldif - Point out that account "dns" is s4 specific

14 years agos4:provision_users.ldif - Fix memberships regarding the denied password RODC replicat...
Matthias Dieter Wallnöfer [Sun, 10 Jan 2010 14:38:55 +0000 (15:38 +0100)]
s4:provision_users.ldif - Fix memberships regarding the denied password RODC replication group

14 years agos3: Remove some unused variables
Volker Lendecke [Sun, 10 Jan 2010 21:42:02 +0000 (22:42 +0100)]
s3: Remove some unused variables

14 years agos3: Fix some nonempty blank lines
Volker Lendecke [Sat, 9 Jan 2010 19:26:46 +0000 (20:26 +0100)]
s3: Fix some nonempty blank lines

14 years agos3: Use sid_check_is_domain instead of a direct sid_equal
Volker Lendecke [Sun, 10 Jan 2010 16:58:12 +0000 (17:58 +0100)]
s3: Use sid_check_is_domain instead of a direct sid_equal

14 years agos3: Use sid_check_is_in_our_domain instead of a direct sid_peek_check_rid
Volker Lendecke [Sun, 10 Jan 2010 16:57:00 +0000 (17:57 +0100)]
s3: Use sid_check_is_in_our_domain instead of a direct sid_peek_check_rid

14 years agos3: Replace most calls to sid_append_rid() by sid_compose()
Volker Lendecke [Sun, 10 Jan 2010 16:39:27 +0000 (17:39 +0100)]
s3: Replace most calls to sid_append_rid() by sid_compose()

14 years agos3: Remove unused samr_make_sam_obj_sd
Volker Lendecke [Sun, 10 Jan 2010 16:30:54 +0000 (17:30 +0100)]
s3: Remove unused samr_make_sam_obj_sd

14 years agos3: Remove the typedef for "auth_serversupplied_info"
Volker Lendecke [Sun, 10 Jan 2010 13:24:22 +0000 (14:24 +0100)]
s3: Remove the typedef for "auth_serversupplied_info"

14 years agos3: Remove the typedef for "auth_usersupplied_info"
Volker Lendecke [Sun, 10 Jan 2010 13:16:04 +0000 (14:16 +0100)]
s3: Remove the typedef for "auth_usersupplied_info"

14 years agos3: Trim libnss_wins.so
Volker Lendecke [Sun, 10 Jan 2010 12:35:37 +0000 (13:35 +0100)]
s3: Trim libnss_wins.so

14 years agos3: Trim down some utilities a bit
Volker Lendecke [Sun, 3 Jan 2010 21:28:33 +0000 (22:28 +0100)]
s3: Trim down some utilities a bit

14 years agos4:provision_users.ldif - Remove foreign security principal S-1-5-17 for now
Matthias Dieter Wallnöfer [Sun, 10 Jan 2010 10:07:16 +0000 (11:07 +0100)]
s4:provision_users.ldif - Remove foreign security principal S-1-5-17 for now

This belongs to the AD IIS stuff where I don't know yet if we should import it.

14 years agos4:provision_users.ldif - Import all essential groups for Windows Server 2008 mode
Matthias Dieter Wallnöfer [Sun, 10 Jan 2010 09:47:30 +0000 (10:47 +0100)]
s4:provision_users.ldif - Import all essential groups for Windows Server 2008 mode

Additionally I had to fix some bugs (especially wrong "groupTypes") and
reordered the objects using the SID (this is easier when enhancing the file).

14 years agos4-ldb: display security descriptors with correct SDL for known SIDs
Andrew Tridgell [Sun, 10 Jan 2010 01:53:07 +0000 (12:53 +1100)]
s4-ldb: display security descriptors with correct SDL for known SIDs

This makes it much easier to compare SDs

14 years agos4-dsdb: added samdb_domain_sid_cache_only()
Andrew Tridgell [Sun, 10 Jan 2010 01:52:22 +0000 (12:52 +1100)]
s4-dsdb: added samdb_domain_sid_cache_only()

14 years agos3: Remove a pointless "else" branch from add_ccache_to_list()
Volker Lendecke [Sat, 9 Jan 2010 17:43:38 +0000 (18:43 +0100)]
s3: Remove a pointless "else" branch from add_ccache_to_list()

14 years agos3: Slightly simplify winbindd_store_creds
Volker Lendecke [Sat, 9 Jan 2010 18:01:35 +0000 (19:01 +0100)]
s3: Slightly simplify winbindd_store_creds

14 years agos3: Fix a segfault in winbindd_dual_ccache_ntlm_auth()
Volker Lendecke [Sat, 9 Jan 2010 19:22:00 +0000 (20:22 +0100)]
s3: Fix a segfault in winbindd_dual_ccache_ntlm_auth()

ntlmssp_update allocates the reply_blob as a child of ntlmssp_state. This means
with ntlmss_end() it will be gone. winbindd_dual_ccache_ntlm_auth used the blob
after the ntlmssp_end().

