20 years agoperl script to convert from sambaAccount to sambaSamAccount; requires Net::LDAP:...
Gerald Carter [Wed, 14 May 2003 04:37:40 +0000 (04:37 +0000)]
perl script to convert from sambaAccount to sambaSamAccount; requires Net::LDAP::LDIF
(This used to be commit 9cde1aa32aed55a3d7cb28881c6acd9800b02065)

20 years ago*****LDAP schema changes*****
Gerald Carter [Wed, 14 May 2003 03:32:20 +0000 (03:32 +0000)]
*****LDAP schema changes*****

New objectclass named sambaSamAccount which uses attribute
prefaced with the phrase 'samba' to prevent future name clashes.

Change in functionality of the 'ldap filter' parameter.  This always
defaults to "(uid=%u)" now and is and'd with the approriate objectclass
depending on whether you are using ldapsam_compat or ldapsam

conversion script for migrating from sambaAccount to
sambaSamAccount will come next.
(This used to be commit 998586e65271daa919e47e1206c0007454cbca66)

20 years agoMove some #ifdefs and function prototypes around to avoid a compiler
Tim Potter [Wed, 14 May 2003 01:10:46 +0000 (01:10 +0000)]
Move some #ifdefs and function prototypes around to avoid a compiler
warning when we have a working version of snprintf()
(This used to be commit 4836c0c0fcbf4be5f52bba60cc8843e8551b59b8)

20 years agoFix pointer return bug in get_unix_attributes()
Tim Potter [Wed, 14 May 2003 00:58:41 +0000 (00:58 +0000)]
Fix pointer return bug in get_unix_attributes()
(This used to be commit 7aa54f0a7f3bcfb5d9560ca3bb88e1549b7ce421)

20 years agospelling
Tim Potter [Wed, 14 May 2003 00:46:43 +0000 (00:46 +0000)]
(This used to be commit 865c11275685c85124b506c9bbd2a8bde2e760b9)

20 years agospelling
Tim Potter [Wed, 14 May 2003 00:27:12 +0000 (00:27 +0000)]
(This used to be commit 1b4f904277b5e9331cdf4491163f856aa78d0453)

20 years agofix up packaging for RH9
Gerald Carter [Tue, 13 May 2003 21:01:41 +0000 (21:01 +0000)]
fix up packaging for RH9
(This used to be commit 9a52b08bfc9e5004030ee03676811543b55be6d5)

20 years agoRestore a number of fixes that idra removed when he merged his
Andrew Bartlett [Tue, 13 May 2003 15:29:23 +0000 (15:29 +0000)]
Restore a number of fixes that idra removed when he merged his
idmap-and-the-rest from HEAD.

These are correctness fixes that were already in 3.0, and a memory leak fix.

The pdb_ldap changes are held back at jerry's request (he is also playing
with pdb_ldap ATM).

Andrew Bartlett
(This used to be commit c7d5e336bd91514a02974044742b058be387e490)

20 years agoFix non-constant initialiser for Sun CC.
Andrew Bartlett [Tue, 13 May 2003 06:13:36 +0000 (06:13 +0000)]
Fix non-constant initialiser for Sun CC.
(This used to be commit c2948b9248016388f9b7c5595b0e1aba7e1ace8d)

20 years agoMore info on Unix permissions.
John Terpstra [Tue, 13 May 2003 06:13:20 +0000 (06:13 +0000)]
More info on Unix permissions.
(This used to be commit 94a3d8e8ad0bb0df2ed29d0c6dceb403197c5e6c)

20 years agoTry to fix up some code in pdb_smbpasswd that assumed we still had a 'uid'.
Andrew Bartlett [Tue, 13 May 2003 01:00:25 +0000 (01:00 +0000)]
Try to fix up some code in pdb_smbpasswd that assumed we still had a 'uid'.

This is to get non-unix accounts going for the build farm.

Andrew Bartlett
(This used to be commit 3daf7cba2393af768b253c419152ea305204f3d8)

20 years agoFix up the build-farm testsuite again - I forgot to move some files into the
Andrew Bartlett [Mon, 12 May 2003 21:56:52 +0000 (21:56 +0000)]
Fix up the build-farm testsuite again - I forgot to move some files into the
template dir, and idra's checkin removed smbpasswd_nua.

Andrew Bartlett
(This used to be commit a7b34564ae4615c2585a82aa248a9cf92905254d)

20 years agoFix obvious compiler warnings.
Jeremy Allison [Mon, 12 May 2003 21:27:54 +0000 (21:27 +0000)]
Fix obvious compiler warnings.
(This used to be commit 2a6d0c2481c3c34351e57c30a85004babdbf99b0)

20 years agocopied form HEAD to get the compile going.
Gerald Carter [Mon, 12 May 2003 20:10:28 +0000 (20:10 +0000)]
copied form HEAD to get the compile going.
Simo, please fix this tomorrow or let me know
what needs to be done.

