samba.git
9 years agos3-secrets: only include secrets.h when needed.
Günther Deschner [Thu, 5 Aug 2010 00:25:37 +0000 (02:25 +0200)]
s3-secrets: only include secrets.h when needed.

Guenther

9 years agos4-dns: ndr_dnsp helper functions
Andrew Tridgell [Tue, 11 May 2010 05:35:43 +0000 (07:35 +0200)]
s4-dns: ndr_dnsp helper functions

these handle the dnsp_name type

9 years agos4-dns: fix dnsp for old build
Andrew Tridgell [Thu, 5 Aug 2010 05:25:38 +0000 (15:25 +1000)]
s4-dns: fix dnsp for old build

9 years agos4-dns: expanded the DNSP code to handle more record type
Andrew Tridgell [Tue, 11 May 2010 05:09:46 +0000 (07:09 +0200)]
s4-dns: expanded the DNSP code to handle more record type

9 years agopidl: added a new type dnsp_name
Andrew Tridgell [Tue, 11 May 2010 05:09:05 +0000 (07:09 +0200)]
pidl: added a new type dnsp_name

This is a name type used in DNS where each DNS component maps to a
length byte followed by the string

We want these to map to a char *, which is why we need to do this in
PIDL

9 years agos4-ldb: added ldif handler for the dnsRecord attribute
Andrew Tridgell [Thu, 5 Aug 2010 05:22:49 +0000 (15:22 +1000)]
s4-ldb: added ldif handler for the dnsRecord attribute

9 years agodns: first version of dsnRecord parser
Andrew Tridgell [Thu, 5 Aug 2010 05:20:13 +0000 (15:20 +1000)]
dns: first version of dsnRecord parser

9 years agos3: avoid global include of ads.h.
Günther Deschner [Thu, 1 Jul 2010 22:32:52 +0000 (00:32 +0200)]
s3: avoid global include of ads.h.

Guenther

9 years agos3-libnet_join: add ads.h as helper header to IDL.
Günther Deschner [Wed, 4 Aug 2010 14:49:43 +0000 (16:49 +0200)]
s3-libnet_join: add ads.h as helper header to IDL.

Guenther

9 years agos3-printing: move AD related printing components to an own file.
Günther Deschner [Wed, 4 Aug 2010 13:31:08 +0000 (15:31 +0200)]
s3-printing: move AD related printing components to an own file.

Guenther

9 years agos3: Convert cli_list() to return NTSTATUS
Volker Lendecke [Mon, 2 Aug 2010 17:22:22 +0000 (19:22 +0200)]
s3: Convert cli_list() to return NTSTATUS

If needed, the callback functions can count themselves

9 years agos3: Use data_blob_null
Volker Lendecke [Mon, 2 Aug 2010 13:25:45 +0000 (15:25 +0200)]
s3: Use data_blob_null

9 years agos4:LDB modules - remove the "kludge_acl" module code
Matthias Dieter Wallnöfer [Wed, 4 Aug 2010 17:43:59 +0000 (19:43 +0200)]
s4:LDB modules - remove the "kludge_acl" module code

Obviously this has been forgotten by Nadya.

9 years agos3-waf: fix the build.
Günther Deschner [Wed, 4 Aug 2010 12:55:10 +0000 (14:55 +0200)]
s3-waf: fix the build.

Guenther

9 years agos4-dsdb: Removed kludge_acl as it is no longer necessary
Nadezhda Ivanova [Wed, 4 Aug 2010 12:22:17 +0000 (15:22 +0300)]
s4-dsdb: Removed kludge_acl as it is no longer necessary

Moved the access check on extended operations to acl module and removed kludge_acl

9 years agosmall optimizations for shadowcopy2 module
Christian Ambach [Wed, 21 Apr 2010 13:10:39 +0000 (15:10 +0200)]
small optimizations for shadowcopy2 module

- remove a compiler warning
- replace strlen usage

Signed-off-by: Christian Ambach <christian.ambach@de.ibm.com>
9 years agofix snapshot content display with hide unreadable
Christian Ambach [Wed, 14 Apr 2010 08:05:56 +0000 (10:05 +0200)]
fix snapshot content display with hide unreadable

With the hide unreadable option set, snapshots are be displayed
as empty with shadow_copy2 and a NFSv4 ACL module.

