samba.git
14 years agor1033: Include new manpages
Jelmer Vernooij [Sat, 5 Jun 2004 19:06:43 +0000 (19:06 +0000)]
r1033: Include new manpages

14 years agor1032: Fix RPC backend segfault
Jelmer Vernooij [Sat, 5 Jun 2004 18:45:27 +0000 (18:45 +0000)]
r1032: Fix RPC backend segfault

14 years agor1031: Move more code dealing with passwords to struct samr_Password.
Andrew Bartlett [Sat, 5 Jun 2004 14:04:45 +0000 (14:04 +0000)]
r1031: Move more code dealing with passwords to struct samr_Password.

Andrew Bartlett

14 years agor1030: added server side schannel support
Andrew Tridgell [Sat, 5 Jun 2004 05:01:38 +0000 (05:01 +0000)]
r1030: added server side schannel support

14 years agor1029: cope wiith samdb_result_passwords() returning a null machine password
Andrew Tridgell [Sat, 5 Jun 2004 04:51:24 +0000 (04:51 +0000)]
r1029: cope wiith samdb_result_passwords() returning a null machine password

14 years agor1028: More consistancy fixes, which should also fix the build.
Andrew Bartlett [Sat, 5 Jun 2004 04:32:50 +0000 (04:32 +0000)]
r1028: More consistancy fixes, which should also fix the build.

Andrew Bartlett

14 years agor1027: More rename:
Andrew Bartlett [Sat, 5 Jun 2004 03:43:00 +0000 (03:43 +0000)]
r1027: More rename:

pwd -> password

(should fix the build).

Andrew Bartlett

14 years agor1026: Spelling.
Tim Potter [Sat, 5 Jun 2004 03:37:27 +0000 (03:37 +0000)]
r1026: Spelling.

14 years agor1025: Rename (across the samr and netlogon pipes, so far)
Andrew Bartlett [Sat, 5 Jun 2004 03:22:10 +0000 (03:22 +0000)]
r1025: Rename (across the samr and netlogon pipes, so far)

pwd -> password
passwd -> password

username -> account_name

Also work on consistant structure feild names between these two pipes,
and fix up some callers to use samr_Password for the netlogon
credential code.

Andrew Bartlett

14 years agor1024: Use samr_Password for the machine password here - this ensures we can
Andrew Bartlett [Sat, 5 Jun 2004 03:14:59 +0000 (03:14 +0000)]
r1024: Use samr_Password for the machine password here - this ensures we can
never pass in something of the wrong length.

Andrew Bartlett

14 years agor1023: Prepare the auth subsystem interfaces for netlogon SamLogon to use.
Andrew Bartlett [Sat, 5 Jun 2004 03:09:38 +0000 (03:09 +0000)]
r1023: Prepare the auth subsystem interfaces for netlogon SamLogon to use.

Andrew Bartlett

14 years agor1022: add the session_info() call to the dcerpc server auth backends
Stefan Metzmacher [Sat, 5 Jun 2004 02:27:30 +0000 (02:27 +0000)]
r1022: add the session_info() call to the dcerpc server auth backends

metze

14 years agor1021: Because auth_serversupplied_info is not reference counted, this may
Andrew Bartlett [Sat, 5 Jun 2004 02:00:02 +0000 (02:00 +0000)]
r1021: Because auth_serversupplied_info is not reference counted, this may
only be called once per authentication.

Andrew Bartlett

14 years agor1020: Add an (untested, until the other end is hooked in) method for
Andrew Bartlett [Sat, 5 Jun 2004 01:54:16 +0000 (01:54 +0000)]
r1020: Add an (untested, until the other end is hooked in) method for
obtaining the session_info from an NTLMSSP authenticated user.

Andrew Bartlett

14 years agor1019: Push the auth subsystem away from using typedef, and over to the 'all
Andrew Bartlett [Sat, 5 Jun 2004 01:39:08 +0000 (01:39 +0000)]
r1019: Push the auth subsystem away from using typedef, and over to the 'all
goodness and light' struct ;-)

Break apart the auth subsystem's return strucutres, into the parts
that a netlogon call cares about, and the parts that are for a local
session.  This is the 'struct session_info' and it will almost
completly replace the current information stored on a vuid, but be
generic to all login methods (RPC over TCP, for example).