(This used to be commit 90e35640bcde1b7dfc6e68eed6ccb257f309fd0d)

20 years agoAnd finally IDMAP in 3_0
Simo Sorce [Mon, 12 May 2003 18:12:31 +0000 (18:12 +0000)]
And finally IDMAP in 3_0

We really need idmap_ldap to have a good solution with ldapsam, porting
it from the prvious code is beeing made, the code is really simple to do
so I am confident it is not a problem to commit this code in.

Not committing it would have been worst.
I really would have been able to finish also the group code, maybe we can
put it into a followin release after 3.0.0 even if it may be an upgrade

The code has been tested and seem to work right, more testing is needed for
corner cases.

Currently winbind pdc (working only for users and not for groups) is
disabled as I was not able to make a complete group code replacement that
works somewhat in a week (I have a complete patch, but there are bugs)

(This used to be commit 0e58085978f984436815114a2ec347cf7899a89d)

20 years agofixing typos in debug statements
Gerald Carter [Mon, 12 May 2003 16:06:05 +0000 (16:06 +0000)]
fixing typos in debug statements
(This used to be commit f59bcb51cfe4e268ba43245d401d212aefdf2b72)

20 years agoEliminate NULL pointers from VFS interface. All hooks now really callable, producing...
Alexander Bokovoy [Mon, 12 May 2003 16:03:16 +0000 (16:03 +0000)]
Eliminate NULL pointers from VFS interface. All hooks now really callable, producing either correct result or returning error if the feature isn't supported in the configuration
(This used to be commit af0a17349e6986eef2e2fd07b4b9f0bcd33bbe1f)

20 years agoGCC accepts unnamed initalization of sub-structure. Let's see how AIX behave with...
Alexander Bokovoy [Mon, 12 May 2003 09:01:50 +0000 (09:01 +0000)]
GCC accepts unnamed initalization of sub-structure. Let's see how AIX behave with this. Previous fix was incorrect.
(This used to be commit 13e1bbd803111361bcd9c28b5086277c912c59a6)

20 years agoRe-enable secure channel for net rpc vampire.
Tim Potter [Mon, 12 May 2003 07:18:36 +0000 (07:18 +0000)]
Re-enable secure channel for net rpc vampire.

Jump out of sam entry processing loop if the return value from
cli_netlogon_sam_sync() isn't OK or STATUS_MORE_ENTRIES.
(This used to be commit 47d8ee3679292ece5d86df11bc56c9b4d71f3d11)

20 years agoFix two bugs that were stopping net rpc vampire from working over secure
Tim Potter [Mon, 12 May 2003 07:16:21 +0000 (07:16 +0000)]
Fix two bugs that were stopping net rpc vampire from working over secure

  - If the domain name passed to create_rpc_bind_req() is empty, use

  - Correctly set the auth_padding field when the send_size is a multiple
    of 8 bytes

I've tested with nt4sp6 and win2ksp0 and it seems to work, although
there are no password hashes transferred from win2k.  The empty
passwords are being protected by the secure channel encryption though.
(This used to be commit a8c11e855611c91e94787387c62ac629232cacfa)

20 years agoA small patch from metze to fix builds on some platforms ...
Richard Sharpe [Mon, 12 May 2003 06:57:32 +0000 (06:57 +0000)]
A small patch from metze to fix builds on some platforms ...
(This used to be commit 5d643b9d03135effa5c8fd84bf1e54a7bd075e2a)

20 years agoAdding more info on File and Directory conventions / differences.
John Terpstra [Mon, 12 May 2003 05:14:19 +0000 (05:14 +0000)]
Adding more info on File and Directory conventions / differences.
(This used to be commit f03a13c0be5d33dd6548de7cd5b8d4e397793701)

20 years agoFix a wrong define check.
Alexander Bokovoy [Mon, 12 May 2003 03:13:41 +0000 (03:13 +0000)]
Fix a wrong define check.
(This used to be commit f1c4f65ae019c61f4a8178bee094dd66506dbcae)

20 years agoWrap into WITH_QUOTAS yet another place in trans2
Alexander Bokovoy [Mon, 12 May 2003 02:53:17 +0000 (02:53 +0000)]
Wrap into WITH_QUOTAS yet another place in trans2
(This used to be commit fea5d85d947965bd4789ab31607fd48d78403696)

20 years agoForgot to actually change default for quotas to -no- in
Alexander Bokovoy [Mon, 12 May 2003 02:45:29 +0000 (02:45 +0000)]
Forgot to actually change default for quotas to -no- in
(This used to be commit 7c2395d7cd98b23d58a024a54e5058101fc9d322)