To prevent multiple conversions of the paths when the acl call
does a VFS_STAT (as the nfs4acl code does), a check was added
to convert_shadow2_name() so it will not touch paths any more
that look like they have already been converted.

Signed-off-by: Christian Ambach <christian.ambach@de.ibm.com>
9 years agos3: Fix shadow copies after the change for in-path @GMT-
Volker Lendecke [Thu, 12 Nov 2009 11:30:58 +0000 (12:30 +0100)]
s3: Fix shadow copies after the change for in-path @GMT-

Signed-off-by: Christian Ambach <christian.ambach@de.ibm.com>
9 years agos3: Fix in-path shadowcopy2 module
Volker Lendecke [Mon, 9 Nov 2009 10:06:30 +0000 (11:06 +0100)]
s3: Fix in-path shadowcopy2 module

Signed-off-by: Christian Ambach <christian.ambach@de.ibm.com>
9 years agomake scannedonly notify the scanner if no .scanned: file was found during rename
olivier [Mon, 2 Aug 2010 22:20:11 +0000 (00:20 +0200)]
make scannedonly notify the scanner if no .scanned: file was found during rename

9 years agos3-dcerpc: fix some uninitialized variables build warnings.
Günther Deschner [Tue, 3 Aug 2010 13:55:20 +0000 (15:55 +0200)]
s3-dcerpc: fix some uninitialized variables build warnings.

Guenther

9 years agos3-spoolss: fix type-punned pointer build warning.
Günther Deschner [Tue, 3 Aug 2010 13:13:57 +0000 (15:13 +0200)]
s3-spoolss: fix type-punned pointer build warning.

Guenther

9 years agos3-spoolss: fix uninitialized variable build warning.
Günther Deschner [Tue, 3 Aug 2010 12:51:27 +0000 (14:51 +0200)]
s3-spoolss: fix uninitialized variable build warning.

Andreas, please check.

Guenther

9 years agos3-printing: make sure to set NTSTATUS code before checking it.
Günther Deschner [Tue, 3 Aug 2010 11:18:01 +0000 (13:18 +0200)]
s3-printing: make sure to set NTSTATUS code before checking it.

Andreas, please check.

Guenther

9 years agos4/vampire: update dNSHostName for the Server object
Anatoliy Atanasov [Mon, 2 Aug 2010 13:47:38 +0000 (16:47 +0300)]
s4/vampire: update dNSHostName for the Server object

9 years agos4-schema: More verbose error log when subClassOf is not found in schema
Kamen Mazdrashki [Tue, 3 Aug 2010 01:26:57 +0000 (04:26 +0300)]
s4-schema: More verbose error log when subClassOf is not found in schema

Error message show failing classSchema object
but not the specific value for the failure,
which makes diagnostics by log files really hard.

9 years agos4: fix comment typos
Kamen Mazdrashki [Tue, 3 Aug 2010 01:14:45 +0000 (04:14 +0300)]
s4: fix comment typos

9 years agos3-spoolss: remove duplicate (and incorrect) header.
Günther Deschner [Mon, 2 Aug 2010 22:12:27 +0000 (00:12 +0200)]
s3-spoolss: remove duplicate (and incorrect) header.

Guenther

9 years agoidl: Fixed a possible crash bug.
Andreas Schneider [Mon, 2 Aug 2010 16:32:40 +0000 (18:32 +0200)]
idl: Fixed a possible crash bug.

9 years agos4:librpc/rpc: avoid using DCERPC_NDR_REF_ALLOC for epm_Map()
Stefan Metzmacher [Mon, 2 Aug 2010 08:21:36 +0000 (10:21 +0200)]
s4:librpc/rpc: avoid using DCERPC_NDR_REF_ALLOC for epm_Map()

metze

9 years agos4:torture/libnet: avoid dereferencing dcerpc_binding_handle internals
Stefan Metzmacher [Mon, 2 Aug 2010 07:53:27 +0000 (09:53 +0200)]
s4:torture/libnet: avoid dereferencing dcerpc_binding_handle internals

metze

9 years agos3-spoolss: Fixed a segfault if a value has no data.
Andreas Schneider [Mon, 2 Aug 2010 11:49:59 +0000 (13:49 +0200)]
s3-spoolss: Fixed a segfault if a value has no data.