14 years agos4-drs: instanceType is always sent, regardless of UDV values
Andrew Tridgell [Sat, 9 Jan 2010 11:08:25 +0000 (22:08 +1100)]
s4-drs: instanceType is always sent, regardless of UDV values

14 years agos4-debug: lower the verbosity of a couple of common log messages
Andrew Tridgell [Sat, 9 Jan 2010 10:43:16 +0000 (21:43 +1100)]
s4-debug: lower the verbosity of a couple of common log messages

14 years agos4-samldb: fixed primaryGroupID when promoting a machine to a DC
Andrew Tridgell [Sat, 9 Jan 2010 09:58:07 +0000 (20:58 +1100)]
s4-samldb: fixed primaryGroupID when promoting a machine to a DC

The machine gets a primaryGroupID of DOMAIN_RID_DCS. This is done
without changing the member attributes of its groups.

14 years agos4-schema: fixed the SDDL for the schema root security descriptor
Andrew Tridgell [Sat, 9 Jan 2010 09:54:16 +0000 (20:54 +1100)]
s4-schema: fixed the SDDL for the schema root security descriptor

This was preventing a DCPROMO client from allowing outgoing
replication

14 years agos4-drs: add a local UDV entry even when no replUpToDateVector present on NC
Andrew Tridgell [Sat, 9 Jan 2010 09:53:27 +0000 (20:53 +1100)]
s4-drs: add a local UDV entry even when no replUpToDateVector present on NC

This allows us to filter correctly for a NC that we have created but
not pulled from anyone.

14 years agos4-drs: give DN of failed replication partition
Andrew Tridgell [Sat, 9 Jan 2010 09:42:23 +0000 (20:42 +1100)]
s4-drs: give DN of failed replication partition

14 years agos4-drs: base is_nc_prefix on instanceType
Andrew Tridgell [Sat, 9 Jan 2010 07:50:30 +0000 (18:50 +1100)]
s4-drs: base is_nc_prefix on instanceType

for extended operations comparing to the ncRoot_dn is not correct

14 years agos4-drs: having no SPNs to change is not an error
Andrew Tridgell [Sat, 9 Jan 2010 07:10:38 +0000 (18:10 +1100)]
s4-drs: having no SPNs to change is not an error

14 years agos4-drs: fixed writespn to ignore add/delete errors
Andrew Tridgell [Sat, 9 Jan 2010 06:42:59 +0000 (17:42 +1100)]
s4-drs: fixed writespn to ignore add/delete errors

When a SPN is added and already exists, it is ignored. Similarly, when
a SPN is deleted and doesn't exist, it is ignored.

14 years agos4-dsdb: added samdb_ldb_val_case_cmp()
Andrew Tridgell [Sat, 9 Jan 2010 06:42:05 +0000 (17:42 +1100)]
s4-dsdb: added samdb_ldb_val_case_cmp()

14 years agos4-drs: moved the DsWriteAccountSpn call to its own file
Andrew Tridgell [Sat, 9 Jan 2010 06:11:01 +0000 (17:11 +1100)]
s4-drs: moved the DsWriteAccountSpn call to its own file

14 years agos4-libnet: dsdb_wellknown_dn() in vampire code
Andrew Tridgell [Sat, 9 Jan 2010 04:57:52 +0000 (15:57 +1100)]
s4-libnet: dsdb_wellknown_dn() in vampire code

14 years agos4-drs: need to set the getncchanges extended_ret on success too
Andrew Tridgell [Sat, 9 Jan 2010 04:12:18 +0000 (15:12 +1100)]
s4-drs: need to set the getncchanges extended_ret on success too

14 years agos4-drs: calculate and send a uptodateness_vector with replication requests
Andrew Tridgell [Sat, 9 Jan 2010 03:29:39 +0000 (14:29 +1100)]
s4-drs: calculate and send a uptodateness_vector with replication requests

This stops us getting objects changes twice if they came via an
indirect path.

14 years agos4-drs: be less verbose when we filter objects by UDV
Andrew Tridgell [Sat, 9 Jan 2010 03:28:00 +0000 (14:28 +1100)]
s4-drs: be less verbose when we filter objects by UDV

14 years agos4-drs: added filtering by udv in getncchanges
Andrew Tridgell [Sat, 9 Jan 2010 02:11:27 +0000 (13:11 +1100)]
s4-drs: added filtering by udv in getncchanges

When a client supplied an uptodateness_vector, we can use it to filter
what objects we return. This greatly reduces the amount of replication
traffic between DCs.