20 years agoForgot to actually change default for quotas to -no- in
Alexander Bokovoy [Mon, 12 May 2003 02:41:41 +0000 (02:41 +0000)]
Forgot to actually change default for quotas to -no- in
(This used to be commit 73cf63a27f4ff53546171aec18952324d1ed983c)

20 years agoBetter isolate quota stuff
Alexander Bokovoy [Mon, 12 May 2003 02:23:50 +0000 (02:23 +0000)]
Better isolate quota stuff
(This used to be commit 73de13159c57de21a59d2e32216812f5d9d2292e)

20 years agoSmall fix for HAVE_SYS_QUOTAS usage, spotted by build farm
Alexander Bokovoy [Mon, 12 May 2003 02:01:31 +0000 (02:01 +0000)]
Small fix for HAVE_SYS_QUOTAS usage, spotted by build farm
(This used to be commit 01cff3e2e4e2724ed112fa62a775055192a2207b)

20 years agoFix up a bit of my sloppy C.
Andrew Bartlett [Mon, 12 May 2003 01:49:03 +0000 (01:49 +0000)]
Fix up a bit of my sloppy C.
(This used to be commit f67cc24acf37a9f46427c993574ecf261d7aec1a)

20 years agoSmall fix for HAVE_SYS_QUOTAS usage, spotted by build farm
Alexander Bokovoy [Mon, 12 May 2003 01:43:59 +0000 (01:43 +0000)]
Small fix for HAVE_SYS_QUOTAS usage, spotted by build farm
(This used to be commit ee5fa840686a1d9789688e7c0f1c3ca8d0175e94)

20 years agoAdd NT quota support. Patch from Stefan (metze) Metzemacher
Alexander Bokovoy [Mon, 12 May 2003 01:20:17 +0000 (01:20 +0000)]
Add NT quota support. Patch from Stefan (metze) Metzemacher
1. Allows to change quota settings for shared mount points from Win2K and WinXP from Explorer properties tab
2. Disabled by default and when requested, will be probed and enabled only on Linux where it works
3. Was tested for approx. two weeks now on Linux by two independent QA teams, have not found any bugs so far
Documentation to follow
(This used to be commit 4bf022ce9e45be85609426762ba2644ac2031326)

20 years agoGive up on the idea of avoiding lp_load() in ntlm_auth....
Andrew Bartlett [Mon, 12 May 2003 00:18:45 +0000 (00:18 +0000)]
Give up on the idea of avoiding lp_load() in ntlm_auth....

Also, we might be given a 0 length challenge, so don't smb_panic() for
smb_xmalloc() of zero size.

Andrew Bartlett
(This used to be commit 4842de04cf2e1528e726dfad070dfe3a82f46fa2)

20 years agoMake it possible to actually use --user-SID and --group-SID on a standard command...
Andrew Bartlett [Mon, 12 May 2003 00:17:44 +0000 (00:17 +0000)]
Make it possible to actually use --user-SID and --group-SID on a standard command line.

Andrew Bartlett
(This used to be commit dd14da756640ba36834a05b9da4759a809c0bb37)

20 years agoFix compile.
Andrew Bartlett [Sun, 11 May 2003 23:49:36 +0000 (23:49 +0000)]
Fix compile.
(This used to be commit ca2e453c7838b6d0ed2d0a45124d162073bbbf99)

20 years agoAdd new files needed for VFS fixes.
Alexander Bokovoy [Sun, 11 May 2003 23:44:26 +0000 (23:44 +0000)]
Add new files needed for VFS fixes.
1. idmap.h is used for unid_t only, agreed by Simo
2. sysquotas.h is used to add quota support to VFS layer and is needed for future NT quota commit
3. vfs_macros.h provides convenient macros to access VFS API.
(This used to be commit 1dd578635913c8340cfc565d7e0c8c5086070084)

20 years agoFix VFS layer:
Alexander Bokovoy [Sun, 11 May 2003 23:34:18 +0000 (23:34 +0000)]
Fix VFS layer:
1. Finally work with cascaded modules with private data storage per module
2. Convert VFS API to macro calls to simplify cascading
3. Add quota support to VFS layer (prepare to NT quota support)

Patch by Stefan (metze) Metzemacher, with review of Jelmer and me
Tested in past few weeks. Documentation to new VFS API for third-party developers to follow
(This used to be commit 91984ef5caa2d13c5d52e1f535bd3bbbae1ec978)

20 years agoRolling in VL's changes.
John Terpstra [Sun, 11 May 2003 19:57:51 +0000 (19:57 +0000)]
Rolling in VL's changes.
(This used to be commit 02244dac83623dabe927f79780cf4b7313022495)

20 years agoSet the password for a newly created trustdom account. Tested against
Volker Lendecke [Sun, 11 May 2003 16:59:06 +0000 (16:59 +0000)]
Set the password for a newly created trustdom account. Tested against
PDCs running NT4SP1, NT4SP6 and Samba 3.0.