Andrew Bartlett

14 years agor1018: fix a const and unsigned int problem in ldb
Andrew Tridgell [Sat, 5 Jun 2004 01:30:27 +0000 (01:30 +0000)]
r1018: fix a const and unsigned int problem in ldb

14 years agor1017: - move to a centralised way of handling talloc/ldb interaction
Andrew Tridgell [Sat, 5 Jun 2004 01:29:49 +0000 (01:29 +0000)]
r1017: - move to a centralised way of handling talloc/ldb interaction

14 years agor1016: - store the schannel session key after it is established
Andrew Tridgell [Sat, 5 Jun 2004 01:29:20 +0000 (01:29 +0000)]
r1016: - store the schannel session key after it is established

- move to a centralised way of handling talloc/ldb interaction

14 years agor1015: commit the schannel session key handling code now, so abartlet and
Andrew Tridgell [Sat, 5 Jun 2004 01:27:02 +0000 (01:27 +0000)]
r1015: commit the schannel session key handling code now, so abartlet and
metze can have a look. Not tested yet, as I'm still writing the server
side schannel code.

14 years agor1014: change the handling of r->out to auto-zero the structure, and auto-copy
Andrew Tridgell [Sat, 5 Jun 2004 01:13:27 +0000 (01:13 +0000)]
r1014: change the handling of r->out to auto-zero the structure, and auto-copy
in,out,ref vars

14 years agor1013: WE NEED ALWAYS TO INIT THE OUTPUT VARS!
Stefan Metzmacher [Fri, 4 Jun 2004 20:36:44 +0000 (20:36 +0000)]
r1013: WE NEED ALWAYS TO INIT THE OUTPUT VARS!
in rpc server code!

add missing ZERO_STRUCT(r->out.info);
in samr_GetDomPwInfo

metze

14 years agor1010: make the dcesrv_crypto code a bit more generic...
Stefan Metzmacher [Fri, 4 Jun 2004 12:32:37 +0000 (12:32 +0000)]
r1010: make the dcesrv_crypto code a bit more generic...
fix type 'cyrpto' -> 'crypto'

metze

14 years agor1009: Make all users of NT and LM passwords use the samr_Password structure.
Andrew Bartlett [Fri, 4 Jun 2004 11:58:46 +0000 (11:58 +0000)]
r1009: Make all users of NT and LM passwords use the samr_Password structure.

This includes the netlogon pipe, for the machine account password
change system.

Andrew Bartlett

14 years agor1006: print out the SVN revision by configure,
Stefan Metzmacher [Fri, 4 Jun 2004 11:25:45 +0000 (11:25 +0000)]
r1006: print out the SVN revision by configure,
so that the build farm can use it

metze

14 years agor1005: there was one NTLMSSP_STATE left in:-(
Stefan Metzmacher [Fri, 4 Jun 2004 09:52:35 +0000 (09:52 +0000)]
r1005: there was one NTLMSSP_STATE left in:-(
...fix the build

metze

14 years agor1004: continue tridge's work on dcerpc server auth/crypto code
Stefan Metzmacher [Fri, 4 Jun 2004 09:46:46 +0000 (09:46 +0000)]
r1004: continue tridge's work on dcerpc server auth/crypto code
I made it much more generic, and we should be able to add a
module interface to this code, so that other DCERPC_AUTH types can be added
via modules...

metze

14 years agor1003: Removed e_ndr stuff - too verbose.
Tim Potter [Fri, 4 Jun 2004 08:00:51 +0000 (08:00 +0000)]
r1003: Removed e_ndr stuff - too verbose.

Merged across populating %needed with hf index members for functions
from old version of eparser.

14 years agor1002: Add helper files for eparser output.
Tim Potter [Fri, 4 Jun 2004 07:59:54 +0000 (07:59 +0000)]
r1002: Add helper files for eparser output.

14 years agor1001: in samba4 we don't(shouldn't) use typedef's anymore...
Stefan Metzmacher [Fri, 4 Jun 2004 07:46:24 +0000 (07:46 +0000)]
r1001: in samba4 we don't(shouldn't) use typedef's anymore...

metze

14 years agor995: - renamed many of our crypto routines to use the industry standard
Andrew Tridgell [Thu, 3 Jun 2004 23:15:16 +0000 (23:15 +0000)]
r995: - renamed many of our crypto routines to use the industry standard
  names rather than our crazy naming scheme. So DES is now called
  des_crypt() rather than smbhash()

- added the code from the solution of the ADS crypto challenge that
  allows Samba to correctly handle a 128 bit session key in all of the
  netr_ServerAuthenticateX() varients. A huge thanks to Luke Howard
  from PADL for solving this one!