9 years agos3-spoolss: Move some debug message to a higher level.
Andreas Schneider [Mon, 2 Aug 2010 11:21:47 +0000 (13:21 +0200)]
s3-spoolss: Move some debug message to a higher level.

9 years agos3-spoolss: Fixed some C++ build warnings.
Andreas Schneider [Mon, 2 Aug 2010 10:38:57 +0000 (12:38 +0200)]
s3-spoolss: Fixed some C++ build warnings.

9 years agos3-spoolss: Use a stackframe to allocat memory.
Andreas Schneider [Mon, 2 Aug 2010 10:09:03 +0000 (12:09 +0200)]
s3-spoolss: Use a stackframe to allocat memory.

This is faster and should be more secure to use if something goes wrong.

9 years agos4-smbtorture: remove another old and now invalid s3 special case during spoolss...
Günther Deschner [Mon, 2 Aug 2010 14:08:30 +0000 (16:08 +0200)]
s4-smbtorture: remove another old and now invalid s3 special case during spoolss testing.

Guenther

9 years agospoolss: fix potential crash bug in spoolss_PrinterEnumValues push path.
Günther Deschner [Mon, 2 Aug 2010 13:52:09 +0000 (15:52 +0200)]
spoolss: fix potential crash bug in spoolss_PrinterEnumValues push path.

Guenther

9 years agos3-spoolss: remove unused MAX_OPEN_PRINTER_EXS define.
Günther Deschner [Mon, 2 Aug 2010 13:31:10 +0000 (15:31 +0200)]
s3-spoolss: remove unused MAX_OPEN_PRINTER_EXS define.

Guenther

9 years agos3-build: fix some c++ build warnings.
Günther Deschner [Sun, 1 Aug 2010 13:34:52 +0000 (15:34 +0200)]
s3-build: fix some c++ build warnings.

Guenther

9 years agos3: Use cups-config --libs
Volker Lendecke [Mon, 2 Aug 2010 09:03:13 +0000 (11:03 +0200)]
s3: Use cups-config --libs

9 years agos4:ldap.py - remove superflous spaces
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 20:12:04 +0000 (22:12 +0200)]
s4:ldap.py - remove superflous spaces

Sorry, forgot to delete them in the last commit

9 years agos4:ldap.py - additional "instanceType" checks
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 19:12:50 +0000 (21:12 +0200)]
s4:ldap.py - additional "instanceType" checks

9 years agos4:instancetype LDB module - add checks requested by MS-ADTS 3.1.1.5.2.2
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 15:34:43 +0000 (17:34 +0200)]
s4:instancetype LDB module - add checks requested by MS-ADTS 3.1.1.5.2.2

We've to test for the WRITE flag if we are performing an NC add. And if it
isn't an NC add then only the WRITE or no flag is allowed.

9 years agos4:objectclass LDB module - consider the "instanceType" when adding NCs
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 15:36:11 +0000 (17:36 +0200)]
s4:objectclass LDB module - consider the "instanceType" when adding NCs

This is requested by MS-ADTS 3.1.1.5.2.2 (NC add operation).

9 years agos4:descriptor LDB module - remove the "forest DN" check
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 15:02:45 +0000 (17:02 +0200)]
s4:descriptor LDB module - remove the "forest DN" check

Also here we have to work with the default base DN.

After some reading I've discovered that this isn't really true. The forest
partition does exist on one or more DCs and is there the same as the default
base DN (which is already checked by the module).
And if we have other DCs which contain child domains then they never contain
data of the forest domain beside the schema and the configuration partition
(which are checked anyway) since a DC can always contain only one domain!

Link: http://www.informit.com/articles/article.aspx?p=26896&seqNum=5
9 years agos4:acl LDB module - remove the "forest DN" check
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 15:02:45 +0000 (17:02 +0200)]
s4:acl LDB module - remove the "forest DN" check

After some reading I've discovered that this isn't really true. The forest
partition does exist on one or more DCs and is there the same as the default
base DN (which is already checked by the module).
And if we have other DCs which contain child domains then they never contain
data of the forest domain beside the schema and the configuration partition
(which are checked anyway) since a DC can always contain only one domain!