(This used to be commit 2143446043b2c29027cf69554caddf41274df709)

20 years agoWhen we have a NT4SP0 PDC trust us, we first have to check the
Volker Lendecke [Sun, 11 May 2003 13:20:27 +0000 (13:20 +0000)]
When we have a NT4SP0 PDC trust us, we first have to check the
the password was correct. So the PDC believed that he had his trust
account correctly added. Later the auth2 naturally failed.

BTW, setting up an interdom trust account is not what I would call
well documented and easy to handle... Working on that now :-)

(This used to be commit e4e44cf3b18231ec5d7326fb42edec741caa147b)

20 years agoOk, this is a hack. On a netsec bind reply I did not see anything
Volker Lendecke [Sun, 11 May 2003 08:08:25 +0000 (08:08 +0000)]
Ok, this is a hack. On a netsec bind reply I did not see anything
useful in the auth verifier yet. So this patch ignores it.

Really checking this would be a lot more intrusive: in rpc_api_pipe we
would have to distinguish between binds and normal requests, or have
more state in the netsec info of cli_state, which is also somewhat

(This used to be commit 8de04fcf680a9bc5054965577eb500e0541ffe66)

20 years agoUpdating Global definitions.
John Terpstra [Sun, 11 May 2003 07:41:48 +0000 (07:41 +0000)]
Updating Global definitions.
(This used to be commit 18b62570c4b0b84ebe665395f08ca7c23e4c126a)

20 years agoExtending Access Control Docs
John Terpstra [Sun, 11 May 2003 07:41:23 +0000 (07:41 +0000)]
Extending Access Control Docs
(This used to be commit c0a1069463ea7239ca603deb5c568151439fea3f)

20 years agoAs discussed on samba-technical - move to 'primaryGroupSid' insted of
Andrew Bartlett [Sun, 11 May 2003 05:59:34 +0000 (05:59 +0000)]
As discussed on samba-technical - move to 'primaryGroupSid' insted of
primaryGroupID (rid).  This is consistant with the move from 'rid' to ntSid
for the primary user identifier.

Also cope with legacy installations where primaryGroupID might have been
stored as 0.

Andrew Bartlett
(This used to be commit 0e432817cb927b41af7b49fb0b5081ffdb46f85e)

20 years agoUsing /dev/urandom for determining an ldap server backoff is a waste of system
Andrew Bartlett [Sun, 11 May 2003 02:26:53 +0000 (02:26 +0000)]
Using /dev/urandom for determining an ldap server backoff is a waste of system
entropy - use sys_random() instead.

Andrew Bartlett
(This used to be commit 640462a365235aa7ce6f817778f022530a25d909)

20 years agoCompletion of Stand-Alone server docs.
John Terpstra [Sat, 10 May 2003 20:54:41 +0000 (20:54 +0000)]
Completion of Stand-Alone server docs.
(This used to be commit 44dbed231743b5caea319c448b2700bb02cc8303)

20 years agoFix typos in StandAloneServer
John Terpstra [Sat, 10 May 2003 15:37:26 +0000 (15:37 +0000)]
Fix typos in StandAloneServer
Remove admonition in ServerType - Until someone gives me a replacement for it
I consider it a non-issue.
(This used to be commit 147676b44a4b393276497419f0bc80bff8fe425a)

20 years agoReverse latest module_dummy patch from metze and me.
Jelmer Vernooij [Sat, 10 May 2003 12:56:20 +0000 (12:56 +0000)]
Reverse latest module_dummy patch from metze and me.
(This used to be commit 9dc8f7b4d46b3d1adf4b22cec2b62600dcb3c899)

20 years agoReverse previous patch from Stefan and me after comments by Andrew Bartlett
Jelmer Vernooij [Sat, 10 May 2003 11:49:51 +0000 (11:49 +0000)]
Reverse previous patch from Stefan and me after comments by Andrew Bartlett
(This used to be commit d817eaf0ecca2d878ab1ffcf7a747a02d71c811e)

20 years agoPatch from metze and me that adds dummy smb_register_*() functions so
Jelmer Vernooij [Sat, 10 May 2003 10:53:48 +0000 (10:53 +0000)]
Patch from metze and me that adds dummy smb_register_*() functions so
that is now possible to, for example, load a module which contains
an auth method into a binary without the auth/ subsystem built in.
(This used to be commit 74d9ecfe2dd7364643d32acb62ade957bd71cd0d)