- restructured the server side rpc authentication to allow for other
  than NTLMSSP sign and seal. This commit just adds the structure, the
  next commit will add schannel server side support.

- added 128 bit session key support to our client side code, and
  testing against w2k3 with smbtorture. Works well.

14 years agor990: fix tdb standalone build
Stefan Metzmacher [Thu, 3 Jun 2004 15:59:26 +0000 (15:59 +0000)]
r990: fix tdb standalone build

metze

14 years agor987: Start replacing ndr_* functions with e_ndr_* and removing stuff related
Tim Potter [Thu, 3 Jun 2004 07:33:36 +0000 (07:33 +0000)]
r987: Start replacing ndr_* functions with e_ndr_* and removing stuff related
to storing data in structure members.

14 years agor986: Attempt #3 at eparser. This time use the 'pull' code as a base as it's
Tim Potter [Thu, 3 Jun 2004 06:58:28 +0000 (06:58 +0000)]
r986: Attempt #3 at eparser.  This time use the 'pull' code as a base as it's
closer to what ethereal does.  Also try to keep as much of the generated
structure wrt pointers/structs/alignment as possible.

14 years agor985: Add stub function for dumping enums to prevent perl warnings in eparser.
Tim Potter [Thu, 3 Jun 2004 00:53:09 +0000 (00:53 +0000)]
r985: Add stub function for dumping enums to prevent perl warnings in eparser.

14 years agor976: - added -W for workgroup to locktest
Andrew Tridgell [Wed, 2 Jun 2004 08:33:29 +0000 (08:33 +0000)]
r976:  - added -W for workgroup to locktest

 - retry connections a few times when reconnecting

14 years agor975: slight improvemet to nt_errstr(), still needs to be fixed properly
Andrew Tridgell [Wed, 2 Jun 2004 08:31:47 +0000 (08:31 +0000)]
r975: slight improvemet to nt_errstr(), still needs to be fixed properly
(getting rid of the static buffer)

14 years agor970: - remove unused wchar_t check
Stefan Metzmacher [Tue, 1 Jun 2004 16:23:29 +0000 (16:23 +0000)]
r970: - remove unused wchar_t check
- move immediate structures check to the compiler checks

metze

14 years agor969: remove some unused checks
Stefan Metzmacher [Tue, 1 Jun 2004 16:17:51 +0000 (16:17 +0000)]
r969: remove some unused checks

metze

14 years agor968: use sinlcude() and no function, that's more portable between
Stefan Metzmacher [Tue, 1 Jun 2004 15:56:22 +0000 (15:56 +0000)]
r968: use sinlcude() and no function, that's more portable between
autoconf versions

metze

14 years agor967: move some configure checks to seperate files
Stefan Metzmacher [Tue, 1 Jun 2004 15:13:27 +0000 (15:13 +0000)]
r967: move some configure checks to seperate files

metze

14 years agor963: Tridge, according to ethereal this is an 'Acct Ctrl' field. My tests also
Volker Lendecke [Tue, 1 Jun 2004 14:36:43 +0000 (14:36 +0000)]
r963: Tridge, according to ethereal this is an 'Acct Ctrl' field. My tests also
indicate this although I could not find any consistent pattern.

I found this as 'net rpc group list local' in Samba3 sets this to 250 and only
gets a fixed, but incomplete list of groups out of W2k3.

I tried to correlate the results I got from w2k3 with the LDAP contents of the
corresponding entries, but I could not find anything. Ethereal only decodes
the lower byte, but to get all it seems necessary to have 0xffff here.

If you have time, could you might want to spend some of it decoding the bits
for SAMR completeness....