Link: http://www.informit.com/articles/article.aspx?p=26896&seqNum=5
9 years agos4:acl LDB module - remove unused call "is_root_base_dn"
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 14:39:45 +0000 (16:39 +0200)]
s4:acl LDB module - remove unused call "is_root_base_dn"

9 years agos4:urgent_replication.py test - adapt the test for the harder delete restrictions
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 15:44:50 +0000 (17:44 +0200)]
s4:urgent_replication.py test - adapt the test for the harder delete restrictions

Otherwise we are not able to delete the "test crossRef" object which points
to the default NC anymore.

9 years agos4:ldap.py - perform tests on the additional delete constraint checks
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 11:09:46 +0000 (13:09 +0200)]
s4:ldap.py - perform tests on the additional delete constraint checks

9 years agos4:objectclass LDB module - implement additional delete constraint checks
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 10:50:56 +0000 (12:50 +0200)]
s4:objectclass LDB module - implement additional delete constraint checks

MS-ADTS 3.1.1.5.5.3

9 years agos3: Fix an uninitialized variable
Volker Lendecke [Sun, 1 Aug 2010 18:15:57 +0000 (20:15 +0200)]
s3: Fix an uninitialized variable

9 years agos3: Fix a C++ warning
Volker Lendecke [Sun, 1 Aug 2010 18:15:39 +0000 (20:15 +0200)]
s3: Fix a C++ warning

9 years agos3: Fix some nonempty blank lines
Volker Lendecke [Mon, 19 Jul 2010 19:00:31 +0000 (21:00 +0200)]
s3: Fix some nonempty blank lines

9 years agos3: Fix an uninitialized variable
Volker Lendecke [Sun, 1 Aug 2010 17:55:06 +0000 (19:55 +0200)]
s3: Fix an uninitialized variable

9 years agos4:ldap.py - add a test for "CN=System" object rename behaviour
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 09:26:08 +0000 (11:26 +0200)]
s4:ldap.py - add a test for "CN=System" object rename behaviour

9 years agos4:subtree_rename LDB module - rename "check_system_flags" into "check_constraints...
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 09:06:18 +0000 (11:06 +0200)]
s4:subtree_rename LDB module - rename "check_system_flags" into "check_constraints" and perform more checks

Always considering MS-ADTS 3.1.1.5.4.1.2.

9 years agos4:subtree_rename LDB module - introduce out of memory checks
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 09:05:26 +0000 (11:05 +0200)]
s4:subtree_rename LDB module - introduce out of memory checks

9 years agos4:dsdb/samdb/ldb_modules/util.c - remove unused variables
Matthias Dieter Wallnöfer [Sun, 1 Aug 2010 09:32:14 +0000 (11:32 +0200)]
s4:dsdb/samdb/ldb_modules/util.c - remove unused variables

9 years agos4:torture/rpc/winreg: fix "strict aliasing" compiler warnings
Stefan Metzmacher [Sun, 1 Aug 2010 09:18:05 +0000 (11:18 +0200)]
s4:torture/rpc/winreg: fix "strict aliasing" compiler warnings

metze

9 years agos4:torture/rpc/winreg: fix the build with non gcc compilers (e.g. on solaris)
Stefan Metzmacher [Sun, 1 Aug 2010 09:17:14 +0000 (11:17 +0200)]
s4:torture/rpc/winreg: fix the build with non gcc compilers (e.g. on solaris)

metze

9 years agos4:ldap.py - performs some "systemFlags" testing
Matthias Dieter Wallnöfer [Fri, 9 Jul 2010 17:22:01 +0000 (19:22 +0200)]
s4:ldap.py - performs some "systemFlags" testing

9 years agos4:subtree_rename LDB module - introduce the "systemFlags" protection rules
Matthias Dieter Wallnöfer [Sun, 4 Jul 2010 18:07:09 +0000 (20:07 +0200)]
s4:subtree_rename LDB module - introduce the "systemFlags" protection rules

This is done in a dedicated call "check_system_flags".