20 years agoTypos and fixes resulting from VL feedback.
John Terpstra [Sat, 10 May 2003 05:26:05 +0000 (05:26 +0000)]
Typos and fixes resulting from VL feedback.
(This used to be commit ab1ff35187924c31a8fa503c0d02c6b54605da4f)

20 years agoFix from to correctly set the flags based
Jeremy Allison [Sat, 10 May 2003 02:05:24 +0000 (02:05 +0000)]
Fix from to correctly set the flags based
on the security entries sent.
(This used to be commit 45953d59f707b58e66b980512afc7f929d360ad5)

20 years agoOptimisation. We don't need to do 2 strlen's when we're setting the stat
Jeremy Allison [Sat, 10 May 2003 00:35:38 +0000 (00:35 +0000)]
Optimisation. We don't need to do 2 strlen's when we're setting the stat
struct in the correct place.
(This used to be commit 4cd8b276715f7e019fcee8fc3ebb2855610c4751)

20 years agoFix for possible segfault in readline handler. Found by
Jeremy Allison [Sat, 10 May 2003 00:30:47 +0000 (00:30 +0000)]
Fix for possible segfault in readline handler. Found by
(This used to be commit 151b7d0d88e14dd17c28e522a3e1e8f64a4a2a87)

20 years agoFix typo.
John Terpstra [Sat, 10 May 2003 00:03:58 +0000 (00:03 +0000)]
Fix typo.
(This used to be commit 7bec28f23c5bef8516e798a0808585ed1a30517e)

20 years agosyncing README to go along with scripts
Gerald Carter [Fri, 9 May 2003 21:52:49 +0000 (21:52 +0000)]
syncing README to go along with scripts
(This used to be commit 94780f0947f7ad7d5bc83e61681148637b59d7ab)

20 years agosyncing import/export smbpasswd file scripts from 2.2
Gerald Carter [Fri, 9 May 2003 21:51:57 +0000 (21:51 +0000)]
syncing import/export smbpasswd file scripts from 2.2
(This used to be commit ee1374cabf38c3d99e66a45316e232d1c2cfbe6a)

20 years agoremoving total print jobs since it is not used anymore
Gerald Carter [Fri, 9 May 2003 21:47:32 +0000 (21:47 +0000)]
removing total print jobs since it is not used anymore
(This used to be commit b87be0dddfcace95527b9a05f8f81cd6d2e86d39)

20 years agoMakefile changes to allow new NTLMv2 patch to work.
Andrew Bartlett [Fri, 9 May 2003 15:25:20 +0000 (15:25 +0000)]
Makefile changes to allow new NTLMv2 patch to work.
(This used to be commit 0c1946e51c7cd18907eb65c93042758196fe74c9)

20 years agoCleanups. My NTLMv2 changes also changed the preference from using an implicit
Andrew Bartlett [Fri, 9 May 2003 15:02:11 +0000 (15:02 +0000)]
Cleanups.  My NTLMv2 changes also changed the preference from using an implicit
structure-memcpy for DATA_BLOB parameters to using a pointer to that DATA_BLOB.

auth_sam calls some of these functions, so I've cleaned it all up to use this
format now.

Also clean up some debug statements to make them easier to read.

Andrew Bartlett
(This used to be commit 0c355c274a6ac084e4bf15a15613dfc007d6c5fc)

20 years agoFinally get NTLMv2 working on the client!
Andrew Bartlett [Fri, 9 May 2003 14:42:20 +0000 (14:42 +0000)]
Finally get NTLMv2 working on the client!

With big thanks to tpot for the ethereal disector, and for the base code
behind this, we now fully support NTLMv2 as a client.

In particular, we support it with direct domain logons (tested with ntlm_auth
--diagnostics), with 'old style' session setups, and with NTLMSSP.

In fact, for NTLMSSP we recycle one of the parts of the server's reply directly...

(we might need to parse for unicode issues later).

In particular, a Win2k domain controller now supplies us with a session key
for this password, which means that doman joins, and non-spnego SMB signing
are now supported with NTLMv2!

Andrew Bartlett
(This used to be commit 9f6a26769d345d319ec167cd0e82a45e1207ed81)

20 years agoFix bug #4 for net rap. Allow more than 50 chars for long form listings of users...
Jim McDonough [Fri, 9 May 2003 12:58:28 +0000 (12:58 +0000)]
Fix bug #4 for net rap.  Allow more than 50 chars for long form listings of users and groups.
(This used to be commit dcc6d9e76c737400aaffdd4f261fd0f191aaeea8)

20 years agoFix comment - we now have 5 types of error...
Andrew Bartlett [Fri, 9 May 2003 09:45:37 +0000 (09:45 +0000)]
Fix comment - we now have 5 types of error...
(This used to be commit 372a574a73b86855cf6efc18349e5ba24067d690)

20 years agoMake sure we always have some client data, not just the hash. An NTLMv2 or
Andrew Bartlett [Fri, 9 May 2003 09:41:08 +0000 (09:41 +0000)]
Make sure we always have some client data, not just the hash.  An NTLMv2 or
LMv2 response less than 24 bytes is just silly.