Volker

14 years agor962: convert 'unsigned' and 'unsigned int' to uint_t
Stefan Metzmacher [Tue, 1 Jun 2004 10:12:52 +0000 (10:12 +0000)]
r962: convert 'unsigned' and 'unsigned int' to uint_t

metze

14 years agor961: convert 'uchar' to 'uint8_t'
Stefan Metzmacher [Tue, 1 Jun 2004 08:30:34 +0000 (08:30 +0000)]
r961: convert 'uchar' to 'uint8_t'

metze

14 years agor960: convert 'unsigned int' to uint_t in the most places
Stefan Metzmacher [Tue, 1 Jun 2004 08:12:45 +0000 (08:12 +0000)]
r960: convert 'unsigned int' to uint_t in the most places

metze

14 years agor959: - a static WERROR function should return a WERROR:-)
Stefan Metzmacher [Tue, 1 Jun 2004 07:23:51 +0000 (07:23 +0000)]
r959: - a static WERROR function should return a WERROR:-)

- some minor format fixes

metze

14 years agor956: More debian updates:
Jelmer Vernooij [Mon, 31 May 2004 18:26:53 +0000 (18:26 +0000)]
r956: More debian updates:
 - add 'pidl' package
 - include manpages

14 years agor955: Update debian package rules... builds now
Jelmer Vernooij [Mon, 31 May 2004 17:14:27 +0000 (17:14 +0000)]
r955: Update debian package rules... builds now

14 years agor954: Start working on typedef parsing.
Tim Potter [Mon, 31 May 2004 03:38:46 +0000 (03:38 +0000)]
r954: Start working on typedef parsing.

14 years agor953: Remove unused Pull and Print parse functions. Remove Push from
Tim Potter [Mon, 31 May 2004 03:27:28 +0000 (03:27 +0000)]
r953: Remove unused Pull and Print parse functions.  Remove Push from
remaining parse function names.

Add some debug comments in parser output to aid in debugging.

14 years agor952: fixed schannel from my last commit
Andrew Tridgell [Sun, 30 May 2004 13:44:40 +0000 (13:44 +0000)]
r952: fixed schannel from my last commit

14 years agor951: from w2k3 behaviour, the netlogon server is supposed to give back the negotiate...
Andrew Tridgell [Sun, 30 May 2004 13:27:14 +0000 (13:27 +0000)]
r951: from w2k3 behaviour, the netlogon server is supposed to give back the negotiate flags it can support.

14 years agor950: - added netr_ServerAuthenticate3(). This is used by WinXP clients who try to...
Andrew Tridgell [Sun, 30 May 2004 13:15:15 +0000 (13:15 +0000)]
r950: - added netr_ServerAuthenticate3(). This is used by WinXP clients who try to login to Samba4, as
  WinXP sees us as an ADS server.

Unfortunately WinXP also uses a set of negotiate_flags that we don't
support yet. Some crypto work needed.

14 years agor948: Tridge suggested that the best way to write the ethereal parser is to
Tim Potter [Sun, 30 May 2004 12:02:57 +0000 (12:02 +0000)]
r948: Tridge suggested that the best way to write the ethereal parser is to
base it on the generator for the ndr pull/push code rather than trying
to get all the alignment and other junk sorted out by hand.

This commit (nearly) gets eparser to the same state it previously was
but based on parser.pm.  We correctly parse simple types, strings and
sids.

14 years agor946: Updated attribute only oplock tester to cover more cases.
Jeremy Allison [Sun, 30 May 2004 08:12:17 +0000 (08:12 +0000)]
r946: Updated attribute only oplock tester to cover more cases.
Jeremy.

14 years agor943: change samba4 to use 'uint8_t' instead of 'unsigned char'
Stefan Metzmacher [Sat, 29 May 2004 08:11:46 +0000 (08:11 +0000)]
r943: change samba4 to use 'uint8_t' instead of 'unsigned char'

metze

14 years agor937: - added a simple QuerySecurity implementation in samr server
Andrew Tridgell [Fri, 28 May 2004 13:23:30 +0000 (13:23 +0000)]
r937:  - added a simple QuerySecurity implementation in samr server

 - moved some sec desc defines into misc.idl

 - fixed pw_len field in UserInfo26

 - made some pipes available on TCP

 - added netr_DsrEnumerateDomainTrusts() to netlogon

 - added templates for remaining netlogon IDL calls (from ethereal)

 - added a unistr_noterm vs unistr error detector in ndr basic decoder

 - added torture test for netr_DsrEnumerateDomainTrusts()

14 years agor935: remove unused variable
Andrew Tridgell [Fri, 28 May 2004 03:11:35 +0000 (03:11 +0000)]
r935: remove unused variable

14 years agor934: on ascii strings STR_TERMINATE_ASCII should trigger STR_TERMINATE behaviour
Andrew Tridgell [Fri, 28 May 2004 03:11:14 +0000 (03:11 +0000)]
r934: on ascii strings STR_TERMINATE_ASCII should trigger STR_TERMINATE behaviour

14 years agor929: Remove more unused code from util_sid.c (the old-style sid code is
Andrew Bartlett [Thu, 27 May 2004 23:35:02 +0000 (23:35 +0000)]
r929: Remove more unused code from util_sid.c (the old-style sid code is
almost gone).