9 years agos4:dsdb/pydsdb.c - import "systemFlags" into Python
Matthias Dieter Wallnöfer [Thu, 8 Jul 2010 19:05:33 +0000 (21:05 +0200)]
s4:dsdb/pydsdb.c - import "systemFlags" into Python

Needed by ldap.py tests

9 years agos4:subtree_rename LDB module - "subren_ctx_init" - fix the "out of memory" return
Matthias Dieter Wallnöfer [Sat, 31 Jul 2010 19:26:38 +0000 (21:26 +0200)]
s4:subtree_rename LDB module - "subren_ctx_init" - fix the "out of memory" return

9 years agos4:torture/winreg.c - fix warnings of Solaris 10 cc through the appropriate casts
Matthias Dieter Wallnöfer [Sat, 31 Jul 2010 18:59:02 +0000 (20:59 +0200)]
s4:torture/winreg.c - fix warnings of Solaris 10 cc through the appropriate casts

9 years agos4:torture/winreg.c - remove unreachable code
Matthias Dieter Wallnöfer [Sat, 31 Jul 2010 18:58:47 +0000 (20:58 +0200)]
s4:torture/winreg.c - remove unreachable code

9 years agos4:torture/winreg.c - add a FIXME comment on a codepart which isn't standard C-compatible
Matthias Dieter Wallnöfer [Sat, 31 Jul 2010 18:57:18 +0000 (20:57 +0200)]
s4:torture/winreg.c - add a FIXME comment on a codepart which isn't standard C-compatible

It seems to be GCC specific since it doesn't compile on Solaris 10 cc:
http://build.samba.org/?function=View+Build;host=sunX;tree=samba_4_0_waf;compiler=cc

9 years agos4:dcesrv_netr_LogonGetDomainInfo - improve the client OS informations update
Matthias Dieter Wallnöfer [Sat, 31 Jul 2010 12:52:44 +0000 (14:52 +0200)]
s4:dcesrv_netr_LogonGetDomainInfo - improve the client OS informations update

As ekacnet pointed out on the mailing list we don't need to do a delete if we
(re)set the values afterwards - only if we don't set any new ones.

9 years agos4:blackbox/newuser: use test specific user names
Stefan Metzmacher [Sat, 31 Jul 2010 09:35:31 +0000 (11:35 +0200)]
s4:blackbox/newuser: use test specific user names

As this test doesn't delete the user accounts at the end,
we should use test specific user names. That lowers the
chance of conflicts with other tests.

metze

9 years agos4:selftest: skip samba4.blackbox.ktpass for now as it's not portable
Stefan Metzmacher [Sat, 31 Jul 2010 09:31:43 +0000 (11:31 +0200)]
s4:selftest: skip samba4.blackbox.ktpass for now as it's not portable

ktutil might not be installed or from MIT.
We should build a samba4ktutil and use that instead,
until then we need to skip this test.

metze

9 years agos4:blackblock/ktpass: use test specific user name
Stefan Metzmacher [Sat, 31 Jul 2010 09:12:10 +0000 (11:12 +0200)]
s4:blackblock/ktpass: use test specific user name

metze

9 years agos4:librpc/rpc: use the correct ndr flags in dcerpc_ndr_validate_in()
Stefan Metzmacher [Sat, 31 Jul 2010 07:00:44 +0000 (09:00 +0200)]
s4:librpc/rpc: use the correct ndr flags in dcerpc_ndr_validate_in()

metze

9 years agos3-printing: remove unused get_local_printer_publishing_data() call.
Günther Deschner [Fri, 30 Jul 2010 22:54:45 +0000 (00:54 +0200)]
s3-printing: remove unused get_local_printer_publishing_data() call.

Guenther

9 years agos3-build: avoid to globally include printing and spoolss headers.
Günther Deschner [Fri, 30 Jul 2010 22:47:20 +0000 (00:47 +0200)]
s3-build: avoid to globally include printing and spoolss headers.

This shrinks precompiled headers by 3MB and will slightly speed up any build.

Guenther

9 years agocleanups: Trailing spaces, line length, etc...
Simo Sorce [Fri, 30 Jul 2010 20:34:53 +0000 (16:34 -0400)]
cleanups: Trailing spaces, line length, etc...

9 years agos3-dcerpc: Use dcerpc_guess_sizes in the server code too.
Simo Sorce [Fri, 30 Jul 2010 18:01:01 +0000 (14:01 -0400)]
s3-dcerpc: Use dcerpc_guess_sizes in the server code too.