Andrew Bartlett
(This used to be commit b4ecdb2e582376d2713f81e8e32a668014905d70)

20 years agoElsewhere, we use a #define for this environment variable name, so do the
Andrew Bartlett [Fri, 9 May 2003 09:37:12 +0000 (09:37 +0000)]
Elsewhere, we use a #define for this environment variable name, so do the
same here.

Andrew Bartlett
(This used to be commit a4556786d28724309321a02afbf5005158440258)

20 years agoWhen checking if a SID is in a domain, make sure that indeed the user RID is
Andrew Bartlett [Fri, 9 May 2003 09:33:51 +0000 (09:33 +0000)]
When checking if a SID is in a domain, make sure that indeed the user RID is
one element longer than the domain sid.

Andrew Bartlett
(This used to be commit c61e5e38776d2de53d120b592a6685158e79ebb8)

20 years agoSync up to head ...
Richard Sharpe [Fri, 9 May 2003 07:43:07 +0000 (07:43 +0000)]
Sync up to head ...
(This used to be commit 045210e129e6e0aef8f847e7ed8714d0d9974e7f)

20 years agoMore edits, hackety hack.
John Terpstra [Fri, 9 May 2003 06:48:00 +0000 (06:48 +0000)]
More edits, hackety hack.
(This used to be commit 044489f218fa90df3a4838980b060738d6dcbffc)

20 years agoSync to the changes in head ...
Richard Sharpe [Fri, 9 May 2003 06:23:55 +0000 (06:23 +0000)]
Sync to the changes in head ...
(This used to be commit 7f76eac5a0f93107d990b0fde651838c38970092)

20 years agoAdded some more diagnostic tests to check out a theory that having either hash
Tim Potter [Fri, 9 May 2003 06:03:11 +0000 (06:03 +0000)]
Added some more diagnostic tests to check out a theory that having either hash

  - auth with ntlmv2 and lmv2 but deliberately break the ntlmv2 hash
  - auth with ntlmv2 and lmv2 but deliberately break the lmv2 hash
  - auth with ntlm and lm but deliberately break the ntlm hash
  - auth with ntlm and lm but deliberately break the lm hash

My theory is that the NTLM or NTLMv2 field must be correct and if it is,
it doesn't matter what the value of the LM or LMv2 field is.

Fixed cosmetic test name display bug.
(This used to be commit 5dcde9451bd0d6a7462b77cf5ed137bfd691adaa)

20 years agoFix up a bunch of problems in editreg.c
Richard Sharpe [Fri, 9 May 2003 05:48:50 +0000 (05:48 +0000)]
Fix up a bunch of problems in editreg.c

Now the build farm will no doubt find more.
(This used to be commit e91e648c9b0841fbffbc8f39e71abade0996a1e7)

20 years agoFix nasty bug pointed out by samba-technical poster. If name is mangled
Jeremy Allison [Fri, 9 May 2003 01:06:27 +0000 (01:06 +0000)]
Fix nasty bug pointed out by samba-technical poster. If name is mangled
then we weren't always correctly detecting that it had a valid stat struct
and so might now return a 'file existed'. Finally realized this when installing
the W2K resource kit as a test case.
(This used to be commit d48069ccd8351e4bff097a7f7500c738870a413d)

20 years agofix bug #47; revert registration of workgroup<1b> to 2.2 behavior
Gerald Carter [Thu, 8 May 2003 21:19:58 +0000 (21:19 +0000)]
fix bug #47; revert registration of workgroup<1b> to 2.2 behavior
(This used to be commit 016f6b4e19c2b8e4f5e1d010cc428ca194650140)

20 years agofixed bug #75; add check for non-zero destlen
Gerald Carter [Thu, 8 May 2003 20:19:10 +0000 (20:19 +0000)]
fixed bug #75; add check for non-zero destlen
(This used to be commit 83bb84f13121267992e78f2d005257932c711f23)

20 years agoHopefully re-enable building heimdal. It needs -lresolv included in LIBS
Jim McDonough [Thu, 8 May 2003 17:02:13 +0000 (17:02 +0000)]
Hopefully re-enable building heimdal.  It needs -lresolv included in LIBS
to link during configure checks.
(This used to be commit 7af282e7ff9c2cccfab97130dc66515a4852c25f)

20 years agoRevert a patch that somehow slipped in...
Volker Lendecke [Thu, 8 May 2003 08:12:57 +0000 (08:12 +0000)]
Revert a patch that somehow slipped in...