14 years agor924: got rid of the global well-known SIDs, instead using const defines in misc.idl
Andrew Tridgell [Thu, 27 May 2004 13:21:35 +0000 (13:21 +0000)]
r924: got rid of the global well-known SIDs, instead using const defines in misc.idl

14 years agor920: a placeholder lsads.idl file (lack of this is why the build farm is unhappy...
Andrew Tridgell [Thu, 27 May 2004 06:36:50 +0000 (06:36 +0000)]
r920: a placeholder lsads.idl file (lack of this is why the build farm is unhappy for samba4)

14 years agor919: - added lsa_QueryInfoPolicy2() to IDL, test suite and server
Andrew Tridgell [Thu, 27 May 2004 06:27:21 +0000 (06:27 +0000)]
r919: - added lsa_QueryInfoPolicy2() to IDL, test suite and server

- added lsa_OpenPolicy2() to server

- added guid handling in samdb

- added a couple more info policy levels in lsa server

- added some DNS info in the provisioning template and script

With the above changes WinXP professional can join a Samba4 domain

14 years agor918: - dcerpc endpoint name are case insensitive
Andrew Tridgell [Thu, 27 May 2004 05:24:32 +0000 (05:24 +0000)]
r918:  - dcerpc endpoint name are case insensitive

this fix allows samba3 to join a samba4 domain using "net rpc join"

14 years agor917: - added the start of a LSA server to samba4.
Andrew Tridgell [Thu, 27 May 2004 04:13:58 +0000 (04:13 +0000)]
r917: - added the start of a LSA server to samba4.

- added start of QueryDomainInfo in samr server

"net rpc info" from samba3 now works against a samba4 server. I
suspect join will work fairly soon.

14 years agor904: - fixed account expiry testing in auth_sam
Andrew Tridgell [Wed, 26 May 2004 08:02:20 +0000 (08:02 +0000)]
r904: - fixed account expiry testing in auth_sam

- added printf style format attribute checking to samdb varargs fns

- fix nt_time_to_unix() for zero and -1 times

14 years agor903: used samdb_result_passwords() in samr_ChangePasswordUser2() and fix the error...
Andrew Tridgell [Wed, 26 May 2004 07:34:49 +0000 (07:34 +0000)]
r903: used samdb_result_passwords() in samr_ChangePasswordUser2() and fix the error handling on a bad change.

With this change WinXP can now successfully change the password on a
Samba4 server via SAMR. After the change you can't login because the
handling of much_change_time seems to be broken in the auth code, but
that should be easy to fix.

14 years agor902: added torture tests for sending rubbish in the domain name field of GetDomPwInfo
Andrew Tridgell [Wed, 26 May 2004 07:33:05 +0000 (07:33 +0000)]
r902: added torture tests for sending rubbish in the domain name field of GetDomPwInfo

14 years agor901: w2k3 completely ignores the domain name argument to GetDomPwInfo,
Andrew Tridgell [Wed, 26 May 2004 07:32:30 +0000 (07:32 +0000)]
r901: w2k3 completely ignores the domain name argument to GetDomPwInfo,
always returning the info for the primary domain. I noticed this
because WinXP sends the wrong information in this field (it sends
\\server_name) and gets away with it

14 years agor900: when DEBUGLEVEL > 10 print the full deocde of all RPC calls in the server
Andrew Tridgell [Wed, 26 May 2004 07:31:19 +0000 (07:31 +0000)]
r900: when DEBUGLEVEL > 10 print the full deocde of all RPC calls in the server

14 years agor899: remove the weird shell patterns from the makerules
Stefan Metzmacher [Wed, 26 May 2004 06:31:24 +0000 (06:31 +0000)]
r899: remove the weird shell patterns from the makerules

they slowdown the build

metze

14 years agor898: - remove some unused macros
Andrew Tridgell [Wed, 26 May 2004 05:40:33 +0000 (05:40 +0000)]
r898: - remove some unused macros

- remove unused lib/smbpasswd.c

- don't set the pkt size twice when doing SMB signing

14 years agor897: - user/group creation needs to create unique names across both the
Andrew Tridgell [Wed, 26 May 2004 04:20:17 +0000 (04:20 +0000)]
r897: - user/group creation needs to create unique names across both the
  Builtin and local domain, as some calls (notably password change
  calls) don't specify a domain name, they just specifiy an account
  name.