14 years agos4-idl: give a enum for attribute cn and a 'NONE' attribute
Andrew Tridgell [Sat, 9 Jan 2010 02:10:28 +0000 (13:10 +1100)]
s4-idl: give a enum for attribute cn and a 'NONE' attribute

The 'NONE' attribute has value 0xFFFFFFFF. Adding this ensures the
compiler will complain if it is set to use 16 bit enums. We rely on
being able to store 32 bits in an attid enum.

14 years agos4-drs: fixed the NC in the getncchanges RID alloc reply
Andrew Tridgell [Fri, 8 Jan 2010 23:12:54 +0000 (10:12 +1100)]
s4-drs: fixed the NC in the getncchanges RID alloc reply

the search happens on a different DN to the NC of the request, but the
reply is with the original NC

14 years agos4-debug: removed debug_ctx(). It didn't catch on :-)
Andrew Tridgell [Fri, 8 Jan 2010 22:05:56 +0000 (09:05 +1100)]
s4-debug: removed debug_ctx(). It didn't catch on :-)

There was only one user, which isn't worth it for the overhead.

14 years agos4-messaging: remove only usage of debug_ctx()
Andrew Tridgell [Fri, 8 Jan 2010 22:05:29 +0000 (09:05 +1100)]
s4-messaging: remove only usage of debug_ctx()

14 years agos4-messaging: fixed a memory leak in messaging_path()
Andrew Tridgell [Fri, 8 Jan 2010 22:04:18 +0000 (09:04 +1100)]
s4-messaging: fixed a memory leak in messaging_path()

It is a bit convoluted to fix, as cluster_id_string() may return a
const string.

14 years agos4-drs: fixed usage of ldb_dn_new()
Andrew Tridgell [Fri, 8 Jan 2010 22:03:45 +0000 (09:03 +1100)]
s4-drs: fixed usage of ldb_dn_new()

14 years agos4-ldb: validate the type of the ldb argument to ldb_dn_new()
Andrew Tridgell [Fri, 8 Jan 2010 22:03:08 +0000 (09:03 +1100)]
s4-ldb: validate the type of the ldb argument to ldb_dn_new()

It has been a common bug to get the first two arguments the wrong way
around

14 years agoFix comment
Simo Sorce [Fri, 8 Jan 2010 21:53:30 +0000 (16:53 -0500)]
Fix comment

14 years agoRe-fix bug 5202 - cannot change ACLs on writable file with "dos filemode=yes"
Jeremy Allison [Fri, 8 Jan 2010 18:17:46 +0000 (10:17 -0800)]
Re-fix bug 5202 - cannot change ACLs on writable file with "dos filemode=yes"

This bug re-occurred for 3.3.x and above.

The reason is that to change a NT ACL we now have to open the file requesting
WRITE_DAC and WRITE_OWNER access. The mapping from POSIX "w" to NT permissions
in posix_acls doesn't add these bits when "dos filemode = yes", so even though
the permission or owner change would be allowed by the POSIX ACL code, the
NTCreateX call fails with ACCESS_DENIED now we always check NT permissions
first.

Added in the mapping from "w" to WRITE_DAC and WRITE_OWNER access.

Jeremy.

14 years agos4:provision_self_join.ldif - Adapt comment after implementation of distributed RIDs
Matthias Dieter Wallnöfer [Fri, 8 Jan 2010 17:18:21 +0000 (18:18 +0100)]
s4:provision_self_join.ldif - Adapt comment after implementation of distributed RIDs

14 years agos4-kdc: Migrate tcp connections to tsocket.
Andreas Schneider [Thu, 7 Jan 2010 11:23:33 +0000 (12:23 +0100)]
s4-kdc: Migrate tcp connections to tsocket.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
14 years agos4:kdc: use LIBSAMBA_TSOCKET
Stefan Metzmacher [Wed, 4 Nov 2009 18:27:20 +0000 (19:27 +0100)]
s4:kdc: use LIBSAMBA_TSOCKET

metze

14 years agos4:kdc: the ->process function returns "bool"
Stefan Metzmacher [Fri, 8 Jan 2010 10:45:59 +0000 (11:45 +0100)]
s4:kdc: the ->process function returns "bool"

metze

14 years agolibcli/util: add tstream_read_pdu_blob_send/recv
Stefan Metzmacher [Thu, 5 Nov 2009 08:55:12 +0000 (09:55 +0100)]
libcli/util: add tstream_read_pdu_blob_send/recv

This will take the some full_request callback function
as the Samba4 packet code.

metze

14 years agos3-time: fix build warnings after we moved to shared time functions.
Günther Deschner [Fri, 8 Jan 2010 10:03:31 +0000 (11:03 +0100)]
s3-time: fix build warnings after we moved to shared time functions.

Bjoern, please check.

Guenther

14 years agos3-docs: mention -K option in pdbedit manpage.
Günther Deschner [Fri, 8 Jan 2010 09:38:46 +0000 (10:38 +0100)]
s3-docs: mention -K option in pdbedit manpage.

Guenther