(This used to be commit 6cde3d4d655bbe1d81e68ec2ec7a23669ac82120)

20 years agoThis puts real netlogon connection caching to winbind. This becomes
Volker Lendecke [Thu, 8 May 2003 08:02:52 +0000 (08:02 +0000)]
This puts real netlogon connection caching to winbind. This becomes
important once we start doing schannel, as there would be a lot more
roundtrips for the second PIPE open and bind. With this patch logging
in to a member server is a matter of two (three if you count the
ack...) packets between us and the DC.

(This used to be commit 5b3cb7725a974629d0bd8b707bc2940c36b8745e)

20 years agoAnother set of updates to the docs.
John Terpstra [Thu, 8 May 2003 07:40:21 +0000 (07:40 +0000)]
Another set of updates to the docs.
(This used to be commit 9abe3b23836ae75bd31fd2af4c7d82f34c27f52a)

20 years agoThis puts back wins.dat into nmbd for easy editing. It leaves most of
Volker Lendecke [Thu, 8 May 2003 07:33:39 +0000 (07:33 +0000)]
This puts back wins.dat into nmbd for easy editing. It leaves most of
the other infrastructure with name owners etc in place. If anybody is
really going to tackle winsrepld, it will probably not be hard to put
the additional info back.

(This used to be commit eb82daa84a5339f28ebf431ee1044b7e1e4a4300)

20 years agoAnother attempt at undoing my bogus patch
Tim Potter [Thu, 8 May 2003 05:10:44 +0000 (05:10 +0000)]
Another attempt at undoing my bogus patch
(This used to be commit 05a684b3be1525aad3589ded9e59c3f012b5ef20)

20 years agoAlways initialise this - it helps callers who use this in a loop...
Andrew Bartlett [Wed, 7 May 2003 23:09:18 +0000 (23:09 +0000)]
Always initialise this - it helps callers who use this in a loop...

Andrew Bartlett
(This used to be commit 97bc047434284527f25e130a72981da704ed1212)

20 years agoFix the spinning bug for 'net rpc user' as well - there are more errors in
Andrew Bartlett [Wed, 7 May 2003 22:56:02 +0000 (22:56 +0000)]
Fix the spinning bug for 'net rpc user' as well - there are more errors in
this world than 'status more entires'...

Also move all the cases to 'NT_STATUS_EQUAL()' to test it.

Andrew Bartlett
(This used to be commit b4645bf0661dadcd077b21bb6f6452ed8b2eb726)

Jeremy Allison [Wed, 7 May 2003 22:19:43 +0000 (22:19 +0000)]
are identical - noticed by "Dr. Tilo Levante" <>.
(This used to be commit f6d7c279bc8354202f2a9b39fec4a4c8ace368a3)

20 years agoFix irritating typo - non arg functions should be fn(void).
Jeremy Allison [Wed, 7 May 2003 21:40:11 +0000 (21:40 +0000)]
Fix irritating typo - non arg functions should be fn(void).
(This used to be commit 395dfd196cf4bcd432a4895d3dd09fefd46cd6d8)

20 years ago%f -> %p (reported by Kurt Pfeifle <>
Jelmer Vernooij [Wed, 7 May 2003 18:49:34 +0000 (18:49 +0000)]
%f -> %p (reported by Kurt Pfeifle <>
(This used to be commit 28a6e8a6947b54169ff484a7255e601dcc7899d7)

20 years agoMake fchown, fchmod conditional for systems that don't have them.
Jeremy Allison [Wed, 7 May 2003 18:28:24 +0000 (18:28 +0000)]
Make fchown, fchmod conditional for systems that don't have them.
(This used to be commit cf78b1e7fe72aec72d03c86c46a8ca49df539c11)

20 years agoSMB Signing with NTLMv2 works!
Andrew Bartlett [Wed, 7 May 2003 12:58:59 +0000 (12:58 +0000)]
SMB Signing with NTLMv2 works!

(well, under certain conditions :-)

There is no length limit on the size of the authentication response added
into the MD5 hash.  (We had previously limited this to lengths like 40, 44 or
64 in attempts to make sense of what the SNIA spec tells us).

Instead, the entire authentication response is added in.

Currently, this only works on a Win2k domain members with a Samba PDC,
becouse our NTLMv2 code currently fails against an Win2k PDC.

However, this splits the problem in half - particularly as the NTLMv2 format
is known, and even has an ethereal disector! (thanks tpot).