9 years agos3-dceprc: Improve dcerpc_guess_sizes() interface
Simo Sorce [Fri, 30 Jul 2010 17:12:35 +0000 (13:12 -0400)]
s3-dceprc: Improve dcerpc_guess_sizes() interface

Make it possible to pass in the NDR padding size so that theoretically
client and server code can decide to use a different alignment.

Pass in the header length as a parameter so that this function can be used for
different type of packets.

Make sure padding size will not make the fragment exceed the maximum length.

Calculate padding taking in account the header length.

9 years agos3-dcerpc: rationalize packet creation in the server code
Simo Sorce [Fri, 30 Jul 2010 16:19:20 +0000 (12:19 -0400)]
s3-dcerpc: rationalize packet creation in the server code

Move all related functions into create_next_packet, but make it transport
neutral (not pass in pipse_struct)

9 years agos3-dcerpc: Make function to guess pdu sizes common.
Simo Sorce [Fri, 30 Jul 2010 15:27:40 +0000 (11:27 -0400)]
s3-dcerpc: Make function to guess pdu sizes common.

9 years agos3-dceprc: consolidate use of dcerpc_push_dcerpc_auth()
Simo Sorce [Fri, 30 Jul 2010 00:20:49 +0000 (20:20 -0400)]
s3-dceprc: consolidate use of dcerpc_push_dcerpc_auth()

9 years agos3-dcerpc: Remove unused functions
Simo Sorce [Fri, 30 Jul 2010 00:21:53 +0000 (20:21 -0400)]
s3-dcerpc: Remove unused functions

9 years agos3-dcerpc: use common spengo wrapper code for client SPNEGO/NTLMSSP
Simo Sorce [Thu, 29 Jul 2010 23:55:44 +0000 (19:55 -0400)]
s3-dcerpc: use common spengo wrapper code for client SPNEGO/NTLMSSP

9 years agos3-dcerpc: add sign/seal support when using SPNEGO/KRB5
Simo Sorce [Thu, 29 Jul 2010 20:34:39 +0000 (16:34 -0400)]
s3-dcerpc: add sign/seal support when using SPNEGO/KRB5

9 years agos3-dcerpc: Add SPNEGO incapsulation for KRB5 auth
Simo Sorce [Wed, 28 Jul 2010 21:06:51 +0000 (17:06 -0400)]
s3-dcerpc: Add SPNEGO incapsulation for KRB5 auth

9 years agorpcclient: Use DCERPC_AUTH_LEVEL_CONNECT if no sign/seal is set for ntlmssp
Simo Sorce [Fri, 30 Jul 2010 00:07:19 +0000 (20:07 -0400)]
rpcclient: Use DCERPC_AUTH_LEVEL_CONNECT if no sign/seal is set for ntlmssp

9 years agos3: Remove "cli" from "struct finfo"
Volker Lendecke [Fri, 30 Jul 2010 14:18:51 +0000 (16:18 +0200)]
s3: Remove "cli" from "struct finfo"

9 years agos4-smbtorture: remove ifdefed symlink test, this is already skipped internally
Günther Deschner [Fri, 30 Jul 2010 09:24:23 +0000 (11:24 +0200)]
s4-smbtorture: remove ifdefed symlink test, this is already skipped internally
in the s3 case.

Guenther

9 years agoRevert "s4-torture: Disable setting REG_BINARY printer data with size 0."
Günther Deschner [Fri, 30 Jul 2010 09:23:06 +0000 (11:23 +0200)]
Revert "s4-torture: Disable setting REG_BINARY printer data with size 0."

This reverts commit 104c419345a42a9a02ae25df482f449dc41a6380.

9 years agoWe should be using the winbindd separator in this case, not hardcoding a \\ value.
Jeremy Allison [Thu, 29 Jul 2010 20:47:27 +0000 (13:47 -0700)]
We should be using the winbindd separator in this case, not hardcoding a \\ value.

Jeremy.

9 years agos3: Do the ftruncate write cache optimization in one place
Volker Lendecke [Thu, 29 Jul 2010 12:48:06 +0000 (14:48 +0200)]
s3: Do the ftruncate write cache optimization in one place

Instead of hand-tuning all the cases that are below this piece of code, this is
a general case that we can catch upfront.