- added the remaining password set levels to SetUserInfo in the samr
  server. We now support all of the password set and change levels
  that we know about in SAMR.

14 years agor896: - use andrews samdb_result_passwords() for the remaining password change mechanisms
Andrew Tridgell [Wed, 26 May 2004 01:16:30 +0000 (01:16 +0000)]
r896: - use andrews samdb_result_passwords() for the remaining password change mechanisms

- added samr_ChangePasswordUser2() (just a subset of samr_ChangePasswordUser3)

14 years agor895: use _t in base ndr fns
Andrew Tridgell [Wed, 26 May 2004 01:14:56 +0000 (01:14 +0000)]
r895: use _t in base ndr fns

14 years agor894: use _t in generated pidl code
Andrew Tridgell [Wed, 26 May 2004 01:14:26 +0000 (01:14 +0000)]
r894: use _t in generated pidl code

14 years agor893: a few more _t conversions
Andrew Tridgell [Wed, 26 May 2004 01:13:12 +0000 (01:13 +0000)]
r893: a few more _t conversions

14 years agor892: Actually add the NTLMSSP self-check torture code this time...
Andrew Bartlett [Tue, 25 May 2004 23:12:52 +0000 (23:12 +0000)]
r892: Actually add the NTLMSSP self-check torture code this time...

Andrew Bartlett

14 years agor891: fix compile
Gerald Carter [Tue, 25 May 2004 18:12:55 +0000 (18:12 +0000)]
r891: fix compile

14 years agor890: convert samba4 to use [u]int8_t instead of [u]int8
Stefan Metzmacher [Tue, 25 May 2004 17:50:17 +0000 (17:50 +0000)]
r890: convert samba4 to use [u]int8_t instead of [u]int8

metze

14 years agor889: convert samba4 to use [u]int16_t instead of [u]int16
Stefan Metzmacher [Tue, 25 May 2004 17:24:24 +0000 (17:24 +0000)]
r889: convert samba4 to use [u]int16_t instead of [u]int16

metze

14 years agor888: more ..int32 -> ..int32_t stuff
Stefan Metzmacher [Tue, 25 May 2004 16:50:09 +0000 (16:50 +0000)]
r888: more ..int32 -> ..int32_t stuff

metze

14 years agor884: convert samba4 to use [u]int32_t instead of [u]int32
Stefan Metzmacher [Tue, 25 May 2004 16:24:13 +0000 (16:24 +0000)]
r884: convert samba4 to use [u]int32_t instead of [u]int32

metze

14 years agor882: - create TORTURE_AUTH SUBSYSTEM
Stefan Metzmacher [Tue, 25 May 2004 15:56:02 +0000 (15:56 +0000)]
r882: - create TORTURE_AUTH SUBSYSTEM
- fix the build because of the missing torture/ntlmssp.c
- this file should go into torture/auth/ !

metze

14 years agor879: fixed a typo in the password fetch wrapper
Andrew Tridgell [Tue, 25 May 2004 14:21:09 +0000 (14:21 +0000)]
r879: fixed a typo in the password fetch wrapper

14 years agor877: This attempt at IDL was accidently included in the pervious commit.
Andrew Bartlett [Tue, 25 May 2004 14:11:39 +0000 (14:11 +0000)]
r877: This attempt at IDL was accidently included in the pervious commit.

Andrew Bartlett

14 years agor874: This patch is a pile of work on NTLMSSP:
Andrew Bartlett [Tue, 25 May 2004 14:06:28 +0000 (14:06 +0000)]
r874: This patch is a pile of work on NTLMSSP:

Samba's NTLMSSP code is now fully talloc based, which should go a long
way to cleaning up the memory leaks in this code.  This also avoids a
lot of extra copies of data, as we now allocate the 'return' blobs on
a caller-supplied context.

I have also been doing a lot of work towards NTLM2 signing and
sealing.  I have this working for sealing, but not for the verifier
(MD5 integrity check on the stream) which is still incorrect.

(I can aim a rpcecho sinkdata from a Win2k3 box to my server, and the
data arrives intact, but the signature check fails.  It does however
match the test values I have...).