Andrew Bartlett
(This used to be commit 7645d3d28afbb8eea502c0e063df3afb3aa812f4)

20 years agoIgnore generated file
Alexander Bokovoy [Wed, 7 May 2003 11:11:39 +0000 (11:11 +0000)]
Ignore generated file
(This used to be commit 1702f3d709147b3aea5aaa7e9ed29c654322de9b)

20 years agoSet our 'global sam name' in one place. For domain controllers, this is
Andrew Bartlett [Wed, 7 May 2003 08:21:06 +0000 (08:21 +0000)]
Set our 'global sam name' in one place.  For domain controllers, this is
lp_workgroup(), for all other server this is global_myname().

This is the name of the domain for accounts on *this* system, and getting
this wrong caused interesting bugs with 'take ownership' on member servers
and standalone servers at Snap.

(They lookup the username that they got, then convert that to a SID - but
becouse the domain out of the smbpasswd entry was wrong, we would fail the

Andrew Bartlett
(This used to be commit 5fc78eba20411f3f5a8ccadfcba5c4ab73180dba)

20 years agoFix missing closing para marker.
John Terpstra [Wed, 7 May 2003 07:50:22 +0000 (07:50 +0000)]
Fix missing closing para marker.
(This used to be commit 3e7860a104a6fe9da81d981e6301d9c3f78648fd)

20 years agoMore updates. Now working on BDC Documentation.
John Terpstra [Wed, 7 May 2003 07:44:01 +0000 (07:44 +0000)]
More updates. Now working on BDC Documentation.
(This used to be commit e38695fa369ba20f23046754084b08ebdc211b5a)

20 years agoWhoops - that wasn't a whitespace syncup after all.
Tim Potter [Wed, 7 May 2003 05:52:44 +0000 (05:52 +0000)]
Whoops - that wasn't a whitespace syncup after all.
(This used to be commit 42d0414ed244b92b665cb231f6756f60391861dd)

20 years agoForce ASCII for client messages. Patch from David Lee <>
Jeremy Allison [Wed, 7 May 2003 02:00:58 +0000 (02:00 +0000)]
Force ASCII for client messages. Patch from David Lee  <>
(This used to be commit f219e8309c7d17b332873e9283ab3c3796e7e799)

20 years agoWe used to use the name "*",0x0 here, but some Windows
Jeremy Allison [Wed, 7 May 2003 01:00:36 +0000 (01:00 +0000)]
We used to use the name "*",0x0 here, but some Windows
servers don't answer that name. However we *know* they
have the name workgroup#1b (as we just looked it up).
So do the node status request on this name instead.
Found at LBL labs.
(This used to be commit 41e3abe8b80026812ea7dd7ad535e8e41e26daa4)

20 years agoMore stuffing, this turkey will soon be done!
John Terpstra [Tue, 6 May 2003 23:57:07 +0000 (23:57 +0000)]
More stuffing, this turkey will soon be done!
(This used to be commit a261d446d03f491db3392da2570e68ad5139fafc)

20 years agoFix <!DOCTYPE>'s
Jelmer Vernooij [Tue, 6 May 2003 21:55:00 +0000 (21:55 +0000)]
Fix <!DOCTYPE>'s
(This used to be commit 16978cac9cd80f7d3fbafef5b8ff0cb60e109b4e)

20 years agoAdd generating text versions of the docs using xmlto
Jelmer Vernooij [Tue, 6 May 2003 21:54:10 +0000 (21:54 +0000)]
Add generating text versions of the docs using xmlto
(This used to be commit 7a85934670e102e06660cf5ff0737d1298f8447b)

20 years agoCheck for xmlto. Not required and not used (yet)
Jelmer Vernooij [Tue, 6 May 2003 21:43:30 +0000 (21:43 +0000)]
Check for xmlto. Not required and not used (yet)
(This used to be commit f5a05cfc8766f02c282838648900c00f64f81e42)

20 years agoMerge from HEAD.
John Terpstra [Tue, 6 May 2003 17:19:33 +0000 (17:19 +0000)]
Merge from HEAD.
(This used to be commit 243a8699cd57ba3b7ff49f8c40f75cc2ffc22e4e)

20 years agoThis commit was manufactured by cvs2svn to create branch 'SAMBA_3_0'.(This used to...
cvs2svn Import User [Tue, 6 May 2003 13:10:11 +0000 (13:10 +0000)]
This commit was manufactured by cvs2svn to create branch 'SAMBA_3_0'.(This used to be commit 8bee59ffcea1495f03b35d38da0eb76955b93f3d)

20 years agosort out some include dependencies
Simo Sorce [Tue, 6 May 2003 13:10:10 +0000 (13:10 +0000)]
sort out some include dependencies
split out privileges from rpc_lsa.h
(This used to be commit 37d7cc8162d02a664095dbe0fc8d7250d1ed51c9)