9 years agoRevert "s3: Avoid pwrite calls for the 1-byte writes with zero content"
Volker Lendecke [Thu, 29 Jul 2010 12:21:14 +0000 (14:21 +0200)]
Revert "s3: Avoid pwrite calls for the 1-byte writes with zero content"

This reverts commit 6763730304627a58139450fd3e03a0ce48e31bb9.

9 years agos3: Avoid an unnecessary ftruncate call
Volker Lendecke [Thu, 29 Jul 2010 15:04:18 +0000 (17:04 +0200)]
s3: Avoid an unnecessary ftruncate call

If we just created the file, it has length 0 by definition. This is still done
while holding the share mode lock, so no race around wrt other cifs clients.

9 years agoFix bug #7589 - ntlm_auth fails to use cached credentials.
Jeremy Allison [Thu, 29 Jul 2010 19:44:00 +0000 (12:44 -0700)]
Fix bug #7589 - ntlm_auth fails to use cached credentials.

In handling the WINBINDD_PAM_AUTH message winbindd canonicalizes a *copy*
of the mapped username, but fails to canonicalize the actual username
sent to the backend domain process. When "winbind default domain"
is set this can lead to credentials being cached with an index of
user: user, not DOMAIN\user. All other code paths that use
canonicalize_username() (WINBINDD_PAM_CHAUTHTOK, WINBINDD_PAM_LOGOFF)
correctly canonicalize the data sent to the backend. All calls
the can cause credentials to be looked up (PAM_CHAUTHTOK etc.)
correctly call canonicalize_username() to create the credential
lookup key.

Jeremy.

9 years agos3-spoolss: Increased debug level for trace output.
Andreas Schneider [Thu, 29 Jul 2010 18:48:01 +0000 (20:48 +0200)]
s3-spoolss: Increased debug level for trace output.

9 years agos3-spoolss: Fixed debug statements and increased level.
Andreas Schneider [Thu, 29 Jul 2010 18:45:07 +0000 (20:45 +0200)]
s3-spoolss: Fixed debug statements and increased level.

9 years agoFix tdb_check() to work with read-only tdb databases. The function tdb_lockall()...
Jeremy Allison [Wed, 28 Jul 2010 12:15:40 +0000 (05:15 -0700)]
Fix tdb_check() to work with read-only tdb databases. The function tdb_lockall() uses F_WRLCK internally, which doesn't work on a fd opened with O_RDONLY. Use tdb_lockall_read() instead.

Jeremy.

9 years agos4-test: Expand DSSYNC test to be executed against all partitions
Kamen Mazdrashki [Fri, 16 Jul 2010 19:52:45 +0000 (22:52 +0300)]
s4-test: Expand DSSYNC test to be executed against all partitions

It is still possible to supply specific partition
to test with on command line using
--option="dssync:partition={PARTITION_DN}"

9 years agos4-test-dssync: Issue warning in case we have diffrences in link attributes
Kamen Mazdrashki [Wed, 14 Jul 2010 02:10:34 +0000 (05:10 +0300)]
s4-test-dssync: Issue warning in case we have diffrences in link attributes

In case differences in messages are:
 1. Attributes with different values, i.e. 'replace'
 2. Those attributes are back-link attributes
then we just warn about those differences.

It turns out windows doesn't send all values for back-link
attributes in replicated_object but in linked_attributes.

I consider this as workaround for the problem.
A better solution might be to merge linked_attribute values
into replacted_objects before comparing those with LDB result
(although this will work for Win2k8 and latter)

9 years agos4-test: Simplify test_GetInfo() - we have what we need through LDB connection
Kamen Mazdrashki [Wed, 14 Jul 2010 00:42:56 +0000 (03:42 +0300)]
s4-test: Simplify test_GetInfo() - we have what we need through LDB connection

we have already opened.

Added 'config_dn' and 'schema_dn' members which are to be used
later for extending the test_FetchData() to be executed
against every partition in case partition is not specified explicitly.

DsSyncText.site_name was removed also as it is not used anymore

9 years agos4-ldb: use TALLOC_CTX type instead of 'void'
Kamen Mazdrashki [Mon, 12 Jul 2010 23:37:58 +0000 (02:37 +0300)]
s4-ldb: use TALLOC_CTX type instead of 'void'