The new torture test is cludged in - when we get a unit test suite
back, I'll happliy put it in the 'right' place....

Andrew Bartlett

14 years agor873: converted samba4 to use real 64 bit integers instead of
Andrew Tridgell [Tue, 25 May 2004 13:57:39 +0000 (13:57 +0000)]
r873: converted samba4 to use real 64 bit integers instead of
structures. This was suggested by metze recently.

I checked on the build farm and all the machines we have support 64
bit ints, and support the LL suffix for 64 bit constants. I suspect
some won't support strtoll() and related functions, so we will
probably need replacements for those.

14 years agor871: add a comment about how samdb_set_password() works
Andrew Tridgell [Tue, 25 May 2004 03:31:46 +0000 (03:31 +0000)]
r871: add a comment about how samdb_set_password() works

14 years agor870: we should issue a rpc fault OP_RANGE_ERROR not a WERR_NOT_SUPPORTED for
Andrew Tridgell [Tue, 25 May 2004 03:04:26 +0000 (03:04 +0000)]
r870: we should issue a rpc fault OP_RANGE_ERROR not a WERR_NOT_SUPPORTED for
functions we don't implement yet so that we don't put uninitialised
result data on the wire (found with valgrind)

14 years agor869: convert SUBSYSTEM PROCESS_MODEL to a *.mk file
Stefan Metzmacher [Tue, 25 May 2004 01:11:46 +0000 (01:11 +0000)]
r869: convert SUBSYSTEM PROCESS_MODEL to a *.mk file

metze

14 years agor868: we should issue a rpc fault OP_RANGE_ERROR not a WERR_NOT_SUPPORTED for
Andrew Tridgell [Tue, 25 May 2004 00:51:47 +0000 (00:51 +0000)]
r868: we should issue a rpc fault OP_RANGE_ERROR not a WERR_NOT_SUPPORTED for
functions we don't implement yet so that we don't put uninitialised
result data on the wire (found with valgrind)

14 years agor867: removed a couple of unused structures
Andrew Tridgell [Tue, 25 May 2004 00:50:41 +0000 (00:50 +0000)]
r867: removed a couple of unused structures

14 years agor866: convert the rest of the binaries to config.mk files
Stefan Metzmacher [Tue, 25 May 2004 00:20:19 +0000 (00:20 +0000)]
r866: convert the rest of the binaries to config.mk files

metze

14 years agor865: Regression test for attribute-only opens not causing an oplock break.
Jeremy Allison [Tue, 25 May 2004 00:05:51 +0000 (00:05 +0000)]
r865: Regression test for attribute-only opens not causing an oplock break.
Samba3 server used to get this wrong.
Jeremy.

14 years agor864: convert the smbd/* code to a config.mk file
Stefan Metzmacher [Mon, 24 May 2004 23:57:53 +0000 (23:57 +0000)]
r864: convert the smbd/* code to a config.mk file

and add server_auth, server_smb, server_rpc as server modules

metze

14 years agor863: Added test to ensure an open and locked file can be truncated by a second open.
Jeremy Allison [Mon, 24 May 2004 23:40:50 +0000 (23:40 +0000)]
r863: Added test to ensure an open and locked file can be truncated by a second open.
This was something the Samba3 server previously got wrong.
Jeremy.

14 years agor862: remove acl and sendfile stuff
Stefan Metzmacher [Mon, 24 May 2004 23:01:20 +0000 (23:01 +0000)]
r862: remove acl and sendfile stuff
it will be readded inside the ntvfs_posix module

metze

14 years agor861: remove the next round of unused stuff
Stefan Metzmacher [Mon, 24 May 2004 22:05:06 +0000 (22:05 +0000)]
r861: remove the next round of unused stuff

metze

14 years agor860: convert SUBSYSTEM TORTURE to a config.mk file
Stefan Metzmacher [Mon, 24 May 2004 22:00:40 +0000 (22:00 +0000)]
r860: convert SUBSYSTEM TORTURE to a config.mk file

metze

14 years agor858: - remove unused account policy stuff it's in our sam.ldb now:-)
Stefan Metzmacher [Mon, 24 May 2004 20:52:03 +0000 (20:52 +0000)]
r858: - remove unused account policy stuff it's in our sam.ldb now:-)
- don't link lib/ldap_escape.c it's currently unused

metze